Infertility

prevalence estimates
1990–2021
Infertility
prevalence estimates
1990–2021
Infertility prevalence estimates, 1990–2021

ISBN 978-92-4-006831-5 (electronic version)

ISBN 978-92-4-006832-2 (print version)

© World Health Organization 2023

Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 IGO licence
(CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo).

Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work
is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific
organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your
work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following
disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not
responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”.

Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World
Intellectual Property Organization (http://www.wipo.int/amc/en/mediation/rules/).

Suggested citation. Infertility prevalence estimates, 1990–2021. Geneva: World Health Organization; 2023. Licence: CC BY-NC-SA 3.0 IGO.

Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris.

Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for
commercial use and queries on rights and licensing, see https://www.who.int/copyright.

Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or
images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the
copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with
the user.

General disclaimers. The designations employed and the presentation of the material in this publication do not imply the
expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its
authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate
border lines for which there may not yet be full agreement.

The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended
by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary
products are distinguished by initial capital letters.

All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published
material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and
use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use.
Contents
Foreword.............................................................................................................................................................................v
Acknowledgements..............................................................................................................................................................vi
Abbreviations.................................................................................................................................................................... viii

Executive summary.............................................................................................................................................................. ix

1. Introduction.....................................................................................................................................................................1

2. Methods...........................................................................................................................................................................3
2.1 Data sources and identification of relevant studies ................................................................................................................................. 3

2.2 Data extraction............................................................................................................................................................................................ 5

2.3 Data analysis .............................................................................................................................................................................................. 5

2.3.1 Risk of bias assessment ................................................................................................................................................................... 5
2.3.2 Descriptive analysis.......................................................................................................................................................................... 5
2.3.3 Meta-analysis and meta-regression ................................................................................................................................................ 6
2.3.4 Sensitivity analyses........................................................................................................................................................................... 6
2.3.5 Rating certainty of evidence ............................................................................................................................................................ 6

3. Results.............................................................................................................................................................................7
3.1 Description of studies................................................................................................................................................................................. 9

3.2 Methodological approaches for estimating infertility prevalence......................................................................................................... 11

3.3 Definitional characteristics ...................................................................................................................................................................... 16

3.3.1 Type of prevalence.......................................................................................................................................................................... 16
3.3.2 Numerator....................................................................................................................................................................................... 16
3.3.3 Denominator................................................................................................................................................................................... 16
3.4 Study population characteristics............................................................................................................................................................. 17
3.4.1 Sample type .................................................................................................................................................................................... 17
3.4.2 Sex of respondents......................................................................................................................................................................... 17
3.4.3 Income level.................................................................................................................................................................................... 17
3.4.4 Regional availability of studies....................................................................................................................................................... 18
3.4.5 Ranges of reported estimates......................................................................................................................................................... 18
3.5 Pooled 12-month infertility estimates .................................................................................................................................................... 19
3.5.1 Pooled primary and secondary infertility...................................................................................................................................... 20
3.5.2 Sensitivity Analyses........................................................................................................................................................................ 21
3.5.3 Pooled infertility estimates stratified by region ........................................................................................................................... 22
3.5.4 Pooled infertility estimates stratified by income, population, and sex of respondents............................................................. 22
3.5.5 Pooled infertility estimates stratified by methodological approach........................................................................................... 23
3.5.6 Meta-regression results by period and lifetime............................................................................................................................. 23
3.6 Certainty of the evidence......................................................................................................................................................................... 23

4. Discussion...................................................................................................................................................................... 25
4.1 Research gaps and measurement challenges......................................................................................................................................... 27
4.1.1 Lack of sufficient studies from some regions or studies with male participants........................................................................ 27
4.1.2 Variation in definitions and in inclusion and exclusion criteria in studies estimating infertility ............................................... 27
4.1.3 Numerous study designs and methods of estimating infertility.................................................................................................. 27
4.1.4 Certainty of estimates and other limitations................................................................................................................................. 28
4.2 Implications for research.......................................................................................................................................................................... 28

4.3 Policy and programmatic implications................................................................................................................................................... 29

4.4 Conclusion ............................................................................................................................................................................................... 29

References......................................................................................................................................................................... 30

Annexes............................................................................................................................................................................. 32

Executive summary Introduction Methods Results Discussion Annexes iii

List of tables
Table 3.1. Five approaches to measuring infertility prevalence identified from the systematic review........................................................... 12
Table 3.2. Range of 12-month period and lifetime infertility prevalence estimates by methodological approach
and other study descriptors................................................................................................................................................................ 13
Table 3.3. Results from sensitivity analyses......................................................................................................................................................... 22
Table 3.4. Pooled lifetime and period infertility prevalence estimates and multivariable odds ratios associations by region and
methodological approach, adjusting for definitional factors and risk of bias.................................................................................. 24

List of figures
Figure 1.1. Wide-ranging findings from previous estimates of infertility............................................................................................................. 2
Figure 1.2. Key questions addressed in these estimates....................................................................................................................................... 2
Figure 2.1. Inclusion criteria................................................................................................................................................................................... 4
Figure 2.2. Exclusion criteria................................................................................................................................................................................... 4
Figure 2.3. Data extracted from studies.................................................................................................................................................................. 5
Figure 3.1. Identification of studies via databases and other methods................................................................................................................ 8
Figure 3.2. Risk of bias of included studies.......................................................................................................................................................... 10
Figure 3.3. Lifetime and period infertility prevalence by methodological approach ........................................................................................ 15
Figure 3.4. Lifetime and period infertility prevalence by country income level................................................................................................. 17
Figure 3.5. Pooled lifetime infertility prevalence estimates................................................................................................................................ 19
Figure 3.6. Pooled period infertility prevalence estimates ................................................................................................................................ 19
Figure 3.7. Pooled lifetime and period infertility prevalence estimates for primary infertility ....................................................................... 20
Figure 3.8. Pooled lifetime and period infertility prevalence estimates for secondary infertility..................................................................... 21

List of annexes
Annex 1. Risk of bias assessment ......................................................................................................................................................................... 33
Annex 2. Studies included in the systematic review............................................................................................................................................ 37
Annex 3. Summary of included studies by region ............................................................................................................................................... 43
Annex 4. Pooled lifetime and period infertility prevalence estimates, by methodological approach.............................................................. 76
Annex 5. Certainty of estimates .......................................................................................................................................................................... 77

Executive summary Introduction Methods Results Discussion Annexes iv

Foreword

This report, the first of its kind in a decade, reveals an understanding of this common challenge that can negatively
important truth: infertility does not discriminate. For millions impact mental health and cause economic hardship.
around the world, the path to parenthood can be difficult to
access, if not impossible. Globally, an estimated 1 out of every Access to sexual and reproductive health services is the primary
6 people are affected by the inability to have a child at some way for people to have the best chance of having the number
point in their life. This is regardless of where they live and what of children they desire. Addressing infertility is also important
resources they have. for achieving the health and gender equality targets of the
2030 Sustainable Development Goals. The data in this report
Infertility affects millions. Even still, it remains understudied, emphasize the need to provide access to prevention, diagnosis
and solutions underfunded, and inaccessible to many, as the and treatment of infertility. However, in most countries, these
result of high costs, social stigma and limited availability. services are inadequate.

The causes of infertility are varied and often complex, and is This report will help policy-makers, civil society organizations,
something that both men and women experience. Indeed, a health service providers, researchers and others stakeholders
wide variety of people, in all regions, may require fertility care. to understand the magnitude of infertility, which is critical for
monitoring, assessing, and improving equitable access to quality
As with any health issue, it is imperative that we know fertility care services, as well as addressing risk factors and
more. This report emphasizes the importance of quantifying consequences of infertility. It is my hope that governments use
infertility, as well as knowing who needs fertility care, and this report to develop evidence-based policies and adopt proven
how risks can be reduced. Using a universal definition and solutions, as part of their efforts to strengthen health systems to
consistent methods to measure infertility will improve our help people fulfil their fertility intentions and live healthier lives.

Dr Tedros Adhanom Ghebreyesus

Director-General, World Health Organization

Executive summary Introduction Methods Results Discussion Annexes v

Acknowledgements

The World Health Organization (WHO) Department of Sexual and of Psychology, Cardiff University, Cardiff, United Kingdom),
Reproductive Health and Research thanks all individuals who Anjani Chandra (National Center for Health Statistics, US
contributed to developing these infertility prevalence estimates. Centers for Disease Control and Prevention, USA), Barbara
Collura (Resolve, McLean, Virginia, USA), Mae Dirac
WHO appreciates the work of the research team that supported (Departments of Health Metrics and Family Medicine, University
the generation of these estimates: Marta Bornstein (Division of Washington, Seattle, USA), Silke Dyer (Department of
of Epidemiology, Ohio State University, Columbus, United Obstetrics and Gynaecology, University of Cape Town, Cape
States of America (USA)), Carie Jean Cox (Independent Town, South Africa), Farid Foroutan (Department of Health
Consultant, Luxembourg City, Luxembourg), Courtney Johnson Research Methods, Evidence, and Impact, McMaster University,
(Department of Epidemiology and Community Health, School Hamilton, Canada), Natalia Gogliormella (National Medically
of Public Health, University of Minnesota, Minneapolis, USA), Assisted Reproduction Programme, Ministry of Health, Buenos
Nedelina Tchangalova (University of Maryland Libraries, Aires, Argentina), Apoorva Jadhav (Policy Evaluation and
College Park, Maryland, USA) and Marie Elizabeth Thoma Communication Division, Office of Population and Reproductive
(Department of Family Science, University of Maryland, Health, Bureau for Global Health, Washington, DC, USA),
Maryland, USA). Neils Keiding (Department of Public Health, University of
Copenhagen, Copenhagen, Denmark), Germaine M. Buck Louis
The estimates presented in this report are also published as a (College of Health and Human Services, George Mason
peer reviewed article: Cox C.M., Thoma M.E., Tchangalova N., University, Fairfax, USA), Manala Makua (National Department
Mburu G., Bornstein M.J., Johnson C.L., and Kiarie J., Infertility of Health, Johannesburg, South Africa), Victoria Mansur
prevalence and the methods of estimation from 1990 to 2021: (National Programme for Assisted Reproduction, Ministry
a systematic review and meta-analysis, Human Reproduction of Health, Buenos Aires, Argentina), Alexander McLain
Open. Nov 12;2022(4):hoac051. doi: 10.1093/hropen/hoac051. (Department of Epidemiology and Biostatistics, Arnold School
of Public Health, University of South Carolina, Columbia, USA),
The following advisory committee members provided inputs Ashraf Nabhan (Department of Obstetrics and Gynaecology, Ain
and feedback on the methodology and protocol used to develop Shams University, Cairo, Egypt), Zozo Nene (Faculty of Health
these estimates: Rudolph Kantum Adageba (Ruma Fertility Sciences, University of Pretoria, South Africa), Steven Ory
and Specialist Hospital, Kumasi, Ghana), Patricia F. Anderson (Herbert Wertheim College of Medicine, Florida International
(Taubman Health Sciences Library, University of Michigan, Ann University, Florida, USA), Antoinette Righarts (Dunedin School
Arbor, USA), Rachid Bezad (Faculty of Medicine and Hospital of Medicine, Dunedin, New Zealand), Iqbal Shah (Department
Les Orangers, Mohammed V University, Rabat, Morocco), Sarah of Global Health and Population, Harvard T.H. Chan School
Bradley (SHOPs Plus Project, Abt Associates, Maryland, USA), of Public Health, Boston, USA), Rémy Slama (Environmental
Silke Dyer (Department of Obstetrics and Gynaecology, Faculty Epidemiology Unit, INSERM, La Tronche, Grenoble, France),
of Health Sciences, University of Cape Town, Cape Town, South Gretchen Stevens (Independent Consultant, Los Angeles, USA),
Africa), Jasmine Fledderjohann (Department of Sociology, Lauren Wise (Department of Epidemiology, Boston University
Lancaster University, Bailrigg, United Kingdom of Great Britain School of Public Health, Boston, USA), Hafida Yartaoui (Family
and Northern Ireland), Marcos Horton (Pregna Medicina Planning Division, Ministry of Health, Rabat, Morocco).
Reproductiva, Buenos Aires, Argentina), LaTeesa James
(Taubman Health Sciences Library, University of Michigan, The following WHO staff participated in the technical
Ann Arbor, USA), Tianjing Li (School of Medicine, University consultation or otherwise reviewed these estimates:
of Colorado, Colorado, USA), Ginny Ryan (Department of Mohamed Mahmoud Ali, Ian Askew, Jenny
Obstetrics and Gynecology, University of Iowa, USA), and Cresswell, Lale Say and Nandita Thatte (Department
Lauren Wise (Department of Epidemiology, Boston University of Sexual and Reproductive Health and Research);
School of Public Health, Boston, USA). Christine Sonja Autenrieth, Zoe Brillantes, Bochen Cao,
and Diana Estevez Fernandez (Division of Data, Analytics and
Additional feedback on the estimates was received from the Delivery for Impact); Karima Gholbzouri, (WHO Regional Office
following participants in the WHO Technical Consultation for the Eastern Mediterranean); Rodolfo Gomez Ponce de Leon
on Estimates of Infertility Prevalence held on 6–8 July 2021: (WHO Regional Office for the Americas / Pan American Health
Rachid Bezad (Faculty of Medicine and Hospital Les Orangers, Organization [PAHO]); and Chandani Anoma Jayathilaka (WHO
Mohammed V University, Rabat, Morocco), Jacky Boivin (School Regional Office for South-East Asia).

Executive summary Introduction Methods Results Discussion Annexes vi

Acknowledgements

The overall process of generating these estimates was led by Copy-editing: Kelly Safreed-Harmon
Gitau Mburu under the guidance of James Kiarie, and with
the support of Ian Askew and Pascale Allotey, (Department of Photo credits :
Sexual and Reproductive Health and Research). Therese Curtin Cover page photo © photobomb
provided administrative support. Page ix top photo © Jan Antonin Kolar on Unsplash
Page ix bottom photo © Harsha KR
The work to generate these estimates was financially supported Page x photo © ccbarr
by the UNDP-UNFPA-UNICEF-WHO-World Bank Special Page 1 photo © WHO/Ahmad Yusni
Programme of Research, Development and Research Training in Page 25 photo © WHO/Yoshi Shimizu
Human Reproduction (HRP), World Health Organization. Page 33 photo © Gifts for moms
Page 37 photo © WHO/Anna Usova
Design and layout: Lushomo Page 77 photo ©WHO/Khadija Farah

Executive summary Introduction Methods Results Discussion Annexes vii

Abbreviations

CI confidence interval

HIC high-income countries

LMIC low- and middle-income countries

OR odds ratio

SDG Sustainable Development Goal

SE standard error

SRHR sexual and reproductive health and rights

TTP time-to-pregnancy

WHO World Health Organization

Executive summary Introduction Methods Results Discussion Annexes viii

Executive
summary

Introduction
Infertility is a disease of the male or female reproductive mental health and to decide the number, timing and spacing
system defined by the failure to achieve a pregnancy after of one’s children.
12 months or more of regular unprotected sexual intercourse.
Infertility impacts millions of people worldwide, often Understanding the magnitude of infertility is critical for
with devastating consequences. Addressing infertility is an developing appropriate interventions, for monitoring access
important component of sexual and reproductive health and to quality fertility care, and for mitigating risk factors for and
rights, but in most countries, infertility policies and services consequences of infertility. Yet there is considerable variation
are inadequate. Addressing infertility is central to achieving in estimates of infertility. Differences in how infertility is defined
Sustainable Development Goal (SDG) 3 – Ensure healthy lives and measured partly contribute to this variation.
and promote well-being for all at all ages – and SDG 5 – Achieve
gender equality and empower all women and girls. Addressing This report provides insight into global and regional infertility
infertility is also central to achieving the human rights to the prevalence by analyzing all relevant studies from 1990 to 2021,
enjoyment of the highest attainable standard of physical and taking into account different study approaches.

Addressing infertility is an important component

of sexual and reproductive health and rights,
and is central to achieving SDG 3 and SGD 5

The highest attainable The ability to decide

standard of physical and the number, timing and
It is central mental health spacing of one’s children
to achieving:

Sustainable Development Sustainable Development

Goal 3: Ensure healthy lives Goal 5: Achieve gender
and promote well-being for equality and empower all
In turn all at all ages. women and girls.
meeting:

Executive summary Introduction Methods Results Discussion Annexes ix

Infertility prevalence estimates, 1990–2021

Methods
These global and regional infertility prevalence estimates were that were reported in the studies, as well as information
generated and reported in accordance with the Guidelines about the study design, study population characteristics and
for Accurate and Transparent Health Estimates Reporting, other factors.
(GATHER), which is widely used for reporting health estimates
of this nature. The process included the following steps : Data analyses were performed to obtain information about
approaches used in estimating infertility, and to generate
A search strategy was developed to identify studies reporting infertility prevalence estimates. These estimates reflected
on infertility prevalence between 1990 and 2021. Several major the pooled findings of the selected studies. The overall
electronic databases were searched, as were websites and lifetime and period prevalence of infertility were estimated,
conference proceedings. and additional analyses provided comparisons of infertility
prevalence in terms of factors such as geographic regions and
Relevant data were extracted from the selected studies. country income levels.
The extracted data included the infertility prevalence rates

Methodological process:

1 Search strategy developed

2 Relevant data extracted

from selected studies

3 Data analyses performed

to obtain information

4 Results

Executive summary Introduction Methods Results Discussion Annexes x

Executive summary

Results
Included studies Global infertility prevalence estimates1
The search identified 12,241 records of potentially relevant Based on data from 1990 to 2021, the 2022 global infertility
studies. Screening of these records led to the selection of prevalence estimates are:
133 studies that met the criteria for inclusion in the systematic • Approximately one in six people have experienced infertility
review. From these, 91 data points were used to generate at some stage in their lives, globally.
pooled 12-month infertility estimates. • Lifetime prevalence of infertility is estimated to be 17.5%
(95% confidence interval [CI]: 15.0, 20.3).
• Period prevalence of infertility is estimated to be 12.6%
(95% [CI]: 10.7, 14.6).

Global infertility prevalence estimates

2022 global infertility prevalence estimates are:

Approximately one
in six people have
experienced infertility
at some stage in their
lives, globally.

17.5% 12.6%
Estimated lifetime Estimated period
prevalence of infertility prevalence of infertility
(95% confidence (95% confidence
interval: 15.0, 20.3). interval: 10.7, 14.6).

Lifetime prevalence is defined as the Period prevalence is defined as the proportion of a

proportion of a population who have population with infertility at a given point or interval
ever experienced infertility in their life. in time, which may be current or in the past.

Regional infertility prevalence estimates The available data indicate that estimated lifetime prevalence
There is some variation in infertility prevalence across regions, of infertility is highest in the WHO Western Pacific Region (23.2%)
but data gaps and overlapping confidence intervals mean that and lowest in the WHO Eastern Mediterranean Region (10.7%).
regional differences identified in this analysis may not be Estimated period prevalence of infertility is highest in the WHO
substantial or conclusive. Some regions had very few studies African Region (16.4%) and lowest in the Eastern Mediterranean
with relevant prevalence estimates, and no studies were Region (10.0%). All confidence intervals for these estimates overlap
identified for the World Health Organization (WHO) South-East based based on 3 or fewer studies, suggesting that the observed
Asia Region. differences may not be substantial or conclusive.

1
Unless otherwise specified, these estimates refer to 12-month period or lifetime prevalence in keeping with the following definition of infertility
adopted by the World Health Organization (WHO): Infertility is a disease of the male or female reproductive system defined by the failure to achieve
a pregnancy after 12 months or more of regular unprotected sexual intercourse

Executive summary Introduction Methods Results Discussion Annexes xi

Infertility prevalence estimates, 1990–2021

Regional infertility prevalence estimates

WHO Eastern Mediterranean WHO European Region
Region
Lifetime infertility prevalence
Lifetime infertility prevalence 16.5% (CI: 14.1, 19.2, n = 18)
10.7% (CI: 3.4, 29.0, n = 3)

Period infertility prevalence

Period infertility prevalence 12.4% (CI: 10.5, 14.6, n = 27)
10.0% (Cl: 5.5, 18.2, n = 3)

WHO Region of the Americas WHO African Region WHO Western Pacific Region
Lifetime infertility prevalence Lifetime infertility prevalence Lifetime infertility prevalence
20.0% (CI: 13.9, 27.9, n = 10) 13.1% (CI: 8.6, 19.4, n = 2) 23.2% (CI: 17.4, 30.2, n = 6)

Period infertility prevalence Period infertility prevalence Period infertility prevalence

10.4% (CI: 7.4, 14.3, n = 5) 16.4% (Cl: 10.0, 25.7, n = 6) 13.0% (CI: 7.8, 20.8, n = 11)

CI = Confidence interval WHO = World Health Organization

n = number of studies No studies were available for the WHO South-East Asia Region

Estimates of infertility prevalence are similar across Lifetime infertility prevalence

countries with different income levels. Lifetime infertility High-income countries
prevalence was 17.8% for high-income countries and 16.5% for
low- and middle-income countries. Period infertility prevalence 17.8%
was 12.6% for high-income countries and 12.6% for low- and 16.5%
middle-income countries. Low- and middle-income
countries

Period infertility prevalence

High-income countries
12.6%
12.6%
Low- and middle-income
countries

Executive summary Introduction Methods Results Discussion Annexes xii

Executive summary

Research gaps, measurement challenges, and implications

for future research
Across the individual studies that contributed data to these Moving forward, researchers must use more consistent,
estimates, reported infertility prevalence varied greatly. This systematic and comprehensive processes to improve the
may partly reflect the many different methods used to measure evidence base relating to infertility prevalence at the global,
infertility. In addition, there is a lack of sufficient studies from regional and national levels.
some regions, as well as variation in definitions and in inclusion
and exclusion criteria in studies estimating infertility, which all
contribute to moderate certainty of pooled estimates.

Research challenges
Across the individual studies that contributed data to these estimates,
reported infertility prevalence varied greatly. This may be due to:

The many different A lack of sufficient Variation in definitions

methods used to studies from in inclusion/exclusion
measure infertility different regions criteria in infertility studies

All of which contribute

to moderate certainty
of pooled estimates.

Recommendations for future infertility prevalence research

1. Estimating prevalence of infertility 3. Reporting estimates

The field needs a standard set of questions for ascertaining Detailed methodological and analytical information
infertility prevalence that could be adopted by Demographic should be provided when reporting estimates of infertility
and Health Surveys and other standard population-based prevalence. When feasible, estimates for total, primary, and
surveys. Questions should be flexible enough to allow for secondary infertility prevalence, with stratification by age
different definitions and approaches in order to facilitate and sex, should be reported.
comparison. At a minimum, 12-month period infertility that
is consistent with the WHO definition should be measured. 4. Making comparisons across studies
Estimates should only be compared where they are as
2. Selecting a methodological approach similar as possible in relation to various study characteristics
When selecting an approach to estimate infertility such as definitions, methodological approaches, and
prevalence, researchers should consider the objectives, data exclusion criteria.
sources, resources, and validity and reliability of data.

Executive summary Introduction Methods Results Discussion Annexes xiii

Infertility prevalence estimates, 1990–2021

Policy and programmatic implications

Valid and reliable estimates of infertility are needed to in relation to the availability, accessibility, and quality of
understand its burden and to facilitate appropriate policy interventions to prevent, diagnose and treat infertility in most
formulation, as well as advocacy, provision, and monitoring of countries. It is anticipated that these estimates will improve our
prevention efforts and fertility care services. These estimates understanding of the prevalence and burden of disease related
clearly show that a large number of people may require to infertility globally and regionally, and will provide a basis for
infertility management and fertility care services in different policy and practice to achieve universal access to fertility care.
regions of the world. Currently, challenges can be observed

Conclusion
Human health and gender equality are central elements of the infertility globally and regionally, a finding that should be used
Sustainable Development Goals, which call on governments to to support the development of policies and practices that
ensure universal access to sexual and reproductive health and will help individuals and couples achieve their desired family
rights. Fertility care is a core part of sexual and reproductive size. Findings also provide insight into how the estimation of
health, and responding to infertility can mitigate gender infertility prevalence can be improved in order to obtain more
inequality. The drive to achieve the Sustainable Development actionable data, including data that allow for more meaningful
Goals therefore must encompass actions to respond more comparisons across settings and time.
effectively to the needs of people with infertility, leaving
no one behind. These estimates show high prevalence of

Executive summary Introduction Methods Results Discussion Annexes xiv

1. Introduction

This section provides background information on how infertility

prevalence has been measured in the past, why there are limitations
to earlier assessments of infertility prevalence, and how this report
advances knowledge and strategic information about infertility
prevalence.

Addressing infertility is an important component of sexual and fertility care services, as well as addressing risk factors for and
reproductive health and rights (SRHR) but has been neglected consequences of infertility. Yet there has been considerable
in the global SRHR agenda. Infertility has devastating societal variability in its previous estimation (Figure 1.1). As many
and health consequences, including social stigma, economic researchers have noted, this variability hinders accurate
hardship, and gender-based violence, as well as poor mental comparisons across regions, populations, and time (7-13).
health (1, 2). Addressing infertility is central to achieving
Sustainable Development Goal (SDG) 3 – Ensure healthy lives Unlike other types of conditions, infertility is defined by the
and promote well-being for all at all ages – and SDG 5 – Achieve absence of an event (i.e., not getting pregnant), usually after a
gender equality and empower all women and girls. defined period of time. The World Health Organization (WHO)
specifies a 12-month duration, defining infertility as “a disease
Furthermore, every human being has a right to the enjoyment of the male or female reproductive system characterized by the
of the highest attainable standard of physical and mental failure to achieve a pregnancy after 12 months or more of regular
health (3). Individuals and couples have the right to decide unprotected sexual intercourse”. (14). However, many studies
the number, timing and spacing of their children (4). Men and have utilized other definitions of infertility that incorporate
women of full age, without any limitation due to race, nationality longer durations, such as 24 or 60 months (11), or involve non-
or religion, have the right to marry and to found a family (5). duration-based definitions to include health conditions that
If infertility is not addressed, it can negate the realization of warrant infertility services or relationship factors, such as single
these essential human rights. Failure to address infertility will persons or same-sex couples as suggested by Zegers-Hochschild
hamper global efforts to ensure universal access to sexual and et al.(15). In addition, while some researchers propose that
reproductive health and rights. Consequently, urgent efforts time-to-pregnancy (TTP) be used to ascertain infertility (16, 17),
are required to improve the prevention, management, and many studies use self-reported or constructed binary measures
treatment of infertility worldwide (6). of infertility (12). In addition, studies vary in their definition and
application of numerators and denominators, study designs,
Understanding the magnitude of infertility is critical for survey instruments, and analytic methods, all of which can affect
monitoring, assessing, and improving equitable access to quality conclusions regarding the true magnitude of infertility (13, 18).

Executive summary Introduction Methods Results Discussion Annexes 1

Infertility prevalence estimates, 1990–2021

Figure 1.1. Wide-ranging findings from previous estimates of infertility

Previous estimates of infertility prevalence suggested 10.5% experienced secondary infertility, (defined by the
that the number of individuals or couples affected by authors as the inability to have an additional live birth).
infertility ranged from 48.5 million couples globally (19) to A 2007 literature review of 25 population surveys found
186 million ever-married women in developing countries that the prevalence of infertility when defined by a duration
alone (20). In addition, a 2012 study that used data from of 12 months or more ranged from 3.5% to 16.7% in more
277 demographic health surveys and Bayesian hierarchical developed nations and from 6.9% to 9.3% in less developed
modeling to generate estimates for 190 countries and nations (21). In 2016, another review and meta-analysis
territories using a duration of five years or more to define of 52 studies reported a mean infertility prevalence of
infertility (19) concluded that 1.9% of women exposed to the 10% worldwide, with pooled prevalence lowest and highest
risk of pregnancy experienced primary infertility, (defined in Australia and Africa, respectively (22).
by the authors as the inability to have any live birth), and

Consequently, it is important to assess how differences generating global and regional estimates. The estimates
in definitions and study designs may affect infertility presented in this report therefore aimed to generate global
estimates. Understanding the impacts of these variations can and regional estimates while taking into account variation
provide important insights for developing a standardized in measurement approaches. The work was guided by two
methodological approach. In addition, pooling of prevalence research questions shown in Figure 1.2.
data through a meta-analytical approach is necessary for

Figure 1.2. Key questions addressed in these estimates

1. W
 hat approaches have been used to estimate 2. W
 hat is the contemporary prevalence of infertility
prevalence of infertility among representative globally, and how do prevalence estimates differ by
populations? methodological approach and study design?

Understanding the magnitude of infertility is

critical for monitoring, assessing, and improving
equitable access to quality fertility care services,
as well as addressing risk factors for and
consequences of infertility.

Executive summary Introduction Methods Results Discussion Annexes 2

2. Methods

This section describes how studies reporting infertility prevalence data

were identified. It also describes which items of data were extracted
from each study and explains how data were analyzed.

These estimates were reported according to the Guidelines Reporting Items for Systematic Reviews and Meta-Analyses
for Accurate and Transparent Health Estimates Reporting (PRISMA) (24) and the Meta-analysis of Observational Studies
(GATHER) (23). A systematic review and meta-analysis in Epidemiology (MOOSE) (25), based on a pre-registered
was conducted in accordance with the updated Preferred protocol (26).

2.1 Data sources and identification of

relevant studies
To identify peer-reviewed publications, the following electronic contemporary. Second, an analysis of trends in infertility
databases were searched: PubMed (US National Library of prevalence in 190 countries and territories had found that
Medicine), Web of Science (Clarivate Analytics), CINAHL (EBSCO), levels of infertility in 2010 were similar to those in 1990 in
Family & Society Studies Worldwide (EBSCO), Public Health most regions of the world (19). Third, it was necessary for
(ProQuest), and Google Scholar. Relevant articles and reviews the range of time to be long enough to capture all relevant
were hand-searched. To identify grey literature, a search was methodological approaches. Finally, Schmidt and Münster (8)
conducted of electronic databases (Public Health [ProQuest] conducted a review of the prevalence of infertility and its
and ProceedingsFirst [OCLC]), relevant websites, and conference measurement in “industrialized countries” spanning 1970
proceedings. Additionally, experts in the field were consulted. to 1992, and the estimates presented here were intended
to extend and expand on this previous work. Records
The search strategy included terms related to infertility retrieved from all searches underwent title, abstract and
(e.g., infertility, subfertility, infecundity, childlessness) and full-text screening, with the aid of reference management
estimation (e.g., estimate, prevalence). Searches were limited software. Two members of the research team independently
to between 1990 and 2021 with no language restrictions. assessed whether studies were eligible for inclusion, based
The year 1990 was selected as the lower bound cut-off for on predetermined criteria (Figure 2.1), with disagreements
four reasons. First, this ensured that the estimates were resolved through discussion to reach consensus.

Executive summary Introduction Methods Results Discussion Annexes 3

Infertility prevalence estimates, 1990–2021

Figure 2.1. Inclusion criteria

General population and clinic-based studies were included • presented original research using primary or secondary
if they met all of the following criteria: data; and
• used one of the following study designs: cross-sectional,
• designed to be a representative sample of a general cohort, case-control (if the control group was a
population of women and/or men; representative sample of the general population and
• reported estimates of the prevalence or cumulative the disease of interest [i.e., cases] was not infertility),
incidence of infertility; or randomized trial (if the study reported an overall
• collected data during or after 1990; estimate for a representative sample of the general
population at baseline, before any interventions were
• specified, in their definition of infertility, a duration of administered).
at least six months in which pregnancy is not reached
or defined infertility as a subjective evaluation of one’s
difficulty conceiving or maintaining a pregnancy;

Studies were considered representative if they recruited, based a clinic that serves the general population (i.e., a primary health
on their study design, all eligible members of a population (i.e., care clinic or an obstetrics and/or gynecology clinic) and were
census) or applied probability-based sampling. Clinic-based representative of the clinic population as a whole.
studies that applied consecutive sampling for 12 or more
months were defined as a census of the clinic population and Studies were excluded based on criteria shown in Figure 2.2.
thus were considered eligible for inclusion. Furthermore, for Duplicate publications that generated multiple estimates of
clinic-based studies to meet the requirement of representing a prevalence of infertility using the same data source, definition
general population, their samples had to have been drawn from of infertility, and approach to estimation were eliminated.

Figure 2.2. Exclusion criteria

Studies were excluded if they met any of the following • measured childlessness without an intention to estimate
criteria: infertility (e.g., a combined measure of voluntary and
involuntary childlessness or a measure that did not
• reported cause-specific prevalence of infertility only, distinguish reasons for involuntary childlessness);
such as tubal factor infertility, or male-factor or female- • included menopausal and/or surgically sterile individuals
factor infertility; in their numerator, which would inflate the numerator
• estimated only the proportion seeking fertility treatment with individuals who have completed their reproductive
or receiving a diagnosis of infertility; life span either naturally or surgically;
• did not use individuals as the unit of analysis (e.g., • did not define their measure of infertility; or
studies in which the prevalence was calculated based on • reported results only as an abstract or unpublished data.
total pregnancies, rather than individuals);

Executive summary Introduction Methods Results Discussion Annexes 4

2. Methods

2.2 Data extraction

Data that were extracted from each study are shown in Figure 2.3.

Figure 2.3. Data extracted from studies

Study characteristics • income level of the country where the study was
conducted, according to World Bank classification at
• study design
time of analysis (2021) (27) (high-income country [HIC],
• data collection details low- and middle-income country [LMIC])
• methodological approach or approaches (prospective • Geographic region (based on World Health Organization
time to pregnancy [TTP] design, retrospective TTP design, regions: African Region, Region of the Americas,
current duration design, self-reported infertility measure South-East Asia Region, European Region, Eastern
[direct], constructed infertility measure [indirect], and Mediterranean Region, Western Pacific Region)
undetermined)
• infertility estimates (reported proportion of study
Definitional characteristics
sample with infertility)
• type of prevalence (period, lifetime)
Study population characteristics • numerator (duration-only, duration and treatment,
self-perceived infertility, intention to conceive)
• sample type (population-based sample, clinic-based (also categorized based on whether intentions were
sample) considered or not, i.e., trying to conceive)
• sex of respondent (female respondent, male • denominator (individuals regardless of risk of pregnancy,
respondent, combined) individuals at risk of pregnancy regardless of intentions,
individuals attempting to conceive)

For studies that included data collected both before and after was extracted. In instances where necessary information was
1990, only estimates calculated from data collected during or not reported in a manuscript, an effort was made to obtain the
after 1990 were extracted. For studies that presented estimates information by e-mailing the corresponding author.
for multiple time periods after 1990, the most recent estimate

2.3 Data analysis

2.3.1 Risk of bias assessment
The risk of bias for each study was assessed using a risk of bias classified as high risk. An overall summary score was generated
tool proposed by Hoy et al. (28) which was slightly modified from the sum of the eight individual items (one point awarded
to better fit with infertility definitions assessed for these for each item labeled as low risk). The overall summary score
estimates (Annex 1). The tool includes eight items assessing was divided into the following tertiles: 1) low risk of bias:
external and internal validity. For each item, studies were rated 6–8 points, 2) moderate risk of bias: 3–5 points, and 3) high
as either low or high risk. Studies that provided insufficient risk of bias: 0–2 points. The potential for publication bias was
information to permit a judgement for a given item were assessed through funnel plots.

2.3.2 Descriptive analysis

First, the methodological approaches used to estimate female reproductive system defined by the failure to achieve
the prevalence of infertility were identified and described. a pregnancy after 12 months or more of regular unprotected
Next, the overall number of studies was reported, and the sexual intercourse,” hereafter referred to as “12-month
numbers of studies by study descriptor variables. In analyzing infertility.” The number of studies and range of estimates of
the estimates of infertility prevalence, the focus was on the 12-month infertility overall and by study descriptor variables
definition adopted by WHO (14), i.e., “a disease of the male or were examined and reported.

Executive summary Introduction Methods Results Discussion Annexes 5

Infertility prevalence estimates, 1990–2021

2.3.3 Meta-analysis and meta-regression

Meta-analysis was applied to estimate period and lifetime the minimum value in sensitivity analyses. Random-effects
prevalence of 12-month infertility overall and stratified by meta-analysis models were used to generate pooled estimates,
income level, region, respondent type, and methodological 95% confidence intervals, I2 statistics (i.e., proportion of
approach. The standard error (SE) of each study’s infertility total variability in point estimates that can be attributed to
prevalence estimate were calculated either: 1) by extracting heterogeneity), and forest plots. A decision to present pooled
the SE directly or calculating from the 95% confidence interval estimates was not solely based on I2 values, but was informed
(CI) ((upper interval–lower interval)/3.92), or when this by consideration that higher I2 values are inevitable where
information was not available, 2) a SE was calculated based on sample sizes are large, and standard errors are precise (29),
the formula for obtaining a standard error from a proportion which was consistent with the included studies. Pooled
(p) ([SQRT(p*(1-p))/N]). Sensitivity analyses were further estimates were stratified by period or lifetime infertility,
conducted to examine the influence of studies that used an income classification (i.e., HIC or LMIC), geographic region,
approximated SE compared to studies in which the SE could be methodological approach, and respondent’s sex. Using a
estimated directly. random-effects meta-analysis model, funnel plots were derived
of the logit transformed prevalence estimates against their
Estimates of 12-month infertility were transformed using the standard errors.
logit function ([ln(p/(1-p)]). Corresponding SEs were logit
transformed using the delta method [SQRT((1/(p*(1-p))2)*(SE2)). Meta-regression using restricted maximum likelihood was
To ensure independence across studies for the meta-analyses applied to generate adjusted period and lifetime prevalence
and assess the sensitivity of analytic choices on selection of estimates of 12-month infertility after accounting for region,
estimates, the maximum or the minimum lifetime and period methodological approach, whether or not numerator included
infertility prevalence estimates were selected for studies in intentions, denominator categories, and risk of bias score. The
which multiple estimates were presented (either in the same covariates in the model were chosen based on variables of
record or a duplicate record, which for the purposes of the interest in estimation (i.e., region, methodological approach)
meta-analysis was defined as an estimate that was generated or having a sufficient number of studies across each variable
from the same data source). categorization and region. The exponentiated regression
coefficient obtained from the meta-regression of the logit
Pooled estimates for studies were generated using the transformed infertility prevalence estimates provides odds
maximum value of the prevalence estimate for studies ratios for a given unit change in the covariate. Stata 16.1 was
presenting multiple estimates, then the same was done using used to conduct meta-analyses and meta-regression (30).

2.3.4 Sensitivity analyses

In meta-analysis, sensitivity analysis was conducted by versus highest-quality studies (risk of bias > 6), and
estimating pooled lifetime and period prevalence of 12-month 3) studies from the general population only versus all studies
infertility stratified by 1) maximum versus minimum values (i.e., population- and clinic-based studies).
for linked studies or SE calculation assumptions, 2) all studies

2.3.5 Rating certainty of evidence

The GRADE framework (31) was used for rating the certainty instead, inconsistency (heterogeneity) was also informed by
in inferences drawn from the estimates. Specifically, the visual inspection of forest plots. Specifically, overlap of point
GRADE guidance relating to overall prognosis was adopted estimates and 95% confidence intervals across the individual
(32) as it is the most applicable to questions on prevalence of a studies was examined. Indirectness was assessed by comparing
condition. The GRADE framework rates certainty in inferences the research questions posed by the authors of individual
as high, moderate, low, or very low. Certainty may decrease studies to those posed in the investigation that yielded these
due to concerns about risk of bias, inconsistency, indirectness, estimates. For imprecision, the width of the 95% CI of the
imprecision, and publication bias. For concerns about risk pooled estimates was assessed. Given that the evidence in
of bias, the sensitivity analyses conducted by the research these estimates will be used in various contexts and settings,
team (exclusion of high risk of bias studies to determine if a formal threshold for rating imprecision was not determined;
conclusions from pooled estimates differ) were used. For rather, consideration was given to whether the upper and lower
concerns about inconsistency, reliance was not placed on the I2 bounds of the 95% CI are sufficiently close to the reported
statistics, as these tend to overestimate heterogeneity amongst pooled estimate. Finally, publication bias was assessed through
studies with large sample sizes and number of events (29); visual inspection of funnel plots.

Executive summary Introduction Methods Results Discussion Annexes 6

3. Results

This section presents findings from the literature search and the
analysis of data extracted from relevant studies.

A total of 133 studies met inclusion criteria, and these studies pool estimates. Overall pooled lifetime prevalence of infertility
reflected five distinct methodological approaches to estimating was 17.5%, and overall pooled period prevalence of infertility
infertility. The findings from all studies that reported 12-month was 12.6%.2 There was some variation in infertility prevalence
infertility prevalence (n = 84) were pooled to calculate pooled across different geographic regions. Unless otherwise specified,
lifetime and period prevalence of infertility at the global and prevalence refers to 12-month infertility, consistent with the
regional levels. Relevant data points from studies were used to WHO definition.

Research process
133 infertility prevalence studies
reflecting 5 distinct methodological
approaches

Relevant data points pooled to calculate

infertility prevalence

Infertility is a disease of the male

or female reproductive system
defined by the failure to achieve
Pooled lifetime Pooled period a pregnancy after 12 months
prevalence of prevalence of or more of regular unprotected
infertility was infertility was sexual intercourse

17.5% 12.6%

2
Unless otherwise specified, these estimates refer to 12-month period or lifetime prevalence in keeping with the following definition of infertility
adopted by the World Health Organization (WHO): Infertility is a disease of the male or female reproductive system defined by the failure to achieve
a pregnancy after 12 months or more of regular unprotected sexual intercourse.

Executive summary Introduction Methods Results Discussion Annexes 7

Infertility prevalence estimates, 1990–2021

A PRISMA flow diagram illustrating the literature search, article selection and final included studies is shown in Figure 3.1.

Figure 3.1. Identification of studies via databases and other methods

Records identified through Records identified through Additional records identified

database searching database searching through other sources
Sept.29 2020 Updated search-March 11, 2021
1. Preliminary searches: n = 91
1. PubMed: n = 8 790 1. PubMed: n = 425 2. Experts and Systematic review
2. Web of Science (Clarivate): n = 1 371 2. Web of Science (Clarivate): n = 92 advisory board: n = 23
3. CINAHL with Full Text (EBSCO): n = 1 999 3. CINAHL with Full Text (EBSCO): n = 115 3. Reference lists: n = 138
Identification

4. Family & Society Studies Worldwide 4. Family & Society Studies 4. Organizations’ websites: n = 10
(EBSCO): n = 1 039 Worldwide (EBSCO): n = 31 5. Proceedings: n = 24
5. Public Health (ProQuest): n = 1 441 5. Public Health (ProQuest): n = 39 6. Google Scholar: n = 920
6. ProceedingFirst (OCLC): n = 322 6. ProceedingFirst (OCLC): n = 0
N = 1 206
N = 14 962 N = 702

Duplicate records removed

Total records
before screening
N = 16 870
N = 4 629

Records with title and

Records excluded
abstract screened
N = 11 275
N = 12 241
Screening

Records sought for

Records not retrieved
full text retrieval
N = 11
N = 966

Full-text records Full text records

assesed for eligibility excluded with reasons
N = 955 N = 822
Ineligible outcome: n = 168
Sub-group study: n = 229
Non-representative sampling: n = 126
Studies included in review
Included

N = 133 Other (mostly duplicates and review articles): n = 82

Linked to other studies: n = 38
Data collection before 1990: n = 40
Definition of infertility not reported or did not meet inclusion criteria: n = 35
Wrong study design: n = 4

Executive summary Introduction Methods Results Discussion Annexes 8

3. Results

3.1 Description of studies

The literature search yielded 16 870 records. Following the The vast majority of studies were cross-sectional in design
removal of duplicates, 12 241 unique records were screened. (n = 115). Thirteen studies used a cohort study design and
The screening resulted in the selection of 133 studies for five used a case-control study design for which only data for
inclusion in this study (Annex 2). An overview of the study the control groups were extracted, which were representative
characteristics and infertility estimates for each study can be samples of a general population.
found in Annex 3.

Total number Most common study design:

cross-sectional (86%)
Most common
infertility definition:
of studies 12-month definition

analysed:
133
Age of study Relationship status of study populations:
populations:
Only
heterogeneous included
participants
who were married or 66 Did not
• Many studies included • Some studies had no studies
participants of age limit in union 11 report the
“reproductive age”, relationship
• Some studies did not status of
defined in different
report age Did not 53 respondents
ways restrict studies
participation
• Some studies had
in accordance with
an age limit beyond
relationship status
reproductive age

The sample in 85 studies included individuals of reproductive Sixty-six studies restricted their sample to individuals who were
age, which was defined differently across studies but often married or in union, 53 studies did not restrict their sample by
confined to individuals aged 15–49 or 20–44 years. relationship status, and a few studies reported both estimates
separately. Eleven studies did not report the relationship status
Eighteen studies provided a lower age limit without an upper of respondents. Some studies explicitly or implicitly excluded
age limit and/or an age limit that extended beyond reproductive individuals not engaged in heterosexual intercourse. One study
age. Fifteen studies limited the sample to a single age or a reported the percent of respondents self-identifying as gay,
smaller age range that captures women in different stages of lesbian, or bisexual.
their reproductive life (e.g., 20–34 years, 30–49 years). For seven
studies, all measuring lifetime prevalence of infertility, the sample The most common definition applied to estimates of
included individuals beyond reproductive age. Ten studies did infertility prevalence was a 12-month definition of infertility
not report the age range of respondents in their analytic sample. in 101 studies. Thirty studies applied a 24-month definition of
Three studies reported estimates for two different age groupings infertility while 14 studies applied a demographic 60-month
and are thus represented in multiple tallies. definition of infertility. Twenty-nine studies applied definitions

Executive summary Introduction Methods Results Discussion Annexes 9

Infertility prevalence estimates, 1990–2021

with durations other than 12, 24, or 60 months (e.g., 6 months, The overall risk of bias was low for 77.4% of studies, moderate
36 months) or with no duration (self-perceived infertility). for 21.1% of studies, and high for 1.5% (Figure 3.2). For
Among the studies that defined infertility by duration, all five out of eight individual items assessed, at least 87.2%
studies measured infertility in months with no studies of studies were rated as low risk. Only one item, the item
measuring infertility in menstrual cycles. Many studies reported measuring the likelihood of non-response, had more than half
estimates for multiple definitions of infertility. Sixty studies (54.9%) of studies rated as high risk. The funnel plots, which
reported total infertility estimates (i.e., primary and secondary were used to examine publication bias, were symmetrical for
infertility combined in a single estimate), and 34 studies studies reporting estimates of lifetime and period prevalence
reported total, primary, and secondary infertility estimates. The of infertility. See more details here: https://doi.org/10.1093/
remaining 39 studies reported some other combination of total, hropen/hoac051.
primary, and/or secondary infertility estimates.

Figure 3.2. Risk of bias of included studies

Low Risk High Risk

Was the sampling frame a true or close

representation of the target population?
70.7% 29.3%

Was some form of random selection used to

select the sample, OR was a census taken?
87.2% 12.8%

Was the likelihood of

non-response bias minimal?
45.1% 54.9%

Were data collected directly from

the subject (as opposed to proxy)?
97.0% 3.0%

Was an acceptable case

definition used in the study?
88.0% 12.0%

Was the study instrument/measure

shown to have reliability and validity?
50.4% 49.6%

Was the same mode of data

collection used for all subjects?
95.5% 4.5%

Were the numerator(s) and denominator(s)

for the parameter of interest appropriate?
91.7% 8.3%

Low Risk Moderate Risk High Risk

Summary item on the overall risk of study bias

(based on number of “yes” responses)
77.4% 21.1% 1.5%

Executive summary Introduction Methods Results Discussion Annexes 10

3. Results

3.2 Methodological approaches for

estimating infertility prevalence
The review comprised studies that fell into six methodological not be determined based on the information provided in the
categories: 1) prospective TTP design, 2) retrospective TTP manuscript. Table 3.1 provides a description of each approach
design, 3) current duration design, 4) self-reported infertility and common applications based on the studies included in the
measure (direct), 5) constructed infertility measure (indirect), review. Table 3.2 reports the number of studies and range of
and 6) undetermined. For 13 studies, the approach could estimates by methodological approach.

Methodological approaches used in infertility prevalence studies

The review comprised studies that fell into 6 methodological categories:

Prospective
1 TTP design
Least common
approach in the
included studies:
prospective TTP
design
Retrospective
2 TTP design

Most common
Current
3 duration design
methodological
approaches in the
included studies:
self-reported
Self-reported infertility measure
and retrospective
4 infertility
TTP design
measure (direct)

Constructed
5 infertility measure
(indirect)

6 Undetermined

Executive summary Introduction Methods Results Discussion Annexes 11

Infertility prevalence estimates, 1990–2021

Table 3.1. Five approaches to measuring infertility prevalence identified from the systematic review

Prospective time- Retrospective Current duration Self-reported Constructed

to-pregnancy time-to-pregnancy design infertility infertility
designa design measure (direct) measure (indirect)

Description • Participants • Participants are • Participants are Participants are • I nfertility status
are enrolled asked to recall the enrolled during queried directly is determined
prior to period PUI or pregnancy a current PUI or about their ability based on the
of unprotected attempt time pregnancy attempt. to conceive either presence or
intercourse (PUI) prior to becoming within a specified absence of a
• Current duration
(incident cohort) pregnant duration of time pregnancy or
(CD) is calculated
or during a period (pregnancy-based (e.g. 12 months) live birth among
as the interval
of unprotected approach). or based on couples exposed
between when the
intercourse their subjective to conception for
• Alternatively, PUI or pregnancy
(prevalent evaluation. a defined period.
participants may attempt began and
cohort).
be asked about date of interview.  xposure to
•E
• Participants are a PUI and/or conception
• CD values are
followed until time-to-pregnancy is inferred
used to estimate
pregnancy, (TTP) in a specified from survey
a summary TTP
infertility time regardless questions and/
distribution for
treatment, or of outcome or a reproductive
the population
study conclusion (historical calendar.
using survival
(administrative prospective
methods and under
censoring). approach).
certain analytic
assumptions.

Sample • [For those • How long had/ • [For those at risk • Have you ever Constructed
questions planning to have you been of pregnancy at experienced based on a series
for querying conceive] Are you trying to become interview]: Series a period of at of questions or
respondents pregnant (asked pregnant? of questions on least 12-months reproductive
at specified dates of last use where you calendar on
• How many months
intervals during of contraception, were having relationship status,
did you have
follow-up)? pregnancy, or birth. unprotected birth history,
regular intercourse
Current duration intercourse contraceptive
• [For those without
is calculated from (or attempting use, and, in some
planning to contraception
start of at-risk to become instances, sexual
conceive] before you
interval to date of pregnant) but activity and desire
pregnancy is became pregnant?
interview. did not become to have another
ascertained by
pregnant? child.
pregnancy testing • [Among those at
over follow-up risk of pregnancy • Have you and
period at interview] How a partner ever
long have you been had difficulty
trying to become conceiving?
pregnant? (number
of months or years)

Common • Assess the • Estimate fecundity • Generate • Estimate Generate

research biologic capacity or measure population- infertility population-
objectives for reproduction infertility based estimates prevalence based estimates
(i.e., fecundity) prevalence of infertility of infertility
• Assess
prevalence prevalence
• Examine the • Identify and/ association
with nationally
relationship or examine risk • Identify and/ between
representative
between risk factors, which or examine risk infertility and
demographic
factors on need to be factors, which need risk factors,
and reproductive
fecundity anchored around to be anchored outcomes, and/
health survey data
the start of the around the start or treatment
PUI or pregnancy of the PUI or seeking
attempt pregnancy attempt behavior

Executive summary Introduction Methods Results Discussion Annexes 12

3. Results

Prospective time- Retrospective Current duration Self-reported Constructed

to-pregnancy time-to-pregnancy design infertility infertility
designa design measure (direct) measure (indirect)

Common applications forb:

Type of Period prevalence Period or lifetime Period prevalence Period or lifetime Period prevalence
prevalencec prevalence prevalence

Duration • 12-months • 12-months • 1 2-months • 1 2-months • 1 2-months

cut-off for
• 24-months • 24-months • 2 4-months  o duration
•N • 6 0-months
infertility
(subjective
measure)

Pregnancy Always considered Sometimes Sometimes Sometimes Commonly not

intentions considered considered considered considered
considered in
numerator

Denominator Those attempting Those ever at risk Those at risk of Ever and not at Ever and not at
considered to conceive of pregnancy or pregnancy or risk of pregnancy risk of pregnancy
attempting to attempting to (e.g. all women of (e.g. all women of
conceive conceive at time of reproductive age) reproductive age)
interview

a
The prospective time-to-pregnancy design approach is considered the gold standard.
b
Common applications are summarized based on the studies included in the systematic review.
c
Period prevalence is defined as the proportion of individuals/couples with infertility at a given point or interval in time, which may be current or past
depending on the study aims. Lifetime prevalence is defined as the proportion of individuals/couples who have ever experienced infertility in their life.

Table 3.2. Range of 12-month period and lifetime infertility prevalence estimates by methodological
approach and other study descriptors

Study Characteristics Number of studiesa Number and range of 12-month total infertility prevalence
estimates (%)b

All studies (studies Number Period Number Lifetime

with 12-month of period Prevalence of lifetime prevalence
estimates) estimatesc estimatesc

Total 133 (84) 69 1.6-34.0 65 3.3-39.7

Methodological approaches

Prospective TTP design 3 (3) 3 13.6-28.0 - -

Retrospective TTP design 34 (24) 25 5.0-32.0 15 3.3-35.3

Current duration design 6 (5) 10 9.4-34.0 - -

Self-reported infertility measure 61 (39) 16 4.0-18.0 45 4.2-39.7

Constructed infertility measure 23 (8) 12 6.0-17.0 - -

Undetermined 13 (6) 3 1.6-13.3 5 10.1-20.9

Definitional characteristics

Numerator (Duration only)

Intentions includedd 65 (46) 22 7.0-32.0 42 4.2-39.7

Intentions not considered 61 (37) 44 1.6-34.0 14 3.3-35.3

Numerator (Duration and/or receipt of care included)

Intentions included⁴ 8 (7) 2 12.0-12.3 8 11.0-26.0

Intentions not considered 2 (2) 1 18.0 1 35.0

Executive summary Introduction Methods Results Discussion Annexes 13

Infertility prevalence estimates, 1990–2021

Study Characteristics Number of studiesa Number and range of 12-month total infertility prevalence
estimates (%)b

All studies (studies Number Period Number Lifetime

with 12-month of period Prevalence of lifetime prevalence
estimates) estimatesc estimatesc

Numerator (Subjective evaluation with or without duration)

Intentions includedd 10 (1) - - 3 11.4-16.4

Intentions not considered or 10 (1) 1 7.74 - -

unknown

Denominator

All regardless of risk of pregnancy 74 (41) 19 1.6-17.0 35 3.3-35.0

Ever at risk of pregnancye 37 (26) 30 4.2-34.0 13 8.2-35.3

Attempting to conceived 40 (30) 20 9.4-32.0 17 5.8-39.7

Study population characteristics

Sample type

General population-based 118 (71) 47 1.6-34.0 65 3.3-39.7

Clinic-based 15 (13) 22 5.0-28.0 - -

Sex of respondent

Female 109 (72) 54 1.6-34.0 56 3.3-39.7

Male 10 (10) 5 7.0-15.3 9 8.2-21.8

Combinedf 18 (9) 10 4.2-28.0 - -

Not reported 1 (-) - - - -

Income Level g

High-income countries 70 (55) 43 5.0-34.0 52 4.2-35.3

Low- and Middle-income countries 65 (29) 26 1.6-32.0 13 3.3-39.7

Regionh

African Region 24 (8) 6 9.5-32.0 4 9.3-15.8

Eastern Mediterranean Region 15 (6) 3 5.2-15.2 4 3.3-21.2

European Region 47 (37) 32 5.0-34.0 25 9.0-31.8

South-East Asia Region 12 (-) - - - -

Region of the Americas 24 (15) 16 4.0-15.7 15 4.2-35.3

Western Pacific Region 29 (19) 12 1.6-28.0 17 8.2-39.7

a
Some studies reported multiple prevalence estimates by applying different definitional or study population characteristics. In these instances,
studies were included in more than one tally.
b
12-month estimates of resolved and unresolved infertility. Outlier 12-month infertility estimate is not reported in Table 3.2. Outliers were deter-
mined based on their magnitude and justification in the respective studies regarding their ability to capture infertility.
c
Some studies reported multiple prevalence estimates by applying different definitional or study population characteristics. In these instances,
multiple estimates from a single study may be included in the same tally.
d
Includes individuals wanting a child and/or trying to conceive.
e
Includes any individual ever at risk of a pregnancy. May include studies that used marital status as a proxy for being at risk of pregnancy.
f
Includes studies that reported estimates for male and female respondents or couple respondents.
g
Defined based on World Bank classifications at the time of the systematic review (The World Bank. Countries and economies. 2021; Available from:
https://data.worldbank.org/country).
h
Defined based on World Health Organization regional groupings (World Health Organization. 2022; Available from: https://www.who.int/about/
who-we-are/regional-offices).

Executive summary Introduction Methods Results Discussion Annexes 14

 3. Results

Of the studies included in these estimates, the self-reported design approach, where participants were recruited from
infertility measure was applied most often followed by the premarital and preconception clinics. The self-reported
retrospective TTP design. The prospective TTP design was the infertility measure approach was the most common approach
least used approach. Six studies reported multiple estimates applied in studies conducted in the WHO African Region, WHO
generated by different approaches and are thus included in the Region of the Americas, and WHO Western Pacific Region. The
count for multiple approaches. Three studies combined two self-reported infertility measure approach was also commonly
approaches to generate a single estimate. In these instances, applied in the WHO European Region; however, in this region,
these studies were categorized based on the primary approach the retrospective TTP design approach was the most widely
used for generating the prevalence estimate. applied approach. The constructed infertility measure was
the most common approach used in studies conducted
Across studies, period prevalence was measured using all in the WHO South-East Asia Region and the WHO Eastern
approaches, whereas lifetime prevalence was mainly measured Mediterranean Region, and was also commonly applied in
using the self-reported infertility measure approach. The studies in the African Region and the Region of the Americas.
12-month definition was the most common definition applied
across all approaches except for the constructed infertility The majority of 12-month infertility estimates were based on
measure for which a 5-year definition was more commonly self-reported infertility measures and retrospective TTP designs
applied. (Table 3.2 and Figure 3.3). Across approaches, duration-based
methods (prospective TTP, retrospective TTP, and current
Use of a retrospective TTP design was more common in high- duration designs) showed larger period estimates and ranges
income countries, particularly Europe, whereas use of the of 12-month infertility (5.0–34.0%) compared with self-reported
constructed infertility measure approach was more common in and constructed measures (4.0–18.0%). Lifetime estimates of
low- and middle-income countries. The self-reported infertility 12-month infertility were available only for retrospective TTP
measure approach was widely applied in studies conducted (3.3–35.3%) and self-reported measures (4.2–39.7%) and were
in both HIC and LMIC. China was the only country in which comparable.
we were able to identify a study that used a prospective TTP

Figure 3.3. Lifetime and period infertility prevalence by methodological approach

45

40

35
Infertility prevalence (%)

30

25

20

15

10

Prospective Retrospective Current duration Self-reported Constructed Undermined

TTP TTP
Methodologic approach
Lifetime Period
TTP = time to pregnancy

Executive summary Introduction Methods Results Discussion Annexes 15

Infertility prevalence estimates, 1990–2021

3.3 Definitional characteristics

3.3.1 Type of prevalence
Eighty-four studies reported a period prevalence and 58 reported lifetime prevalence (Table 3.2). Period and lifetime estimate
a lifetime prevalence with some reporting both. One hundred ranges of 12-month infertility were both wide and comparable to
thirty-four estimates of 12-month infertility were extracted from one another.
84 studies of which 69 were period prevalence and 65 were

3.3.2 Numerator
The majority of studies overall and those reporting 12-month subjective evaluation (i.e., perceived infertility) with or without a
infertility estimates used a numerator defined by duration only specified duration. The range of period and lifetime estimates of
(Table 3.2). Among duration-only estimates, about half included infertility among studies that defined the numerator by duration
intentions (mainly defined as those trying to conceive) in the only did not vary considerably by whether the numerator
numerator and the other half did not. Some studies reported considered intentions (7.0–32.0%, 4.2–39.7%, respectively) or
both. A few studies incorporated duration and receipt of care did not consider intentions (1.6–34.0% with one outlier removed,
in the numerator. Twenty studies used a numerator defined by 3.3–35.3%, respectively).

3.3.3 Denominator
Over half of the studies included individuals regardless of their categories was more evenly divided than for all studies. Period
risk of pregnancy in the denominator. The remaining studies infertility estimate ranges were lower when the denominator
were split in how they defined their denominator between those included individuals regardless of risk (1.6–17.0%) compared
ever at risk of pregnancy and those attempting to conceive to individuals ever at risk (4.2–34.0%) or individuals attempting
(Table 3.2). Some studies provided multiple estimates in their to conceive (9.4–32.0%). Lifetime infertility estimates were
publication using different denominators. Among studies relatively similar across denominator categorizations.
reporting 12-month estimates, the distribution among the three

Executive summary Introduction Methods Results Discussion Annexes 16

 3. Results

3.4 Study population characteristics

3.4.1 Sample type
Most studies drew their sample from the general population general population studies compared to clinic-based studies.
(n = 118) whereas only 15 studies drew their sample from a clinic There were no lifetime estimates of 12-month infertility available
population. Period infertility estimate ranges were similar for for clinic-based studies.

3.4.2 Sex of respondents

A majority of studies included estimates based on female responses from both male and female respondents or from
respondents (n = 109) while only 10 studies included estimates couples. One study did not report the sex of respondents. The
based on male respondents (five studies that included separate range for period and lifetime estimates of 12-month infertility
estimates for female and male respondents were counted in was smaller and lower for male respondents compared to female
both tallies). Eighteen studies included estimates that combined respondents.

3.4.3 Income level

The proportion of study populations from HIC and LMIC was studies presented estimates from both HIC and LMIC. Overall,
similar at 51.9% and 48.1% respectively, whereas 65.5% of the the range of period and lifetime estimates of 12-month infertility
12-month infertility estimates were from HIC (Table 3.2). Two were similar within and across HIC and LMIC (Figure 3.4)

Figure 3.4. Lifetime and period infertility prevalence by country income level

45

40

35
Infertility prevalence (%)

30

25

20

15

10

High-income Low- and middle-income

Country income level

Lifetime Period

Executive summary Introduction Methods Results Discussion Annexes 17

Infertility prevalence estimates, 1990–2021

3.4.4 Regional availability of studies 3.4.5 Ranges of reported estimates

The European Region was the region represented in the greatest Overall, period infertility prevalence estimate ranges were
proportion of studies (35.3% of the total number of studies). The largest for the African (9.5-32.0%), European (5.0-34.0%), and
Eastern Mediterranean Region and South-East Asia Region were Western Pacific regions (1.6-28.0%) compared to the Region of
the least represented regions with 11.3% and 9.0% of the total the Americas (4.0-15.7%) and Eastern Mediterranean Region
number of studies, respectively (Table 3.2). The regions reporting (5.2-15.2%). Lifetime infertility prevalence estimate ranges
the greatest number of 12-month infertility estimates were the were largest for the Region of the Americas (4.2-35.3%), the
European Region, the Region of the Americas, and the Western European Region (9.0-31.8%), and the Western Pacific Region of
Pacific Region. Very few 12-month estimates were available for the (8.2-39.7%), and smallest for the African Region (9.3-15.8%)
the African and Eastern Mediterranean regions, and no 12-month (Table 3.2).
estimates were available for the South-East Asia Region.

Regional availability of data and wide ranges of 12-month infertility estimates

Region most represented Regions least represented in analyzed studies:

in analyzed studies:
WHO Eastern WHO South-East
WHO European
Mediterranean Asia Region
Region
Region 9% of the total number
35.3% of the total of studies, none of which
number of studies, of 11.3% of the total number provided data for 12 month
which 37 studies provided of studies, of which 6 studies infertility estimates
data for 12 month provided data for 12-month
infertility estimates infertility estimates

WHO European Region

Period infertility prevalence
estimate ranges in the
European region were
5.0-34.0%

WHO Region of the Americas

Period infertility prevalence
ranges for the Region of the
Americas were
4.0 - 15.7% WHO South-East
Asia Region
No estimates were
available.
WHO Eastern Mediterranean
Region WHO African Region WHO Western Pacific Region
Period infertility prevalence Overall, period infertility Period infertility prevalence
ranges in the Eastern prevalence estimate ranges were ranges for the Western
Mediterranean region were largest for the African region Pacific region were
5.2 - 15.2% (9.5-32.0%) 1.6 -28.0%

Lifetime infertility prevalence estimate ranges were largest for the Region of the
Americas (4.2-35.3%) European Region (9.0-31.8%) and Western Pacific Region (8.2-39.7%)
and smallest for the African Region (9.3-15.8%)

Executive summary Introduction Methods Results Discussion Annexes 18

3. Results

3.5 Pooled 12-month infertility estimates

We pooled all 12-month infertility prevalence estimates using estimates from 43 studies). Overall pooled lifetime and
meta-analysis and stratified by whether the measure was period prevalence estimates were 17.5% (95% CI: 15.0, 20.3)
estimating lifetime prevalence (n = 39 independent estimates and 12.6% (95% CI: 10.7, 14.6), respectively (Figure 3.5 and
from 37 studies) or period prevalence (n = 52 independent Figure 3.6, respectively).

Figure 3.5. Pooled lifetime infertility Figure 3.6. Pooled period infertility prevalence
prevalence estimates estimates
invlogit(ES) Weight invlogit(ES) Weight
Study with 95% CI (%) Study with 95% CI (%)
African African
Geelhoed (2002) 0.16 [0.14, 0.18] 2.58 Bello (2010) 0.32 [0.28, 0.36] 1.98
Somé (2016) 0.10 [0.08, 0.14] 2.46 Fledderjohann (2016) 0.17 [0.15, 0.19] 2.00
Heterogeneity: τ2 = 0.10, I2 = 86.75%, H2 = 7.55 0.13 [0.09, 0.19] Kouman (2005) 0.10 [0.08, 0.13] 1.94
Test of θi = θj: Q(1) = 7.55, p = 0.01 Polis (2017) 0.31 [0.28, 0.35] 1.99
Sundby (1998) 0.09 [0.09, 0.11] 2.00
Eastern Mediterranean Walraven (2001) 0.10 [0.08, 0.12] 1.98
Ahmadi Asr Badr (2006) 0.03 [0.03, 0.04] 2.55 Heterogeneity: τ2 = 0.49, I2 = 98.55%, H2 = 68.94 0.16 [0.10, 0.26]
Dovom (2014) 0.21 [0.19, 0.24] 2.58 Test of θi = θj: Q(5) = 339.83, p = 0.00
Esmaeilzadeh (2012) 0.16 [0.14, 0.18] 2.59
Heterogeneity: τ2 = 1.16, I2 = 99.33%, H2 = 150.26 0.11 [0.03, 0.29] Eastern Mediterranean
Test of θi = θj: Q(2) = 240.69, p = 0.00 Hassan (1997) 0.12 [0.12, 0.12] 2.02
Mirzaei (2018) 0.05 [0.04, 0.06] 1.98
European Sharif (2020) 0.15 [0.15, 0.16] 2.02
Bhattacharya (2009) 0.19 [0.18, 0.21] 2.62 Heterogeneity: τ2 = 0.38, I2 = 99.83%, H2 = 598.90 0.10 [0.05, 0.18]
Brunetti (1994) 0.10 [0.07, 0.15] 2.27 Test of θi = θj: Q(2) = 250.72, p = 0.00
Buckett (1997) 0.17 [0.15, 0.20] 2.56
Cabrera-Leon (2015) 0.18 [0.18, 0.18] 2.63 European
Datta (2016) 0.13 [0.12, 0.13] 2.62 Ajrouche (2014) 0.18 [0.16, 0.20] 2.00
Gunnell (1994) 0.26 [0.25, 0.28] 2.62 Björvang (2020) 0.10 [0.08, 0.12] 1.96
Gyorffy (2014) 0.10 [0.08, 0.12] 2.55 Eustache (2004) 0.05 [0.03, 0.07] 1.81
Hærvig (2018) 0.18 [0.16, 0.20] 2.61 Gokler (2014) 0.13 [0.10, 0.16] 1.95
Khaer Pedersen (1994) 0.17 [0.12, 0.23] 2.35 Guldbrandsen (2014) 0.16 [0.16, 0.16] 2.02
Klemetti (2010) 0.20 [0.18, 0.22] 2.59 Hallen (2011) 0.07 [0.04, 0.11] 1.70
Kuppers-Chinnow (1997) 0.32 [0.29, 0.34] 2.60 Hoenderboom (2020) 0.17 [0.15, 0.18] 2.01
Oakley (2010) 0.16 [0.15, 0.17] 2.62 Hollegaard (2007) 0.17 [0.16, 0.19] 2.01
Philippov (1998) 0.17 [0.15, 0.18] 2.61 Jensen (2001) 0.10 [0.07, 0.15] 1.82
Rostad (2013) 0.13 [0.12, 0.14] 2.62 Jensen (2001) 0.08 [0.05, 0.12] 1.73
Soares (2011) 0.12 [0.10, 0.14] 2.58 Jensen (2001) 0.10 [0.06, 0.15] 1.77
Sundby (1996) 0.10 [0.09, 0.11] 2.61 Jensen (2001) 0.08 [0.05, 0.12] 1.82
Terävä (2008) 0.16 [0.15, 0.17] 2.62 Kirkegaard (2014) 0.10 [0.09, 0.11] 2.02
Wulff (1997) 0.24 [0.21, 0.28] 2.55 Magnus (2021) 0.12 [0.12, 0.13] 2.02
Heterogeneity: τ2 = 0.15, I2 = 99.04%, H2 = 104.06 0.16 [0.14, 0.19] Muller (2006) 0.11 [0.08, 0.15] 1.85
Test of θi = θj: Q(17) = 757.31, p = 0.00 Muller (2006) 0.09 [0.06, 0.13] 1.82
Muller (2006) 0.05 [0.03, 0.09] 1.57
The Americas Muller (2006) 0.08 [0.05, 0.13] 1.70
Anyalechi (2019) 0.14 [0.12, 0.15] 2.60 Nguyen (2007) 0.12 [0.12, 0.12] 2.02
Cairncross (2020) 0.25 [0.23, 0.26] 2.62 Sarac (2018) 0.09 [0.08, 0.09] 2.01
Crawford (2015) 0.25 [0.22, 0.29] 2.58 Slama (2006) 0.34 [0.18, 0.55] 1.36
Crawford (2015) 0.06 [0.05, 0.07] 2.58 Slama (2012) 0.24 [0.19, 0.30] 1.91
Crawford (2015) 0.10 [0.08, 0.12] 2.56 Taylor (1999) 0.14 [0.12, 0.17] 1.98
Feuntes (1994) 0.26 [0.19, 0.34] 2.34 Toft (2005) 0.15 [0.12, 0.18] 1.95
Gleason (2020) 0.24 [0.22, 0.26] 2.61 Toft (2005) 0.19 [0.15, 0.23] 1.94
Jacobson (2018) 0.35 [0.32, 0.38] 2.60 Toft (2005) 0.27 [0.22, 0.32] 1.94
McQuillan (2003) 0.35 [0.31, 0.39] 2.57 Van der Avoort (2003) 0.15 [0.10, 0.22] 1.78
Nelson (2011) 0.20 [0.16, 0.25] 2.47 Heterogeneity: τ2 = 0.21, I2 = 99.23%, H2 = 129.61 0.12 [0.10, 0.15]
Heterogeneity: τ2 = 0.48, I2 = 98.85%, H2 = 87.04 0.20 [0.14, 0.28] Test of θi = θj: Q(26) = 968.05, p = 0.00
Test of θi = θj: Q(9) = 697.82, p = 0.00
The Americas
Western Pacific Balakrishnan (1993) 0.07 [0.06, 0.07] 2.02
He (2020) 0.21 [0.20, 0.22] 2.63 Bushnik (2012) 0.16 [0.14, 0.17] 2.00
Herbert (2009b) 0.17 [0.15, 0.20] 2.58 Dulberg (1993) 0.09 [0.07, 0.10] 1.98
Kreisel (2020) 0.40 [0.34, 0.45] 2.53 Feuntes (1994) 0.10 [0.09, 0.11] 2.01
Righarts (2015) 0.25 [0.23, 0.28] 2.59 Thoma (2013) 0.16 [0.08, 0.27] 1.52
Righarts (2021) 0.15 [0.14, 0.17] 2.61 Heterogeneity: τ2 = 0.15, I2 = 97.31%, H2 = 37.13 0.10 [0.07, 0.14]
van Roode (2015) 0.26 [0.22, 0.31] 2.53 Test of θi = θj: Q(4) = 174.15, p = 0.00
Heterogeneity: τ2 = 0.20, I2 = 98.08%, H2 = 52.13 0.23 [0.17, 0.30]
Test of θi = θj: Q(5) = 127.67, p = 0.00 Western Pacific
Cai (2011) 0.15 [0.14, 0.17] 2.00
Overall 0.18 [0.15, 0.20] Hu (2020) 0.26 [0.23, 0.29] 1.99
Heterogeneity: τ2 = 0.33, I2 = 99.48%, H2 = 193.20 Huang (2013) 0.13 [0.13, 0.14] 2.02
Test of θi = θj: Q(38) = 2117.16, p = 0.00 Mena (2020) 0.15 [0.14, 0.16] 2.02
Test of group differences: Qb(4) = 7.65, p = 0.11 Meng (2015) 0.14 [0.12, 0.15] 2.00
0.02 0.05 0.12 0.27 0.50 Taylor (1999) 0.20 [0.18, 0.23] 1.99
Random-effects REML model Wang (2018) 0.28 [0.25, 0.31] 1.99
Xingping (2006) 0.02 [0.02, 0.02] 2.02
Yang (2017) 0.11 [0.11, 0.12] 2.02
Zhang (2014) 0.04 [0.04, 0.05] 2.01
Zhou (2018) 0.25 [0.24, 0.26] 2.02
Heterogeneity: τ2 = 0.91, I2 = 99.92%, H2 = 1222.37 0.13 [0.08, 0.21]
Test of θi = θj: Q(10) = 40712.84, p = 0.00
Overall 0.13 [0.11, 0.15]
Heterogeneity: τ2 = 0.40, I2 = 99.81%, H2 = 521.01
Test of θi = θj: Q(51) = 152792.99, p = 0.00
Test of group differences: Qb(4) = 2.89, p = 0.58
0.02 0.05 0.12 0.27 0.50
Random-effects REML model

Executive summary Introduction Methods Results Discussion Annexes 19

Infertility prevalence estimates, 1990–2021

3.5.1 Pooled primary and secondary infertility

For primary 12-month infertility, pooled lifetime and period 12-month infertility, pooled lifetime and period prevalence was
prevalence was 9.6% (95% CI: 6.3, 14.3, n = 12) and 9.0% (95% 6.5% (95% CI: 3.9, 10.7, n = 10) and 4.9% (95%: 2.7, 8.8, n = 17),
CI: 6.6, 12.2, n = 33), respectively (Figure 3.7). For secondary respectively (Figure 3.8).

Figure 3.7. Pooled lifetime and period infertility prevalence estimates for primary infertility
invlogit(ES) Weight
Study with 95% CI (%)
Lifetime
Ahmadi Asr Badr (2006) 0.02 [0.02, 0.03] 2.22
Akhondi (2019) 0.20 [0.20, 0.21] 2.26
Bhattacharya (2009) 0.10 [0.10, 0.11] 2.26
Buckett (1997) 0.11 [0.09, 0.13] 2.23
Cabrera-Leon (2015) 0.06 [0.06, 0.06] 2.26
Esmaeilzadeh (2012) 0.12 [0.10, 0.14] 2.24
Geelhoed (2002) 0.07 [0.05, 0.08] 2.22
Gunnell (1994) 0.16 [0.15, 0.18] 2.25
Kazemijaliseh (2015) 0.17 [0.15, 0.20] 2.25
Philippov (1998) 0.04 [0.03, 0.05] 2.23
Somé (2016) 0.07 [0.05, 0.09] 2.18
Vahidi (2009) 0.25 [0.24, 0.26] 2.26
Heterogeneity: τ2 = 0.63, I2 = 99.90%, H2 = 965.97 0.10 [0.06, 0.14]
Test of θi = θj: Q(11) = 1427.41, p = 0.00

Period
Akre (1999) 0.08 [0.08, 0.08] 2.26
Albayrak (2007) 0.06 [0.05, 0.07] 2.24
Bach (2015) 0.21 [0.19, 0.23] 2.25
Balakrishnan (1993) 0.11 [0.11, 0.12] 2.26
Bushnik (2012) 0.21 [0.18, 0.24] 2.24
Cai (2011) 0.08 [0.06, 0.09] 2.24
Chen (2015) 0.12 [0.11, 0.13] 2.26
Dovom (2014) 0.06 [0.05, 0.08] 2.22
Esmaeilzadeh (2012) 0.04 [0.03, 0.07] 2.11
Feuntes (1994) 0.03 [0.03, 0.04] 2.25
Gokler (2014) 0.05 [0.03, 0.07] 2.17
Hassan (1997) 0.04 [0.04, 0.05] 2.26
Hosseini (2012) 0.03 [0.03, 0.04] 2.23
Joffe (2000) 0.15 [0.13, 0.17] 2.25
Karmaus (1999) 0.23 [0.21, 0.26] 2.25
Keiding (2021) 0.28 [0.26, 0.31] 2.25
Keiding (2021) 0.28 [0.25, 0.31] 2.25
Keiding (2021) 0.38 [0.32, 0.44] 2.22
Keiding (2021) 0.24 [0.19, 0.29] 2.21
Kouman (2005) 0.03 [0.02, 0.05] 2.13
Meng (2015) 0.14 [0.12, 0.16] 2.25
Mirzaei (2018) 0.03 [0.02, 0.03] 2.22
Samarakoon (2007) 0.04 [0.03, 0.05] 2.23
Slama (2012) 0.26 [0.17, 0.38] 2.08
Sundby (1998) 0.03 [0.03, 0.04] 2.24
Thoma (2013) 0.24 [0.12, 0.43] 1.86
Vahidi (2009) 0.03 [0.03, 0.04] 2.25
Wulff (1997) 0.06 [0.04, 0.09] 2.18
Xingping (2006) 0.01 [0.01, 0.01] 2.26
Zargar (1997) 0.15 [0.14, 0.16] 2.26
Zhang (2013) 0.13 [0.12, 0.15] 2.25
Zhang (2014) 0.13 [0.03, 0.03] 2.26
Zhou (2018) 0.15 [0.15, 0.16] 2.26
Heterogeneity: τ2 = 0.97, I2 = 99.94%, H2 = 1635.18 0.09 [0.07, 0.12]
Test of θi = θj: Q(32) = 100588.06, p = 0.00

Overall 0.09 [0.07, 0.12]

Heterogeneity: τ2 = 0.86, I2 = 99.98%, H2 = 4325.08
Test of θi = θj: Q(44) = 191678.85, p = 0.00
Test of group differences: Qb(1) = 0.06, p = 0.81
0.02 0.05 0.12 0.27 0.50
Random-effects REML model

Executive summary Introduction Methods Results Discussion Annexes 20

3. Results

Figure 3.8. Pooled lifetime and period infertility prevalence estimates for secondary infertility
invlogit(ES) Weight
Study with 95% CI (%)

Lifetime
Ahmadi Asr Badr (2006) 0.01 [0.01, 0.02] 3.67
Bhattacharya (2009) 0.07 [0.06, 0.08] 3.74
Buckett (1997) 0.07 [0.05, 0.09] 3.69
Cabrera-Leon (2015) 0.11 [0.11, 0.11] 3.76
Dovom (2014) 0.08 [0.06, 0.10] 3.71
Esmaeilzadeh (2012) 0.02 [0.01, 0.03] 3.62
Geelhoed (2002) 0.11 [0.09, 0.13] 3.73
Gunnell (1994) 0.16 [0.14, 0.17] 3.75
Philippov (1998) 0.13 [0.11, 0.14] 3.74
Somé (2016) 0.04 [0.03, 0.07] 3.61
Heterogeneity: τ2 = 0.73, I2 = 99.43%, H2 = 174.92 0.07 [0.04, 0.11]
Test of θi = θj: Q(9) = 411.13, p = 0.00

Period
Balakrishnan (1993) 0.05 [0.05, 0.06] 3.75
Cai (2011) 0.08 [0.06, 0.09] 3.73
Chauhan (2015) 0.02 [0.01, 0.03] 3.61
Feuntes (1994) 0.04 [0.03, 0.04] 3.74
Gokler (2014) 0.08 [0.06, 0.10] 3.69
Hassan (1997) 0.08 [0.07, 0.08] 3.75
Hosseini (2012) 0.02 [0.01, 0.02] 3.68
Kouman (2005) 0.07 [0.05, 0.09] 3.69
Meng (2015) 0.11 [0.08, 0.16] 3.64
Mirzaei (2018) 0.02 [0.02, 0.03] 3.67
Polis (2017) 0.34 [0.30, 0.39] 3.73
Thoma (2013) 0.05 [0.05, 0.06] 3.74
Wulff (1997) 0.03 [0.02, 0.05] 3.57
Xingping (2006) 0.00 [0.00, 0.00] 3.75
Zhang (2013) 0.35 [0.33, 0.37] 3.75
Zhang (2014) 0.01 [0.01, 0.01] 3.73
Zhou (2018) 0.10 [0.09, 0.10] 3.75
Heterogeneity: τ2 = 1.69, I2 = 99.88%, H2 = 822.78 0.05 [0.03, 0.09]
Test of θi = θj: Q(16) = 45756.15, p = 0.00

Overall 0.05 [0.04, 0.08]

Heterogeneity: τ2 = 1.32, I2 = 99.96%, H2 = 2710.55
Test of θi = θj: Q(26) = 306027.25, p = 0.00
Test of group differences: Qb(1) = 0.51, p = 0.47
0.00 0.02 0.12 0.50
Random-effects REML model

3.5.2 Sensitivity Analyses

For studies that presented more than one 12-month infertility
17.5%
Estimated lifetime
estimate, sensitivity analyses showed minimal variation in
prevalence of infertility
lifetime and period estimates when selecting the minimum (95% confidence interval
value over the maximum value for infertility prevalence [CI]: 15.0, 20.3).
(Table 3.3). Similarly, restricting analyses to only higher-quality
studies with a bias score of 7 or 8 (n = 28 for lifetime, n = 16 for
period), general population studies (n = 39 for lifetime, n = 30
for period), or studies in which the standard errors could be
directly ascertained from the publication (n = 28 for lifetime, 12.6%
n = 39 for period), rather than approximated, also showed little Estimated period
difference in overall infertility prevalence compared with the prevalence of infertility
main findings (Table 3.3). (95% CI: 10.7, 14.6).

Executive summary Introduction Methods Results Discussion Annexes 21

Infertility prevalence estimates, 1990–2021

Table 3.3. Results from sensitivity analyses

Criterion Pooled lifetime infertility, % Pooled period infertility, %

(95% CI) (95% CI)

Overall estimates 17.5 (15.0, 20.3) 12.6 (10.7, 14.6)

Sensitivity analysis criterion applied

Linked studies replaced with minimum value 14.5 (12.3, 17.1) 11.7 (10.0, 13.7)

Limited to high quality studies (bias score >7) 18.1 (15.7, 20.8) 13.9 (10.5, 18.2)

General population studies only 17.5 (15.0, 20.3) 12.4 (9.9, 15.5)

Limited to studies in which standard errors could be 19.0 (16.4, 21.9) 12.2 (10.2, 14.6)
obtained directly

CI = confidence interval

3.5.3 Pooled infertility estimates stratified by region

Regional differences in pooled lifetime infertility prevalence intervals overlapped (Figure 3.6). The highest pooled estimate
showed some variation in magnitude, yet all confidence intervals of period infertility prevalence was in the African Region (16.4%,
overlapped (Figure 3.5). The Western Pacific Region had the 95% CI: 10.0, 25.7, n = 6) followed by the Western Pacific Region
highest prevalence of lifetime infertility (23.2%, 95% CI: 17.4, 30.2, (13.0%, 95% CI: 7.8, 20.8, n = 11), the European Region (12.4%,
n = 6), followed by the Region of the Americas (20.0%, 95% CI: 95% CI: 10.5, 14.6, n = 27), the Region of the Americas (10.4%, 95%
13.9, 27.9, n = 10), the European Region (16.5%, 95% CI: 14.1, 19.2, CI: 7.4, 14.3, n = 5), and the Eastern Mediterranean Region (10.0%,
n = 18), and the African Region (13.1%, 95% CI: 8.6, 19.4, n = 2). 95% CI: 5.2, 18.2, n = 3). The number of studies for lifetime and
The lowest magnitude was found in the Eastern Mediterranean period estimates varied across regions, contributing to variation in
Region (10.7%, 95% CI: 3.4, 29.0, n = 3). Similarly, the magnitude estimates. Notably, no studies conducted in the South-East Asian
of period infertility estimates varied by region, but all confidence Region provided overall 12-month infertility prevalence estimates.

3.5.4 Pooled infertility estimates stratified by income, population,

and sex of respondents
When stratified by country income classifications, pooled Pooled period infertility prevalence was 12.6% (95% CI: 10.8,
lifetime infertility prevalence was 17.8% (95% CI: 15.3, 20.7, 14.7, n = 31) for HIC and 12.6% (95% CI: 9.2, 16.9, n = 21)
n = 30) for HIC and 16.5% (95% CI: 10.4, 25.0, n = 9) for LMIC. for LMIC.

High- Low- and High- Low- and

income middle-income income middle-income
countries countries countries countries

17.8% 16.5% 12.6% 12.6%

Estimated lifetime Estimated period

prevalence of infertility prevalence of infertility

Executive summary Introduction Methods Results Discussion Annexes 22

3. Results

In terms of respondent population (female, male, combined), the 95% CI: 10.5, 14.6, n = 12). This pattern was also observed for
majority of lifetime and period prevalence estimates were based pooled period estimates, but based on only two studies that
on female respondents (n = 37, n = 41, respectively) compared to used male respondents. Pooled period infertility estimates were
male respondents (n = 12, n = 2, respectively) or combined sex 12.6% (95% CI: 10.6, 15.0, n = 41) based on female respondents,
(n = 0, n = 9, respectively). Pooled lifetime infertility prevalence 8.7% (95% CI: 0.51, 14.4, n = 2) based on male respondents, and
estimates were higher when study respondents were female 12.6% (95% CI: 8.2, 18.8, n = 9) based on combined respondents
(17.5%, 95% CI: 14.9, 20.5, n = 37) compared with male (12.4%, (female, male, couple) (data not shown).

3.5.5 Pooled infertility estimates stratified by methodological

approach
The methodological approach varied based on reporting of TTP (21.8%, 95% CI: 13.7, 32.9, n = 3) and current duration
lifetime or period prevalence estimates. Minimal differences approaches (26.2%, 95% CI: 19.9, 33.6, n = 4) followed by a
were found in pooled lifetime infertility estimates across retrospective TTP approach (12.9%, 95% CI: 10.7, 15.6, n = 24)
the three methodological approaches that were used, with (Annex 4). Self-reported and constructed approaches were
estimates of 16.7% (95% CI: 10.3, 26.0, n = 9), 17.6% (95% similar with pooled period infertility prevalence of 10.6%
CI: 15.0, 20.7, n = 27), and 18.5% (95% CI: 15.6, 21.8, n = 3) for (95% CI: 8.1, 13.8, n = 12) and 10.9% (95% CI: 8.0, 14.6, n = 6),
retrospective TTP, self-reported infertility, and undetermined respectively. The lowest infertility prevalence was found for the
approaches, respectively (Annex 4). Corresponding I2 three studies in which the approach could not be determined
percentages were 99.6, 98.6, and 87.1, respectively. In contrast, (6.2%, 95% CI: 1.6, 20.8, n = 3). Corresponding I2 percentages
period estimates were highest when using prospective were 97.5, 68.9, 99.5, 99.0, 98.8, and 99.9, respectively.

3.5.6 Meta-regression results by period and lifetime

Meta-regression results yielded patterns similar to unadjusted Similarly, meta-regression results for period infertility
pooled lifetime infertility prevalence (Table 3.4). Although prevalence were consistent with unadjusted results
confidence intervals overlapped, the magnitude of the odds (Table 3.4). Relative to studies from the European Region,
ratios showed generally higher lifetime infertility prevalence infertility estimates from the African Region were associated
in the Americas (odds ratio [OR]: 1.33, 95% CI: 0.81, 2.18) and with the largest magnitude of association (OR: 1.95, 95%
Western Pacific regions (OR: 1.34, 95% CI: 0.72, 2.49) and lower CI: 1.02, 3.72) followed by the Western Pacific Region (OR: 1.32,
magnitude in the African (OR: 0.60, 95% CI: 0.24, 1.26) and 95% CI: 0.77, 2.27) and Eastern Mediterranean regions (OR:
Eastern Mediterranean (OR: 0.64, 95% CI: 0.31, 1.30) regions 1.11, 95% CI: 0.51, 2.42). Region of the Americas (OR: 0.88, 95%
relative to the European Region after adjustment for definitional CI: 0.41, 1.87) had a lower magnitude of association relative
characteristics and bias scores. This corresponded to pooled to estimates from the European Region, although confidence
adjusted lifetime infertility prevalence estimates of 20.5%, intervals overlapped for all regions with the exception of the
23.4%, 13.0%, and 10.8% for each respective region compared to African Region. These odds ratio associations were consistent
the European Region (16.8%). In the same model, differences by with adjusted period prevalence estimates, which showed the
methodological approach were minimal with adjusted lifetime highest prevalence in the African Region (18.1%) followed by
infertility prevalence estimates of 17.8% for retrospective TTP, the Western Pacific Region (14.2%), European Region (12.6%),
18.5% for undetermined, and 18.2% for the self-reported direct Region of the Americas (11.2%), and Eastern Mediterranean
measures. This corresponded to odds ratio associations of 0.85 Region (10.1%).
(95% CI: 0.49, 1.47) for retrospective TTP and 1.02 (95% CI: 0.49,
2.12) for undetermined relative to self-reported direct measures.

3.6 Certainty of the evidence

The certainty in the overall estimates of 12-month lifetime and as shown in the forest plots. Estimates from individual studies
period prevalence, as well as primary and secondary 12-month were non-overlapping. None of the subgroup hypotheses
prevalence (lifetime and period) is moderate. The certainty was proposed by the review explained the observed heterogeneity,
rated down from high to moderate due to serious inconsistency implying that unknown factors, or factors not reported or
(Annex 5, Tables A–F), based on overlap of point-estimates measured by the individual studies, may be the true underlying
and 95% confidence intervals reported by individual studies explanation for the observed heterogeneity.

Executive summary Introduction Methods Results Discussion Annexes 23

Infertility prevalence estimates, 1990–2021

Table 3.4. Pooled lifetime and period infertility prevalence estimates and multivariable odds
ratios associations by region and methodological approach, adjusting for definitional factors
and risk of bias

Study covariates Infertility prevalence, % (95% CI) Multivariable modela odds ratio (95% CI)

Lifetime prevalence (n = 39 estimates)b

WHO regionc

African Region 13.0 (4.6, 21.3) 0.60 (0.24, 1.26)

Eastern Mediterranean Region 10.8 (5.0, 16.6) 0.64 (0.31, 1.30)

European Region 16.8 (13.4, 20.2) Ref

Region of the Americas 20.5 (15.2, 25.8) 1.33 (0.81, 2.18)

Western Pacific Region 23.4 (15.9, 31.0) 1.34 (0.72, 2.49)

Methodological approach

Prospective TTP - -

Retrospective TTP 17.8 (12.8, 22.9) 0.85 (0.49, 1.47)

Current duration - -

Self-reported direct measure 18.2 (15.2, 21.1) Ref

Constructed measure - -

Undetermined 18.5 (9.3, 27.5) 1.02 (0.49, 2.12)

Period prevalence (n = 52 estimates)b

WHO regionc

African Region 18.1 (11.7, 24.5) 1.95 (1.02, 3.72)

Eastern Mediterranean Region 10.1 (4.4, 15.7) 1.11 (0.51, 2.42)

European Region 12.6 (10.2, 15.0) Ref

Region of the Americas 11.2 (6.2, 16.2) 0.88 (0.41, 1.87)

Western Pacific Region 14.2 (10.1, 18.2) 1.32 (0.77, 2.27)

Methodological approach

Prospective TTP 21.8 (11.1, 32.6) 1.42 (0.53, 3.84)

Retrospective TTP 13.1 (10.5, 15.7) 1.10 (0.65, 1.85)

Current duration 26.0 (14.6, 37.3) 2.43 (1.17, 5.05)

Self-reported direct measure 10.9 (7.8, 14.1) Ref

Constructed measure 11.1 (6.7, 15.5) 1.31 (0.69, 2.47)

Undetermined 6.2 (2.5, 9.9) 0.48 (0.22, 1.07)

CI = confidence interval
TTP = time-to-pregnancy
(-) indicates that 12-month estimates were not found for respective categories.
a
Models were adjusted for region, methodological approach (prospective TTP, retrospective TTP, current duration, self-reported binary measure,
constructed binary measure, undetermined), numerator included intentions, denominator categories (all regardless of risk, ever at risk,
attempting to become pregnant), and risk of bias score (0-8).
b
Lifetime prevalence (I2 = 98.6%), period prevalence (I2 = 99.2%)
c
Overall lifetime and period 12-month estimates were not found for studies conducted in the South-East Asia Region.

Executive summary Introduction Methods Results Discussion Annexes 24

4. Discussion

This section summarizes key findings about infertility prevalence and

reflects on implications for stakeholders in the sexual and reproductive
health field. It highlights how findings in this report should guide
improvements in how research on infertility prevalence is conducted.
In addition, findings in this report provide a basis for raising awareness
about the widespread nature of infertility and about the importance
of ensuring that fertility services are a key element of sexual and
reproductive health and rights in all countries.

Global estimates of infertility are needed to guide planning and time-to-pregnancy design, self-reported infertility measure,
coordination of infertility prevention, diagnosis, and treatment and constructed infertility measure.
efforts. This report is an important milestone in understanding
the contemporary prevalence of infertility. It uses evidence from 4. There is some variation in infertility prevalence across
all eligible studies conducted between 1990 and 2021 identified regions, but data gaps and overlapping confidence
through a comprehensive and systematic analysis of publicly intervals mean that regional differences identified in this
available data. analysis may not be substantial or conclusive. The Western
Pacific Region had the highest prevalence of lifetime infertility
Key findings:3 (23.2%), followed by the Region of the Americas (20.0%) and
the European Region (16.5%). Period infertility was highest
1. Infertility affects a large proportion of the global population, in the African Region (16.4%), followed by the Western Pacific
with approximately one in six people experiencing infertility Region (13.0%) and the European Region (12.4%). The Eastern
in their lifetimes. Lifetime prevalence of infertility is estimated Mediterranean Region had the lowest lifetime and period
to be 17.5% (95% CI: 15.0, 20.3). Period prevalence of infertility infertility, at 10.7% and 10.0%, respectively. The number of
is estimated to be 12.6% (95% CI: 10.7, 14.6). studies for lifetime and period estimates varied across regions,
contributing to uncertainty about findings. Notably, no studies
2. For primary infertility, estimated pooled lifetime and period from the South-East Asia Region provided 12-month infertility
prevalence were 9.6% and 9.0% respectively. For secondary prevalence data that could be used in these analyses.
infertility, estimated pooled lifetime and period prevalence
were 6.5% and 4.9% respectively. 5. Estimates of infertility prevalence are similar across
countries regardless of country income level. Lifetime
3. Methodological approaches to measuring infertility vary infertility prevalence was 17.8% for high-income countries
greatly. Five methodological approaches for measuring and 16.5% for low- and middle-income countries. Period
infertility were identified in this report: prospective time-to- infertility prevalence was 12.6% for high-income countries
pregnancy design, current duration design, retrospective and 12.6% for low- and middle-income countries.

3
Unless otherwise specified, these estimates refer to 12-month period or lifetime prevalence in keeping with the following definition of infertility
adopted by the World Health Organization (WHO): Infertility is a disease of the male or female reproductive system defined by the failure to achieve
a pregnancy after 12 months or more of regular unprotected sexual intercourse.

Executive summary Introduction Methods Results Discussion Annexes 25

Infertility prevalence estimates, 1990–2021

Lifetime and period prevalence of infertility: why both matter

Overall pooled lifetime and period prevalence estimates of Surprisingly, this study found that the range of 12-month
infertility in this report were 17.5% and 12.6%, respectively. infertility prevalence estimates was broad and did not
Period and lifetime measures of infertility prevalence vary substantially by lifetime or period prevalence. This
provide different types of information, both of which are may be due to the majority of studies capturing lifetime
important. Contemporary estimates of period infertility prevalence from reproductive-aged individuals who may not
prevalence help countries identify service needs and have completed childbearing. The wide range of estimates
target resources, whereas estimates of lifetime infertility persisted even after accounting for definitional or study
prevalence provide an understanding of the burden of population characteristics. A prior systematic review (11) also
infertility over people’s lifetime. found considerable heterogeneity in infertility estimates.

Disaggregating primary and secondary infertility

Infertility can be primary or secondary. Primary infertility Distinguishing between primary and secondary infertility is
is when a pregnancy has never been achieved by a important for considering the role of potential etiological
person, and secondary infertility is when at least one prior factors. While primary infertility data may be useful for
pregnancy has been achieved. making comparisons across time and settings (33, 34), high
rates of secondary infertility are associated with infection-
These estimates report that for primary infertility, pooled related pathology resulting from postpartum infections,
lifetime and period prevalence were 9.6% and 9.0% unsafe abortions (35, 36), and some sexually transmitted
respectively. For secondary infertility, pooled lifetime and infections (37). Thus, excluding secondary infertility
period prevalence were 6.5% and 4.9% respectively. from infertility prevalence estimates would result in an
underestimation or misrepresentation of the total burden of
infertility in settings with high prevalence of such factors.

The use of meta-analysis in these estimates facilitated the use of a demographic infertility measure (5-year exposure
pooling of prevalence data, and found a higher magnitude of period) and different regional groupings than those applied in
overall lifetime prevalence of 12-month infertility compared the estimates presented in this report. (These estimates use
to period prevalence, as was expected, and a high level of the country groupings for the six WHO regions.) The range of
heterogeneity across studies, which was anticipated. The infertility estimates for HIC and LMIC were similar in this review,
regional prevalence estimates and data availability reported which is consistent with another (non-systematic) review by
here differ from those reported in a previous study by Boivin et al. (21) that examined 12- and 24-month infertility
Mascarenhas et al. (19), which could be due the latter’s exclusive estimates.

Executive summary Introduction Methods Results Discussion Annexes 26

4. Discussion

4.1 Research gaps and measurement

challenges
These estimates of infertility prevalence should be interpreted in the context of key research gaps and measurement challenges.

4.1.1 Lack of sufficient studies from some regions or studies with

male participants
The estimates presented in this report reflect major gaps in the being lower than period prevalence in the African Region.
availability of studies for certain regions of the world. A larger Furthermore, no studies from the South-East Asia Region
proportion of eligible studies were from the European Region provided overall 12-month infertility prevalence estimates.
(35.3% of included studies), whereas the Eastern Mediterranean These gaps are particularly notable in light of the clear need
Region and South-East Asia Region were the least represented to generate country-level estimates of infertility prevalence
(11.3% and 9.0% of included studies, respectively). The lack to inform national policies and services. Also, there were few
of a sufficient number of studies across regions precluded the studies with male participants, and pooled lifetime and period
generation and comparison of regional differences in primary infertility estimates reported by male respondents were lower
and secondary 12-month infertility prevalence. Strikingly, than for female respondents. Further work will be needed to
the pooled lifetime estimate of infertility for the African provide estimates based on whether infertility is due to male,
Region includes only two studies (38, 39), which may explain female or unexplained factors.
the unexpected and illogical finding of lifetime prevalence

4.1.2 Variation in definitions and in inclusion and exclusion criteria in

studies estimating infertility
Different approaches, definitions and types of study amplify or mask true magnitude and differences in the reported
populations have been used in estimates of infertility estimates. At the same time, such inclusion and exclusion
prevalence, resulting in a high level of heterogeneity. Although criteria may also serve practical program purposes. For
most studies reported either an overall infertility prevalence instance, including receipt of fertility care in the numerator
estimate or overall, primary, and secondary infertility may provide important information related to access to care.
prevalence estimates, this was not consistent across all Restricting the numerator to those intending to conceive
studies. Very few studies applied a consistent definition and (slightly more than 50% of studies in this study) may generate
methodological approach across different regions. Numerous a more useful estimate for predicting service needs (40, 41),
studies used definitions of infertility that do not align with the whereas not restricting the numerator to those with intentions
12-month definition of infertility used by WHO. may be more useful for examining risk factors associated
with infertility (42). Similarly, including individuals outside
Decisions about which populations to include in studies also of reproductive age may have a bearing on how a program
varied among studies. Some studies exclude people based on identifies services for who is at risk of infertility (e.g., non-
relationship status, use of infertility treatment, or pregnancy contracepting, non-sexually active), yet such definitional issues
intentions and reproductive years. These differences may can affect estimates.

4.1.3 Numerous study designs and methods of estimating infertility

Differences in the type of methodological approach used and population has been shown to result in different infertility
the way in which an approach is operationalized make it difficult prevalence estimates (43-45). In addition, there was significant
to ascertain true regional differences in infertility prevalence, variation in the level of detail provided in publications included
and to conduct analyses of trends over time (8, 11). The use of in these estimates, making it difficult in some cases to discern
different numerators or denominators within the same study definitional characteristics.

Executive summary Introduction Methods Results Discussion Annexes 27

Infertility prevalence estimates, 1990–2021

4.1.4 Certainty of estimates and other limitations

The certainty in the estimates of 12-month period and lifetime rated as low or moderate risk of bias, with only 1.5% of studies
prevalence, as well as primary and secondary 12-month rated as high risk, and excluding high-risk studies had minimal
prevalence (lifetime and period) was rated moderate due to impact on the overall estimates. Funnel plots were also found to
inconsistency. None of the subgroup hypotheses explained the be symmetrical, suggesting low potential for publication bias,
observed heterogeneity. In addition, more than 50% of studies although caution should generally be exercised in interpreting
in these estimates either reported a response rate below 75% funnel plots for this purpose (46).
or failed to report a response rate. However, most studies were

4.2 Implications for research

The findings in these estimates have implications for researching The majority of studies that met criteria for inclusion in the
prevalence of infertility, including selecting methodological analyses reported here provided estimates of the prevalence of
approaches, reporting estimates and making comparisons 12-month infertility, suggesting that population-level estimates
across studies. There is a clear need to use consistent, systematic can be generated globally using the WHO definition of infertility.
and comprehensive processes to collect and report infertility Flexibility may be required to accommodate different needs
prevalence data at the global, regional, and country levels. The relating to population-level estimates; however, at a minimum,
moderate certainty of these estimates indicates a need to improve studies should collect data that measure infertility prevalence
how research on infertility is conducted so that inferences drawn based on the WHO definition.
from estimates can be presented with greater confidence.

Recommendations for future infertility prevalence research

1. Estimating prevalence of infertility Assumptions regarding whether participants were at risk of

pregnancy (e.g., married couples considered a proxy for risk
The field needs a standard set of questions for ascertaining
status), should be outlined. Both an estimate of infertility
infertility prevalence that could be adopted by Demographic
prevalence and corresponding standard error (or confidence
and Health Surveys and other standard population-based
interval) should be provided. This is particularly important
surveys. These questions should be qualitatively examined
for studies using complex survey designs or survival analysis
to ensure comparability in interpretation and relevance
in which the standard error cannot be calculated from the
across different contexts. Questions should be flexible
sample size and prevalence estimate. Also, when feasible,
enough to allow for different definitions and approaches
estimates for total, primary, and secondary infertility
in order to facilitate comparison. At a minimum, 12-month
prevalence should be reported, with stratification by
period infertility should be measured in accordance with
age and sex.
the WHO definition. Other dimensions such as intentions
and receipt of fertility care should be captured. Multiple
4. Making comparisons across studies
approaches could also be incorporated to allow for
comparisons across methodologies. Measures should allow Estimates should only be compared when they are as similar
for disaggregation by lifetime and period infertility, primary as possible in relation to various study characteristics such
and secondary infertility, and male and female respondents. as definitions, methodological approaches, and exclusion
criteria. Thus, estimates should be compared when
2. Selecting methodological approach stratified and matched according to key parameters, such as
methodological approach, definition, primary or secondary
Researchers should consider research objectives, data
infertility, intention or risk, sex, and age. When differences
sources, resources, and validity and reliability when
exist, the reasons for the differences should be examined.
selecting an approach to estimate infertility prevalence (i.e.,
For studies in which a standard approach can be applied
prospective time-to-pregnancy design, retrospective time-
to make comparisons across different geographic settings,
to-pregnancy design, current duration design, self-reported
the interpretation of estimates would be aided by the
infertility measure, and constructed infertility measure).
reporting of additional information on contraceptive use and
method mix (including fertility awareness-based methods),
3. Reporting estimates
contraceptive failures, fertility intentions, sexual behavior
Researchers should provide detailed methodological and (frequency, timing, abstinence), parity and gravidity, timing
analytical information when reporting estimates of infertility of pregnancy awareness, postpartum breastfeeding duration
prevalence. It is especially important to specify the survey and lactational amenorrhea, infertility treatment-seeking,
question or questions used to generate the estimates as and availability and use of infertility treatment.
well as clearly defining the numerator and denominator.

Executive summary Introduction Methods Results Discussion Annexes 28

4. Discussion

4.3 Policy and programmatic implications

Valid and reliable prevalence estimates are needed to Currently, there are policy and programmatic challenges
understand the burden of infertility and to inform policies, related to the low availability, accessibility, and quality
advocacy, service provision, and monitoring of fertility care. of interventions to address infertility in most countries.
Having the right data is essential for generating people-centred Prevention, diagnosis and treatment of infertility is often not
and evidence-based policies and services to mitigate the impact prioritized in national population and development policies,
of infertility, and would enable fulfillment of the rights of reproductive health strategies, or health financing. These
individuals to found a family and decide the number, timing and estimates improve our understanding of the prevalence and
spacing of their children. These infertility prevalence estimates burden of infertility, and they also provide a basis for the
clearly show that a large number of people in different regions formulation of policies and services to advance universal
of the world experience infertility and require infertility access to fertility care.
prevention, diagnosis and treatment.

More needs to be done to:

Estimate infertility prevalence by country

Disaggregate infertility estimates by cause (male factor, female factor, both male
and female factors, and unexplained factors) and by age

Develop a set of questions that can be used in nationally representative

demographic health and population surveys to generate infertility prevalence

Promote and ensure consistency in definitions and measures used in

infertility research

Enhance inclusion of infertility in health policies, services, and financing, and

achieve universal access to fertility care for all

4.4 Conclusion
Human health and gender equality are central elements of the presented in this report show high prevalence of infertility
Sustainable Development Goals, which call on governments to globally and regionally, and can be used to support the
ensure universal access to sexual and reproductive health and development of policies and practices that will help more
rights. Fertility care is a core part of sexual and reproductive individuals and couples achieve their desired family size.
health, and responding to infertility can mitigate gender Findings also provide insight into how the estimation of
inequality. The drive to achieve the Sustainable Development infertility prevalence can be improved in order to obtain more
Goals therefore must encompass actions to respond more actionable data, including data that allow for more meaningful
effectively to the needs of people with infertility. The estimates comparisons across settings and time.

Executive summary Introduction Methods Results Discussion Annexes 29

Infertility prevalence estimates, 1990–2021

References
1. Thoma M, Fledderjohann J, Cox C, Kantum Adageba R. 13. Thoma M. Measuring infertility: searching for consensus.
Biological and Social Aspects of Human Infertility: A Global J Womens Health (Larchmt). 2015;24(7):541-3.(https://doi.
Perspective. 2021. In: Oxford Encyclopedia of Sexual and org/10.1089%2Fjwh.2015.5399, accessed on March 30, 2023).
Reproductive Health Oxford University Press, Oxford
[Internet]. Oxford University Press.(https://doi.org/10.1093/ 14. International Classification of Diseases, 11th Revision (ICD-11)
acrefore/9780190632366.013.184, accessed on March 30, 2023). Geneva: World Health Organization; 2018. (https://www.who.
int/classifications/icd/en/, accessed on July 18, 2022).
2. Wang Y, Fu Y, Ghazi P, Gao Q, Tian T, Kong F, et al. Prevalence
of intimate partner violence against infertile women in low- 15. Zegers-Hochschild F, Adamson GD, Dyer S, Racowsky C,
income and middle-income countries: a systematic review De Mouzon J, Sokol R, et al. The international glossary
and meta-analysis. Lancet Glob Health. 2022;10(6):e820- on infertility and fertility care, 2017. Hum Reprod.
e30.(https://doi.org/10.1016/S2214-109X(22)00098-5, accessed 2017;32(9):1786-801.(https://doi.org/10.1093/humrep/dex234,
on March 30, 2023). accessed on March 30, 2023).

3. International Covenant on Economic, Social Cultural Rights 16. Joffe M, Key J, Best N, Keiding N, Scheike T, Jensen TK.
(CESCR). New York: United Nations General Assembly; 1966. Studying time to pregnancy by use of a retrospective design.
Am J Epidemiol. 2005;162(2):115-24.(https://doi.org/10.1093/
4. Convention on the Elimination of All Forms of Discrimination aje/kwi172, accessed on March 30, 2023).
Against Women, Treaty Series, vol. 1249, Part IV; Article 16.
New York: United Nations General Assembly; 1979. 17. Bonde JP, Joffe M, Sallmén M, Kristensen P, Olsen J, Roeleveld
N, et al. Validity issues relating to time-to-pregnancy
5. The Universal Declaration of Human Rights (UDHR). New York: studies of fertility. Epidemiol. 2006;17(4):347-9.(https://
United Nations. General Assembly; 1948.(https://www.un.org/ doi.org/10.1097/01.ede.0000210239.80406.46, accessed on
sites/un2.un.org/files/2021/03/udhr.pdf, accessed on March March 30, 2023).
30, 2023).
18. Dyer SJ. International estimates on infertility prevalence
6. Starrs AM, Ezeh AC, Barker G, Basu A, Bertrand JT, Blum R, et and treatment seeking: potential need and demand for
al. Accelerate progress-sexual and reproductive health and medical care. Hum Reprod. 2009;24(9):2379-80; author reply
rights for all: report of the Guttmacher-Lancet Commission. 80-3.(https://doi.org/10.1093/humrep/dep219, accessed on
Lancet. 2018;391(10140):2642-92.(https://doi.org/10.1016/ March 30, 2023).
S0140-6736(18)30293-9, accessed on March 30, 2023).
19. Mascarenhas MN, Flaxman SR, Boerma T, Vanderpoel S,
7. Thonneau P, Spira A. Prevalence of infertility: international Stevens GA. National, regional, and global trends in infertility
data and problems of measurement. Eur J Obstet Gynecol prevalence since 1990: a systematic analysis of 277 health
Reprod Biol. 1991;38(1):43-52.(https://doi.org/10.1016/0028- surveys. PLoS Med. 2012;9(12):e1001356.(https://doi.org/10.
2243(91)90206-z, accessed on March 30, 2023). 1371/journal.pmed.1001356, accessed on March 30, 2023).

8. Schmidt L, Münster K. Infertility, involuntary infecundity, 20. Rutstein SO, Shah IH. Infecundity, Infertility, and Childlessness
and the seeking of medical advice in industrialized countries in Developing Countries. DHS Comparative Reports No. 9.
1970-1992: a review of concepts, measurements and results. Calverton, Maryland, USA: ORC Macro and the World Health
Hum Reprod. 1995;10(6):1407-18.(https://doi.org/10.1093/ Organization; 2004.(https://dhsprogram.com/pubs/pdf/cr9/
humrep/10.6.1407, accessed on March 30, 2023). cr9.pdf, accessed on March 30, 2023).

9. Guzick DS, Swan S. The decline of infertility: apparent or 21. Boivin J, Bunting L, Collins JA, Nygren KG. International
real? Fertil Steril. 2006;86(3):524-6.(https://doi.org/10.1016/j. estimates of infertility prevalence and treatment-seeking:
fertnstert.2006.05.027, accessed on March 30, 2023). potential need and demand for infertility medical care. Hum
Reprod. 2007;22(6):1506-12.(https://doi.org/10.1093/humrep/
10. Olive DL, Pritts EA. Estimating infertility: the devil is dem046, accessed on March 30, 2023).
in the details. Fertil Steril. 2006;86(3):529-30.(https://
doi.org/10.1016/j.fertnstert.2006.05.026, accessed on 22. Direkvand MA, Delpisheh A, Sayehmiri K. An investigation of
March 30, 2023). the worldwide prevalence of infertility as a systematic review.
Qom Univ Med Sci J. 2016;10(1):76-87.(https://journal.muq.
11. Gurunath S, Pandian Z, Anderson RA, Bhattacharya S. Defining ac.ir/article-1-443-en.html, accessed on March 30, 2023).
infertility — a systematic review of prevalence studies. Hum
Reprod Update. 2011;17(5):575-88.(https://doi.org/10.1093/ 23. Stevens GA, Alkema L, Black RE, Boerma JT, Collins GS,
humupd/dmr015, accessed on March 30, 2023). Ezzati M, et al. Guidelines for Accurate and Transparent Health
Estimates Reporting: the GATHER statement. Lancet (London,
12. Stanford JB. What is the true prevalence of infertility? Fertil England). 2016;388(10062):e19-e23.(https://doi.org/10.1016/
Steril. 2013;99(5):1201-2. s0140-6736(16)30388-9, accessed on March 30, 2023).

Executive summary Introduction Methods Results Discussion Annexes 30

References

24. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, 36. Sharma S, Mittal S, Aggarwal P. Management of infertility in
Mulrow CD, et al. The PRISMA 2020 statement: an updated low resource countries. BJOG. 2009;116 Suppl 1:77-83.(https://
guideline for reporting systematic reviews. BMJ (Clinical doi.org/10.1111/j.1471-0528.2009.02311.x, accessed on
research ed). 2021;372:n71.(https://doi.org/10.1136/bmj.n71, March 30, 2023).
accessed on March 30, 2023).
37. Chemaitelly H, Majed A, Abu-Hijleh F, Blondeel K,
25. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Matsaseng TC, Kiarie J, et al. Global epidemiology of
Rennie D, et al. Meta-analysis of observational studies in Neisseria gonorrhoeae in infertile populations: systematic
epidemiology: a proposal for reporting. Meta-analysis Of review, meta-analysis and metaregression. Sex Transm
Observational Studies in Epidemiology (MOOSE) group. JAMA. Infect. 2021;97(2):157-69.(https://doi.org/10.1136/
2000;283(15):2008-12. sextrans-2020-054515, accessed on March 30, 2023).

26. Cox C, Thoma M, Tchangalova N, Mburu G, Kiarie J. Infertility 38. Geelhoed DW, Nayembil D, Asare K, Schagen van Leeuwen
estimates and their methods of estimation from 1990 - 2020: JH, van Roosmalen J. Infertility in rural Ghana. Int J Gynaecol
a systematic review and meta-analysis protocol. 2020. Obstet. 2002;79(2):137-42.(https://doi.org/10.1016/s0020-
(https://www.crd.york.ac.uk/prospero/display_record. 7292(02)00237-0, accessed on July 18, 2022).
php?RecordID=211704; accessed on July 18, 2022).
39. Eric SN, Justine B, Jean NP. Prevalence of the Infertility among
27. World Bank. Countries and economies [Internet]. Washington couples in ouagadougou (Burkina Faso): a population-based
DC: World Bank; 2021. (https://data.worldbank.org/country, survey. The Open Public Health J. 2016;9(1).(http://dx.doi.org/
accessed on July 18, 2022). 10.2174/1874944501609010088, accessed on March 30, 2023).

28. Hoy D, Brooks P, Woolf A, Blyth F, March L, Bain C, et al. 40. White L, McQuillan J, Greil AL. Explaining disparities
Assessing risk of bias in prevalence studies: modification in treatment seeking: the case of infertility. Fertil
of an existing tool and evidence of interrater agreement. J Steril. 2006;85(4):853-7.(https://doi.org/10.1016/j.
Clin Epidemiol. 2012;65(9):934-9.(https://doi.org/10.1016/j. fertnstert.2005.11.039, accessed on March 30, 2023).
jclinepi.2011.11.014, accessed on March 30, 2023).
41. Greil AL, Slauson-Blevins KS, Tiemeyer S, McQuillan J,
29. Rücker G, Schwarzer G, Carpenter JR, Schumacher M. Undue Shreffler KM. A new way to estimate the potential unmet need
reliance on I(2) in assessing heterogeneity may mislead. BMC for infertility services among women in the United States.
Med Res Methodol. 2008;8:79.(https://doi.org/10.1186/1471- J Womens Health (Larchmt). 2016;25(2):133-8.(https://doi.
2288-8-79, accessed on March 30, 2023). org/10.1089/jwh.2015.5390, accessed on March 30, 2023).

30. StataCorp. Stata Statistical Software: Release 16. College 42. Slama R, Ballester F, Casas M, Cordier S, Eggesbø M, Iniguez
Station, TX: StataCorp LLC; 2019. C, et al. Epidemiologic tools to study the influence of
environmental factors on fecundity and pregnancy-related
31. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso- outcomes. Epidemiol Rev. 2014;36(1):148-64.(https://doi.
Coello P, et al. GRADE: an emerging consensus on rating org/10.1093/epirev/mxt011, accessed on March 30, 2023).
quality of evidence and strength of recommendations.
BMJ. 2008;336(7650):924-6.(https://doi.org/10.1136/ 43. Larsen U. Research on infertility: which definition should we
bmj.39489.470347.ad, accessed on March 30, 2023). use? Fertil Steril. 2005;83(4):846-52.(https://doi.org/10.1016/j.
fertnstert.2004.11.033, accessed on July 18, 2022).
32. Iorio A, Spencer FA, Falavigna M, Alba C, Lang E, Burnand B, et
al. Use of GRADE for assessment of evidence about prognosis: 44. Crawford S, Fussman C, Bailey M, Bernson D, Jamieson DJ,
rating confidence in estimates of event rates in broad categories Murray-Jordan M, et al. Estimates of lifetime infertility from
of patients. BMJ (Clinical research ed). 2015;350:h870.(https:// three states: the behavioral risk factor surveillance system.
doi.org/10.1136/bmj.h870, accessed on March 30, 2023). J Womens Health (Larchmt). 2015;24(7):578-86.(https://doi.
org/10.1089%2Fjwh.2014.5102, accessed on March 30, 2023).
33. Basso O, Juul S, Olsen J. Time to pregnancy as a
correlate of fecundity: differential persistence in trying 45. Jacobson MH, Chin HB, Mertens AC, Spencer JB,
to become pregnant as a source of bias. Int J Epidemiol. Fothergill A, Howards PP. “Research on Infertility:
2000;29(5):856-61.(https://doi.org/10.1093/ije/29.5.856, Definition Makes a Difference” Revisited. Am J Epidemiol.
accessed on March 30, 2023). 2018;187(2):337-46.(https://doi.org/10.1093/aje/kwx240,
accessed on March 30, 2023).
34. Olsen J, Rachootin P. Invited commentary: monitoring
fecundity over time--if we do it, then let’s do it right. Am J 46. Sterne JA, Egger M. Funnel plots for detecting bias in meta-
Epidemiol. 2003;157(2):94-7.(https://doi.org/10.1093/aje/ analysis: guidelines on choice of axis. J Clin Epidemiol.
kwf178, accessed on July 18, 2022) 2001;54(10):1046-55. (https://doi.org/10.1016/s0895-
4356(01)00377-8, accessed on March 30, 2023).
35. Larsen U. Primary and secondary infertility in sub-Saharan
Africa. Int J Epidemiol. 2000;29(2):285-91.(https://doi.
org/10.1093/ije/29.2.285, accessed on March 30, 2023).

Executive summary Introduction Methods Results Discussion Annexes 31

Annexes

Executive summary Introduction Methods Results Discussion Annexes 32

Annex 1. Risk of
bias assessment

The Risk of Bias Tool is designed to assess the risk of bias modified by Hoy et al. (2012). We slightly modified the Hoy
in population-based prevalence studies. It was initially et al. (2012) tool to align with the purpose of our study as
developed by Leboeuf-Yde and Lauritsen (1995) and then shown below.

ITEM RISK OF BIAS CRITERIA FOR ANSWERS ADDITIONAL NOTES AND EXAMPLES
# ITEM

External validity

1 Was the • Yes (LOW RISK): The The sampling frame is a list of the sampling units in the target
sampling sampling frame was a true population and the study sample is drawn from this list. Clinic-
frame a true or close representation of based sampling frames are considered high risk given that they
or close the target population. only represent those individuals seeking care.
representation
of the target • No (HIGH RISK): The Examples:
population? sampling frame was NOT a • The sampling frame was a list of almost every individual within
true or close representation the target population. The answer is: Yes (LOW RISK).
of the target population.
• The cluster sampling method was used and the sample of
clusters/villages was drawn from a list of all villages in the target
population. The answer is: Yes (LOW RISK).
• The sampling frame was a list of just one particular ethnic group
within the overall target population, which comprised many
groups. The answer is: No (HIGH RISK).
• The sampling frame included all eligible patients attending
a primary care clinic that serves the target population over a
12-month period. The answer is No (HIGH RISK).
• The sampling frame included pregnant women only. The answer
is: No (HIGH RISK).

Executive summary Introduction Methods Results Discussion Annexes 33

Infertility prevalence estimates, 1990–2021

ITEM RISK OF BIAS CRITERIA FOR ANSWERS ADDITIONAL NOTES AND EXAMPLES
# ITEM

2 Was some form • Yes (LOW RISK): A census A census collects information from every unit in the sampling frame.
of random was undertaken, OR, some Clinic-based studies that recruit all eligible patients within a time
selection used form of random selection period of 12 months or longer are considered a census. In a survey,
to select the was used to select the only part of the sampling frame is sampled. In these instances,
sample, OR, sample (e.g., simple random random selection of the sample helps minimize study bias.
was a census sampling, stratified random
undertaken? sampling, cluster sampling, Examples:
systematic sampling). • The sample was selected using simple random sampling. The
answer is: Yes (LOW RISK).
• No (HIGH RISK): A census
was NOT undertaken, AND • The target population was the village and every person in the
some form of random village was sampled. The answer is: Yes (LOW RISK).
selection was NOT used to
select the sample. • A census of the patient population was taken at a clinic by
sampling all eligible patients over a 12-month period. The answer
is: Yes (LOW RISK).
• The target population was a region within a country but only the
nearest villages to the capital city were selected in order to save
on the cost of fuel. The answer is: No (HIGH RISK).
• In a case-control study, controls were selected to match the cases on
certain characteristics such as age. The answer is: No: (HIGH RISK)

3 Was the • Yes (LOW RISK): The Examples:

likelihood of response rate for the • The response rate was 83%. The answer is : Yes (LOW RISK).
non-response study was >/=75%, OR, an
bias minimal? analysis was performed • The response rate was 68%; however, the researchers did an
that showed no significant analysis and found no significant difference between responders
difference in relevant and non-responders in terms of age, sex, occupation and or
demographic characteristics socioeconomic status. The answer is: Yes (LOW RISK).
between responders and • The response rate was 68%; however, the researchers applied
no responders OR authors weighting methods to account for differences between
applied weighting methods responders and non-responders in terms of age, sex, occupation
to account for differences and or socioeconomic status. The answer is: Yes (LOW RISK).
between responders and
non-responders. • The response rate was 65% and the researchers did NOT carry
out an analysis to compare relevant demographic characteristics
• No (HIGH RISK): The between responders and non-responders. The answer is: No
response rate was <75%, (HIGH RISK).
and if any analysis
comparing responders and • The response rate was 69% and the researchers did an analysis
non-responders was done, and found a significant difference in age, sex and socio-economic
it showed a significant status between responders and non-responders and no
difference in relevant procedures were applied to account for differences. The answer
demographic characteristics is: No (HIGH RISK).
between responders and
non-responders.

Internal validity

4 Were data • Yes (LOW RISK): All data A proxy is a representative of the subject.
collected were collected directly from
directly from the subjects. Examples:
the subjects • All eligible subjects in the household were interviewed separately.
(as opposed to • No (HIGH RISK): In some The answer is: Yes (LOW RISK).
a proxy)? instances, data were
collected from a proxy. • A representative of the household was interviewed and
questioned about the presence of infertility in at least one
household member, including his or her partner. The answer is:
No (HIGH RISK).
• Medical records (proxy) are used to identify those with infertility.
Some individuals with known infertility will not be captured in
medical records since not all patients are asked about infertility.
The answer is No (HIGH RISK).

Executive summary Introduction Methods Results Discussion Annexes 34

Annex 1. Risk of bias assessment

ITEM RISK OF BIAS CRITERIA FOR ANSWERS ADDITIONAL NOTES AND EXAMPLES
# ITEM

5 Was an • Yes (LOW RISK): An Acceptable case definitions of infertility include those that define
acceptable acceptable case definition infertility as a failure to achieve either a clinical pregnancy or live
case definition was used. In instances birth after > 12 months of regular unprotected intercourse overall or
used in the where multiple definitions > 6 months for ages 35 and over, consistent with minimum clinical
study? are used, at least one criterion for defining infertility (ACOG). Given that some studies
acceptable case definition avoid use of 12 months due to heaping, definitions >9 months will
was used. be considered acceptable. Intentions may or may not be included
in the definition. Definitions that include menopausal women and
• No (HIGH RISK): An surgically sterile men or women are not acceptable.
acceptable case definition
was NOT used. Examples:
• Clinical definition: failure to achieve a clinical pregnancy after 12
months or more of regular unprotected sexual intercourse. The
answer is: Yes (LOW RISK)
• Epidemiological definition: failure to achieve a clinical pregnancy
(or live birth) after 24 months or more of regular unprotected
sexual intercourse. The answer is: Yes (LOW RISK)
• Demographic definition: failure to achieve a live birth after 5 years
or more of regular unprotected sexual intercourse. The answer is:
Yes (LOW RISK)
• Failure to achieve a clinical pregnancy after 6 months of regular
unprotected sexual intercourse. The answer is: No (HIGH RISK).
• Difficulty achieving a clinical pregnancy (no duration specified).
The answer is: No (HIGH RISK)

6 Was the study • Yes (LOW RISK): The study Self-reported time-to-pregnancy/ time-trying-to-conceive
instrument/ instruments/items had been instruments (retrospective or prospective) are considered low risk
items that shown to have reliability given that several studies have shown these measures to be fairly
measured the and validity (if this was reliable and valid (though not all). Binary self-reported instruments
parameter of necessary), e.g., test-retest, that include a duration are also considered low risk (i.e., ‘Have you
interest (e.g., piloting, validation in a ever had a time, lasting 12 months or longer, when you and a partner
prevalence previous study, etc. were trying for a pregnancy but it didn’t happen?’). Studies that use
of infertility) ICD codes to identify cases of infertility are acceptable.
shown to have • No (HIGH RISK): The study
reliability and instruments/items had Instruments that do not specify a duration of time are considered
validity (if NOT been shown to have high risk unless compared to a valid and reliable measure since the
necessary)? reliability or validity (if this definition of infertility is duration-based and these types of measures
was necessary). have not been validated. Current duration measures are considered
high risk since these measures have not been validated for measuring
infertility. Studies that use the reproductive calendar to indirectly
classify women as infertile are considered high risk. Studies that use
proxy measures for unprotected sex are considered high risk (e.g.,
assume those that are married are having unprotected sex).

Examples:
• The authors used a reproductive calendar to determine time to
pregnancy or live birth. The answer is: Yes (LOW RISK).
• The authors ask participants if they ever tried for 12-months or
longer to get pregnant. The answer is: Yes (LOW RISK).
• The authors ask participants if they ever had difficulty conceiving
and do not compare results to a valid and reliable measure. The
answer is: No (HIGH RISK)
• The authors use a reproductive calendar to estimate current
duration at risk and do not compare to a valid and reliable
measure. The answer is: No (HIGH RISK)
• The authors don’t explicitly state or sufficiently describe the
question (or the ICD codes) used to determine infertility status of
respondents. The answer is: No (HIGH RISK)
• A reproductive calendar is used to indirectly determine whether a
woman is classified as infertile (no direct questions on infertility).
The answer is: No (HIGH RISK).
• The authors assume that all married couples are having
unprotected sex. The answer is: No (HIGH RISK).

Executive summary Introduction Methods Results Discussion Annexes 35

Infertility prevalence estimates, 1990–2021

ITEM RISK OF BIAS CRITERIA FOR ANSWERS ADDITIONAL NOTES AND EXAMPLES
# ITEM

7 Was the same • Yes (LOW RISK): The same The mode of data collection is the method used for collecting
mode of data mode of data collection was information from the subjects. The most common modes are face-
collection used for all subjects. to-face interviews, telephone interviews and self-administered
used for all questionnaires.
subjects? • No (HIGH RISK): The same
mode of data collection was Examples:
NOT used for all subjects. • All eligible subjects had a face-to-face interview. The answer is: Yes
(LOW RISK).
• Some subjects were interviewed over the telephone and some
filled in postal questionnaires. The answer is: No (HIGH RISK).

8 Were the • Yes (LOW RISK): The paper There may be errors in the calculation and/or reporting of the
numerator(s) presented appropriate numerator and/or denominator.
and numerator(s) AND
denominator(s) denominator(s) for the Examples:
for the parameter of interest (e.g., • There were no errors in the reporting of the numerator(s) AND
parameter the prevalence of infertility). denominator(s) for the prevalence of infertility. The answer is: Yes
of interest (LOW RISK).
appropriate? • No (HIGH RISK): The paper
did present numerator(s) • In reporting the overall prevalence of infertility in both men and
AND denominator(s) for the women), the authors accidentally used the population of women
parameter of interest but as the denominator rather than the combined population. The
one or more of these were answer is: No (HIGH RISK).
inappropriate.

Summary assessment

9 Summary item • LOW RISK OF BIAS: Further 1-point is awarded to each item labeled as “yes (LOW RISK). Items
on the overall research is very unlikely to 1 - 8 are summed and level of risk is determined by the following
risk of study change our confidence in tertiles:
bias the estimate.
• Low risk of bias: 6 - 8
• MODERATE RISK OF BIAS:
Further research is likely to • Moderate risk of bias: 3 - 5
have an important impact
on our confidence in the • High risk of bias: 0 - 2
estimate and may change
the estimate.
• HIGH RISK OF BIAS: Further
research is very likely to
have an important impact
on our confidence in the
estimate and is likely to
change the estimate.

Executive summary Introduction Methods Results Discussion Annexes 36

Annex 2. Studies
included in the
systematic review

Ahmadi Asr Badr Y, Madaen K, Haj Ebrahimi S, Ehsan Nejad AH, Balakrishnan TR, Fernando R. Infertility among Canadians:
Koushavar H. Prevalence of infertility in Tabriz in 2004. Urol J an analysis of data from the Canadian Fertility Survey (1984)
2006;3:87–91. and General Social Survey (1990). The Prevalence of Infertility
in Canada: Research Studies of the Royal Commission on New
Ajrouche R, Rudant J, Orsi L, Petit A, Baruchel A, Nelken B, Reproductive Technologies 1993;6:, p. 107–162. Minister of
Pasquet M, Michel G, Bergeron C, Ducassou S, et al. Maternal Supply and Services Canada Ottawa: Ottawa, Canada.
reproductive history, fertility treatments and folic acid
supplementation in the risk of childhood acute leukemia: the Barden-O’Fallon J. Associates of self-reported fertility status and
ESTELLE Study. Cancer Causes Control 2014;25:1283–1293. infertility treatment-seeking in a rural district of Malawi. Hum
Reprod 2005;20:2229–2236.
Akhondi MM, Ranjbar F, Shirzad M, Ardakani ZB, Kamali K,
Mohammad K. Practical difficulties in estimating the prevalence Bello B, Kielkowski D, Heederik D, Wilson K. Time-to-pregnancy
of primary infertility in Iran. Int J Fertil Steril 2019;13:113–117. and pregnancy outcomes in a South African population. BMC
Public Health 2010;10:565–572.
Akre O, Cnattingius S, Bergström R, Kvist U, Trichopoulos D,
Ekbom A. Human fertility does not decline: evidence from Bernhard P, Makunde RW, Magnussen P, Lemnge MM. Genital
Sweden. Fertil Steril 1999;71:1066–1069. manifestations and reproductive health in female residents of a
wuchereria bancrofti-endemic area in Tanzania. Trans R Soc Trop
Albayrak E, Günay O. State and trait anxiety levels of childless Med Hyg 2000;94:409–412.
women in Kayseri, Turkey. Eur J Contracept Reprod Health Care
2007;12:385–390. Bhattacharya S, Porter M, Amalraj E, Templeton A, Hamilton M,
Lee AJ, Kurinczuk JJ. The epidemiology of infertility in the North
Anyalechi GE, Hong J, Kreisel K, Torrone E, Boulet S, Gorwitz R, East of Scotland. Hum Reprod 2009;24:3096–3107.
Kirkcaldy RD, Bernstein K. Self-reported infertility and associated
pelvic inflammatory disease among women of reproductive Björvang RD, Gennings C, Lin P-I, Hussein G, Kiviranta H,
age: National Health and Nutrition Examination Survey, United Rantakokko P, Ruokojärvi P, Lindh CH, Damdimopoulou P,
States, 2013-2016. Sex Transm Dis 2019;46:446–451. Bornehag C-G. Persistent organic pollutants, pre-pregnancy use
of combined oral contraceptives, age, and time-to-pregnancy in
Bach CC, Bech BH, Nohr EA, Olsen J, Matthiesen NB, Bossi R, the SELMA cohort. Environ Health 2020;19:1–14.
Uldbjerg N, Bonefeld-Jørgensen EC, Henriksen TB. Serum
perfluoroalkyl acids and time to pregnancy in nulliparous Bolumar F, Olsen J, Boldsen J, European Study Group on
women. Environ Res 2015;142:535–541. Infertility Subfecundity. Smoking reduces fecundity: a European
multicenter study on infertility and subfecundity. Am J
Epidemiol 1996;143:578–587.

Executive summary Introduction Methods Results Discussion Annexes 37

Infertility prevalence estimates, 1990–2021

Boulet SL, Warner L, Adamski A, Smith RA, Burley K, Dovom MR, Tehrani FR, Abedini M, Amirshekari G, Hashemi S,
Grigorescu V. Behavioral risk factor surveillance system state- Noroozzadeh M. A population-based study on infertility and its
added questions: leveraging an existing surveillance system to influencing factors in four selected provinces in Iran (2008-2010).
improve knowledge of women’s reproductive health. J Womens Iran J Reprod Med 2014;12:561–566.
Health 2016;25:565–570.
Dulberg CS, Stephens T. The prevalence of infertility in Canada,
Brunetti P, Morabia A, Campana A, Marcus-Steiff J. Biometrical 1991–1992: analysis of three national surveys. The Prevalence of
study of reproduction conditions in the general-population: Infertility in Canada: Research Studies of the Royal Commission
method and initial results of surveys carried out in chambery- on New Reproductive Technologies 1993;6:, p. 61–106. Minister of
grenoble and martigny. Population 1994;49:27–60. Supply and Services Canada Ottawa: Ottawa, Canada.

Buckett W, Bentick B. The epidemiology of infertility in a rural Ekudayo O, Titilayo A, Anuodo O, Babalola O. Female genital
population. Acta Obstet Gynecol Scand 1997;76:233–237. cutting and infertility in marriage: a cross-sectional study
among women in Nigeria. Int J Educ Res 2020;8:65–78.
Bushnik T, Cook JL, Yuzpe AA, Tough S, Collins J. Estimating the
prevalence of infertility in Canada. Hum Reprod 2012;27:738–746. Ericksen K, Brunette T. Patterns and predictors of infertility
among African women: a cross-national survey of twenty-seven
Cabrera-Leon A, Lopez-Villaverde V, Rueda M, Moya-Garrido MN, nations. Soc Sci Med 1996;42:209–220.
Cabrera-León A. Calibrated prevalence of infertility in 30- to
49-year-old women according to different approaches: a cross- Esmaeilzadeh S, Delavar MA, Zeinalzadeh M, Mir M-RA.
sectional population-based study. Hum Reprod 2015;30:2677– Epidemiology of infertility: a population-based study in Babol,
2685. Iran. Women Health 2012;52:744–754.

Cai X, Song R, Long M, Wang S, Ma Y, Li X, Ai H, Shan X, Fu L, Eustache F, Auger J, Cabrol D, Jouannet P. Are volunteers
Liu Y. [A cross-sectional study on the current status of female delivering semen samples in fertility studies a biased
infertility in three counties of Xinjiang Uygur Autonomous population? Hum Reprod 2004;19:2831–2837.
Region]. Zhonghua Yi Xue Za Zhi 2011;91:3182–3185.
Fledderjohann J, Johnson DR. Impaired fertility and perceived
Cairncross ZF, Ahmed SB, Dumanski S, Nerenberg K, Metcalfe A. difficulties conceiving in Ghana: measurement problems and
Infertility and the risk of cardiovascular disease: findings from prospects. J Biosoc Sci 2016;48:431–456.
the Study of Women’s Health Across the Nation (SWAN). CJC
Open 2020; Fledderjohann J, Trinitapoli J, Billari F. Who seeks help?
Responses to perceived fertility impairments in Malawi.
Chandra A, Stephen EH. Infertility and impaired fecundity in International Popular Conference [Internet] 2017; International
the United States, 1982–2010: data from the National Survey of Union for the Scientific Study of Population (IUSSP). Available
Family Growth. Natl Health Stat Report 2013;1–19. from: https://iussp.confex.com/iussp/ipc2017/meetingapp.cgi/
Paper/3525.
Chauhan S, Kulkarni R, Agarwal D. Prevalence and factors
associated with chronic obstetric morbidities in Nashik district, Fuentes A, Devoto L. Infertility after 8 years of marriage: a pilot
Maharashtra. Indian J Med Res 2015;142:479–488. study. Hum Reprod 1994;9:273–278.

Chen J, Zhong C, Liang H, Yang Y, Zhang O, Gao E, Chen A, Geelhoed DW, Nayembil D, Asare K, Schagen van Leeuwen JH,
Yuan W, Wang J, Sun F, et al. The relationship between age at Roosmalen J van. Infertility in rural Ghana. Int J Gynaecol Obstet
menarche and infertility among Chinese rural women. Eur J 2002;79:137–142.
Obstet Gynecol Reprod Biol 2015;194:68–72.
Gleason JL, Shenassa ED, Thoma ME. Stressful life events, the
Crawford S, Fussman C, Bailey M, Bernson D, Jamieson DJ, incidence of infertility, and the moderating effect of maternal
Murray-Jordan M, Kissin DM. Estimates of lifetime infertility responsiveness: a longitudinal study. J Dev Orig Health Dis
from three states: the behavioral risk factor surveillance system. 2020;1–9.
J Womens Health 2015;24:578–586.
Gokler ME, Unsal A, Arslantas D. The prevalence of infertility and
Damone AL, Earnest A, Joham AE, Teede HJ, Moran LJ, Loxton D. loneliness among women aged 18-49 years who are living in semi-
Depression, anxiety and perceived stress in women with rural areas in western Turkey. Int J Fertil Steril 2014;8:155–162.
and without PCOS: a community-based study. Psychol Med
2019;49:1510–1520. Guldbrandsen K, Håkonsen LB, Ernst A, Toft G, Lyngsø J, Olsen J,
Ramlau-Hansen CH. Age of menarche and time to pregnancy.
Datta J, Palmer MJ, Tanton C, Gibson LJ, Jones KG, Macdowall Hum Reprod 2014;29:2058–2064.
W, Glasier A, Sonnenberg P, Field N, Mercer CH, et al. Prevalence
of infertility and help seeking among 15 000 women and men. Gunnell DJ, Ewings P. Infertility prevalence, needs assessment
Hum Reprod 2016;31:2108–2118. and purchasing. J Public Health Med 1994;16:29–35.

Executive summary Introduction Methods Results Discussion Annexes 38

Annex 2. Studies included in the systematic review

Győrffy Z, Dweik D, Girasek E. Reproductive health and burn-out Jacobson MH, Chin HB, Mertens AC, Spencer JB, Fothergill A,
among female physicians: nationwide, representative study Howards PP. “Research on infertility: definition makes a
from Hungary. BMC Womens Health 2014;14:121. difference” Revisited. Am J Epidemiol 2018;187:337–346.

Hærvig KK, Kierkegaard L, Lund R, Bruunsgaard H, Osler M, Jensen TK, Slama R, Ducot B, Suominen J, Cawood EHH,
Schmidt L. Is male factor infertility associated with midlife Andersen AG, Eustache F, Irvine S, Auger S, Jouannet P, et al.
low-grade inflammation? A population based study. Hum Fertil Regional differences in waiting time to pregnancy among
(Camb) 2018;21:146–154. fertile couples from four European cities. Hum Reprod
2001;16:2697–2704.
Hallén M, Sandblom G, Nordin P, Gunnarsson U, Kvist U,
Westerdahl J. Male infertility after mesh hernia repair: a Joffe M. Time trends in biological fertility in Britain. Lancet
prospective study. Surgery 2011;149:179–184. 2000;355:1961–1965.

Hassan KE. Prevalence of infertility and its impact on marital Karmaus W, Juul S. Infertility and subfecundity in population-
fertility, Egypt, 1993. 1997. Available from: http://www.zohry. based samples from Denmark, Germany, Italy, Poland and
com/dwb/khassan/pub/infertility.pdf. Spain. Eur J Public Health 1999;9:229–235.

He Y, Zheng D, Shang W, Wang X, Zhao S, Wei Z, Song X, Shi X, Katole A, Saoji A. Prevalence of primary infertility and its
Zhu Y, Wang S, et al. Prevalence of oligomenorrhea among associated risk factors in urban population of central India: a
women of childbearing age in China: a large community-based community-based cross-sectional study. Indian J Community
study. Womens Health 2020;16:1-9. Med 2019;44:337–341.

Herbert D, Lucke J, Dobson A. Infertility, medical advice and Kazemijaliseh H, Ramezani Tehrani F, Behboudi-Gandevani S,
treatment with fertility hormones and/or in vitro fertilisation: a Hosseinpanah F, Khalili D, Azizi F. The prevalence and causes
population perspective from the Australian Longitudinal Study of primary infertility in Iran: a population-based study. Glob J
on Women’s Health. Aust N Z J Public Health 2009a;33:358–364. Health Sci 2015;7:226–232.

Herbert DL, Lucke JC, Dobson AJ. Infertility in Australia circa Keiding N, Ali MM, Eriksson F, Matsaseng T, Toskin I, Kiarie J. The
1980: an historical population perspective on the uptake of use of time to pregnancy for estimating and monitoring human
fertility treatment by Australian women born in 1946-51. Aust N fecundity from demographic and health surveys. Epidemiology
Z J Public Health 2009b;33:507–514. 2021;32:27–35.

Hoenderboom BM, Bergen JEAM van, Dukers-Muijrers NHTM, Kirkegaard I, Uldbjerg N, Tabor A, Henriksen TB. Longer
Götz HM, Hoebe CJPA, Vries HJC de, Broek IVF van den, Vries F time-to-pregnancy in spontaneously conceived pregnancies
de, Land JA, Sande MAB van der, et al. Pregnancies and time to is associated with lower PAPP-A and free [beta]-hCG in
pregnancy in women with and without a previous chlamydia first trimester screening for Down syndrome. Prenat Diagn
trachomatis infection. Sex Transm Dis 2020;47:739–747. 2014;34:235–240.

Hollegaard S, Vogel I, Thorsen P, Jensen IP, Mordhorst C-H, Klemetti R, Raitanen J, Sihvo S, Saarni S, Koponen P. Infertility,
Jeune B. Chlamydia trachomatis C-complex serovars are a risk mental disorders and well-being: a nationwide survey. Acta
factor for preterm birth. In Vivo 2007;21:107–112. Obstet Gynecol Scand 2010;89:677–682.

Hosseini J, Emadedin M, Mokhtarpour H, Sorani M. Prevalence Klouman E, Manongi R, Knut-Inge Klepp. Self-reported and
of primary and secondary infertility in four selected provinces in observed female genital cutting in rural Tanzania: associated
Iran, 2010-2011. Iran J Obstet Gynecol Infertil 2012;15:1–7. demographic factors, HIV and sexually transmitted infections.
Trop Med Int Health 2005;10:105–1115.
Hu P, Cai C, Vinturache A, Hu Y, Gao Y, Zhang J, Lu M, Gu H,
Qiao J, Tian Y, et al. Maternal preconception body mass index Kreisel KM, Ikerdeu E, Cash HL, De Jesus SL, Kamb ML,
and time-to-pregnancy in Shanghai Women, China. Women Anderson T, Barrow RY, Sugiyama MS, Basilius K, Madraisau S.
Health 2020;60:1014–1023. An evaluation of infertility among women in the Republic of
Palau, 2016. Hawaii J Health Soc Welf 2020;79:7–15.
Huang J -t, Tang Y -g. Incidence of infertility and its influencing
factors among married residents in Guangdong province. Küppers-Chinnow M, Karmaus W. Prävalenz von verminderter
China Public Health = Zhongguo Gong Gong Wei Sheng Fruchtbarkeit und Inanspruchnahme ärztlicher Hilfe.
2013;29:0194–0197. Geburtshilfe Frauenheilkd 1997;57:89–95.

Jacob MC, McQuillan J, Greil AL. Psychological distress by type Larsen U. Primary and secondary infertility in sub-Saharan
of fertility barrier. Hum Reprod 2007;22:885–894. Africa. Int J Epidemiol 2000;29:285–291.

Executive summary Introduction Methods Results Discussion Annexes 39

Infertility prevalence estimates, 1990–2021

Larsen U. Infertility in central Africa. Trop Med Int Health Nasrabad HBR, Abbasi-Shavazi MJ, Hosseini-Chavoshi M,
2003;8:354–367. Karegar-Shoraki MR. Trend and patterns of childlessness in Iran.
Proceedings of the XXVII International Population Conference of
Larsen U. Research on infertility: which definition should we the IUSSP. 2013; pp. 26–31. Busan, Korea.
use? Fertil Steril 2005;83:846–852.
Nelson DB, Sammel MD, Patterson F, Lin H, Gracia CR,
Louis JF, Thoma ME, Sørensen DN, McLain AC, King RB, Freeman EW. Effects of reproductive history on symptoms of
Sundaram R, Keiding N, Buck Louis GM. The prevalence of menopause: a brief report. Menopause 2011;18:1143–1148.
couple infertility in the United States from a male perspective:
evidence from a nationally representative sample. Andrology Nguyen RH, Wilcox AJ, Skjærven R, Baird DD. Men’s body mass
2013;1:741–748. index and infertility. Hum Reprod 2007;22:2488–2493.

Magnus MC, Fraser A, Rich-Edwards JW, Magnus P, Lawlor DA, Oakley LL. The epidemiology of infertility: measurement,
Håberg SE. Time-to-pregnancy and risk of cardiovascular disease prevalence and an investigation of early life and reproductive
among men and women. Eur J Epidemiol 2021;36:383–391. risk factors. 2010; London School of Hygiene & Tropical
Medicine.
Mascarenhas MN, Cheung H, Mathers CD, Stevens GA. Measuring
infertility in populations: constructing a standard definition for Passey M, Mgone CS, Lupiwa S, Suve N, Tiwara S, Lupiwa T,
use with demographic and reproductive health surveys. Popul Clegg A, Alpers MP. Community based study of sexually
Health Metr 2012a;10:17. transmitted diseases in rural women in the highlands of Papua
New Guinea: prevalence and risk factors. Sex Transm Infect
Mascarenhas MN, Flaxman SR, Boerma T, Vanderpoel S, 1998;74:120–127.
Stevens GA. National, regional, and global trends in infertility
prevalence since 1990: a systematic analysis of 277 health Pedersen KK, Hagen C, Eshoj O. Infertility and pregnancy
surveys. PLoS Med 2012b;9:e1001356. outcome in women with insulin-dependent diabetes mellitus.
An epidemiological study. Ugeskr Laeger 1994;156:6196–6200.
McQuillan J, Greil AL, White L, Jacob MC. Frustrated fertility:
infertility and psychological distress among women. J Marriage Philippov OS, Radionchenko AA, Bolotova VP, Voronovskaya
Fam 2003;65:1007–1018. NI, Potemkina TV. Estimation of the prevalence and causes
of infertility in western Siberia. Bull World Health Organ
Mena GP, Mielke GI, Brown WJ. Do physical activity, sitting time 1998;76:183–187.
and body mass index affect fertility over a 15-year period in
women? Data from a large population-based cohort study. Hum Pick WM, Obermeyer CM. Urbanization, household composition
Reprod 2020;35:676–683. and the reproductive health of women in a South African city.
Soc Sci Med 1996;43:1431–1441.
Meng Q, Ren A, Zhang L, Liu J, Li Z, Yang Y, Li R, Ma L. Incidence
of infertility and risk factors of impaired fecundity among newly Polis CB, Cox CM, Tunçalp Ö, McLain AC, Thoma ME. Estimating
married couples in a Chinese population. Reprod Biomed Online infertility prevalence in low-to-middle-income countries: an
2015;30:92–100. application of a current duration approach to Demographic and
Health Survey data. Hum Reprod 2017;32:1064–1074.
Merritt MA, De Pari M, Vitonis AF, Titus LJ, Cramer DW, Terry KL.
Reproductive characteristics in relation to ovarian cancer risk by Priestley SR. Impaired fertility in Jamaica: evidence from
histologic pathways. Hum Reprod 2013;28:1406–1417. fertility surveys. West Indian Med J 2012;61:716–725.

Miller-Fellows SC, Howard L, Kramer R, Hildebrand V, Furin J, Purkayastha N, Sharma H. Prevalence and potential
Mutuku FM, Dunstan Mukoko, Ivy JA, King CH. Cross-sectional determinants of primary infertility in India: evidence from
interview study of fertility, pregnancy, and urogenital Indian demographic health survey. Clin Epidemiol Glob Health
schistosomiasis in coastal Kenya: documented treatment in 2021;9:162–170.
childhood is associated with reduced odds of subfertility among
adult women. PLoS Negl Trop Dis 2017;11:e0006101. Raatikainen K, Harju M, Hippeläinen M, Heinonen S. Prolonged
time to pregnancy is associated with a greater risk of adverse
Mirzaei M, Namiranian N, Dehghani Firouzabadi R, Gholami S. The outcomes. Fertil Steril 2010;94:1148–1151.
prevalence of infertility in 20-49 years women in Yazd, 2014-2015:
a cross-sectional study. Int J Reprod Biomed 2018;16:683–688. Rao N, Esber A, Turner A, Mopiwa G, Banda J, Norris A. Infertility
and self-rated health among Malawian women. Women Health
Muller A, Slama R, Labbé-Declèves C, Jouannet P, Bujan L, 2018;58:1081–1093.
Mieusset R, Le Lannou D, Guerin J-F, Benchaib M, Spira A.
Geographic variations in probability of pregnancy in four cities
of France. Rev Epidemiol Sante Publique 2006;54:55–60.

Executive summary Introduction Methods Results Discussion Annexes 40

Annex 2. Studies included in the systematic review

Righarts A, Dickson NP, Ekeroma A, Gray AR, Parkin L, Gillett Somé EN, Boncoungou J, Poda JN. Prevalence of the infertility
WR. The burden of infertility in New Zealand: a baseline survey among couples in Ouagadougou (Burkina Faso): A population-
of prevalence and service use. Aust N Z J Obstet Gynaecol based survey. Open Public Health J 2016;9:88-97.
2021;61:439–447.
Song S. Assessing the impact of in utero exposure to famine on
Righarts A, Dickson NP, Parkin L, Gillett WR. Infertility and fecundity: evidence from the 1959-61 famine in China. Popul
outcomes for infertile women in Otago and Southland. N Z Med Stud 2013;67:293–308.
J 2015;128:43–53.
Sundby J, Mboge R, Sonko S. Infertility in the Gambia: frequency
Risch HA, Marrett LD, Howe GR. Parity, contraception, infertility, and health care seeking. Soc Sci Med 1998;46:891–899.
and the risk of epithelial ovarian cancer. Am J Epidemiol
1994;140:585–597. Sundby J, Schei B. Infertility and subfertility in Norwegian
women aged 40-42: prevalence and risk factors. Acta Obstet
Roode T van, Dickson NP, Righarts AA, Gillett WR. Cumulative Gynecol Scand 1996;75:832–837.
incidence of infertility in a New Zealand birth cohort to age
38 by sex and the relationship with family formation. Fertil Steril Taponen S, Ahonkallio S, Martikainen H, Koivunen R,
2015;103:1053-1058.e2. Ruokonen A, Sovio U, Hartikainen A-L, Pouta A, Laitinen J,
King V, et al. Prevalence of polycystic ovaries in women with
Rostad B, Schmidt L, Sundby J, Schei B. Has fertility declined self‐reported symptoms of oligomenorrhoea and/or hirsutism:
from mid-1990s to mid-2000s? Acta Obstet Gynecol Scand Northern Finland Birth Cohort 1966 Study. Hum Reprod
2013;92:1284–1289. 2004;19:1083–1088.

Rutstein SO, Shah IH. Infecundity, infertility, and childlessness Taylor GM, Faragher EB, Chantler E, Seif MW. Fecundity in the
in developing countries [Internet]. 2004; OCR Macro and World modern city: a comparison of couples attending antenatal
Health Organization. Available from: https://www.who.int/ clinics in Manchester (UK) and Melbourne (Australia). J Obstet
reproductivehealth/publications/infertility/DHS_9/en/. Gynaecol 1999;19:489–495.

Safarinejad MR. Infertility among couples in a population-based Terävä A-N, Gissler M, Hemminki E, Luoto R. Infertility and the
study in Iran: prevalence and associated risk factors. Int J Androl use of infertility treatments in Finland: prevalence and socio-
2008;31:303–314. demographic determinants 1992–2004. Eur J Obstet Gynecol
Reprod Biol 2008;136:61–66.
Samarakoon S, Rajapaksa L, Seneviratne HR. Prevalence of
primary and secondary infertility in the Colombo District. Thoma ME, McLain AC, Louis JF, King RB, Trumble AC,
Ceylon J Med Sci 2007;45:83–91. Sundaram R, Buck Louis GM. Prevalence of infertility in
the United States as estimated by the current duration
Sarac M, Koc I. Prevalence and risk factors of infertility in Turkey: approach and a traditional constructed approach. Fertil Steril
evidence from demographic and health surveys, 1993-2013. J 2013;99:1324–1331.
Biosoc Sci 2018;50:472–490.
Toft G, Axmon A, Giwercman A, Thulstrup AM, Rignell-Hydbom A,
Sharif SN, Azizi Kutenaee M, Darsareh F, Roozbeh N. Prevalence Pedersen HS, Ludwicki JK, Zvyezday V, Zinchuk A, Spano M, et al.
and risk factors of infertility in a southern port city of Iran. Fertility in four regions spanning large contrasts in serum levels
Hormozgan Medical Journal 2020;24: e99412. of widespread persistent organochlorines: a cross-sectional
study. Environ Health 2005;4:26.
Singh BP, Shukla U. Inability to conceive and treatment-seeking
behaviour in Uttar Pradesh state in India. Can Stud Popul Udgiri R, Patil VV. Comparative study to determine the
2015;42:1–12. prevalence and socio-cultural practices of infertility in rural and
urban field practice area of tertiary care hospital, Vijayapura,
Slama R, Ducot B, Carstensen L, Lorente C, La Rochebrochard Karnataka. Indian J Community Med 2019;44:129–133.
E de, Leridon H, Keiding N, Bouyer J. Feasibility of the current-
duration approach to studying human fecundity. Epidemiology Unisa S. Childlessness in Andhra Pradesh, India: treatment-
2006;17:440–449. seeking and consequences. Reprod Health Matters 1999;7:54–64.

Slama R, Hansen OKH, Ducot B, Bohet A, Sorensen D, Vahidi S, Ardalan A, Mohammad K. Prevalence of primary
Giorgis Allemand L, Eijkemans MJC, Rosetta L, Thalabard JC, infertility in the Islamic Republic of Iran in 2004-2005. Asia Pac J
Keiding N, et al. Estimation of the frequency of involuntary Public Health 2009;21:287–293.
infertility on a nation-wide basis. Hum Reprod 2012;27:1489–1498.
Van der Avoort IAM, Van Golde RJT, Tuerlings JHAM, Kiemeney LA,
Soares S, Rodrigues T, Barros H. [Infertility prevalence in the city Meuleman EJH, Braat DDM, Kremer JAM. Underestimation of
of Porto]. Acta Med Port 2011;24:699–706. subfertility among relatives when using a family history: taboo
bias. J Androl 2003;24:285–288.

Executive summary Introduction Methods Results Discussion Annexes 41

Infertility prevalence estimates, 1990–2021

Walraven G, Scherf C, West B, Ekpo G, et al. The burden of Xingping G, Yu W, Qiufang H. Prevalence of infertility in rural
reproductive-organ disease in rural women in the Gambia, West areas of Shanxi Province. Chin J Fam Plan 2006;14:358.
Africa. Lancet 2001;357:1161–1167.
Yang F, Li L, Chen J-P, Liu X-Q, Zhong C-L, Yang Y, Ren Y-F, Yuan W,
Wang B, Zhou W, Zhu W, Chen L, Wang W, Tian Y, Shen L, Liang H, Miao M-H. Couple’s infertility in relation to male
Zhang J. Associations of female exposure to bisphenol A with smoking in a Chinese rural area. Asian J Androl 2017;19:311–315.
fecundability: evidence from a preconception cohort study.
Environ Int 2018;117:139–145. Yang Y-Q, Shen H, Chen J, Chen Z-W. [A prevalence survey
of infertility in Beijing, China]. Zhonghua Yi Xue Za Zhi
Weiss HA, Troisi R, Rossing MA, Brogan D, Coates RJ, Gammon MD, 2011;91:313–316.
Potischman N, Swanson CA, Brinton LA. Fertility problems and
breast cancer risk in young women: a case-control study in the Zargar AH, Wani AI, Masoodi SR, Laway BA, Salahuddin M.
United States. Cancer Causes Control 1998;9:331–339. Epidemiologic and etiologic aspects of primary infertility in the
Kashmir region of India. Fertil Steril 1997;68:637–643.
Woodall PA, Kramer MR. Schistosomiasis and infertility in East
Africa. Am J Trop Med Hyg 2018;98:1137. Zhang H, Wang S, Zhang S, Wang T, Deng X. Increasing trend of
prevalence of infertility in Beijing. Chin Med J 2014;127:691–695.
Wu S, Tian J, Wang M, Pan B, Lü H, Wang Z, Li H. [The effect
of cadmium pollution on reproductive health in females]. Zhang X -h, Zhang R. Analysis on infertility prevalence and
Zhonghua Liu Xing Bing Xue Za Zhi 2004;25:852–855. its affecting factors in Gansu province. Reprod Contracept
2013;33:184–192.
Wulff M, Högberg U, Stenlund H. Infertility in an industrial
setting: a population-based study from Northern Sweden. Acta Zhou Z, Zheng D, Wu H, Li R, Xu S, Kang Y, Cao Y, Chen X, Zhu Y,
Obstet Gynecol Scand 1997;76:673–679. Chen Z, et al. Epidemiology of infertility in China: a population-
based study. BJOG 2018;125:432–441.

Executive summary Introduction Methods Results Discussion Annexes 42

Annex 3. Summary of included studies
by region
AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

AFRICAN REGION
Somé et al. Burkina Faso Cross- Households: Women: Retrospective Childbearing age women/ All women/men in Lifetime 12 Infertility - women 10.4 (7.9-13.5)
(2016) sectional 480 18 - 45 Time-to- men who have never born union/living with Primary infertility 6.8 (4.8-9.4)
Men: Pregnancy (TTP) a child and who had been partner Secondary infertility 3.6 (2.2-5.7)
18 - 55 Design seeking a child for more than
12 months Infertility - men
Primary infertility 9.3 (7.0-12.2)
Secondary infertility 4.8 (3.2-7.2)
4.4 (2.9-6.7)

Sundby et al. Gambia Cross- 3 000 15 - 49 Self-reported Women with no pregnancy All women Period 12 Infertility 9.5
(1998) sectional infertility measure or live children born despite Primary infertility 3.3
(Direct) being married and not
having used family planning
for at least a year

Women who are married, not All women Period 36 Secondary 6.2
using contraceptives and not subfertility
breastfeeding and have had
no birth of a child for the last
three years.

Walraven et Gambia Cross- 871 < 45 years Self-reported Women trying to conceive Married women not Period 12 Infertility 9·8 (8·2–11·6)
al. (2001) sectional infertility measure for at least one year using contraception,
(Direct) without success despite postmenarchal and
regular (one time per premenopausal
week) sexual intercourse,
no use of contraception,
postmenarchal and
premenopausal.

Executive summary Introduction Methods Results Discussion Annexes 43

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Geelhoed et Ghana Cross- Women: 1073 Reproductive Self-reported Women/men who All women/men Lifetime 12 Infertility - women 11.8
al. (2002) sectional Men: 1064 age infertility measure experienced failure to Primary infertility 0.6
(Direct) achieve conception after at Secondary infertility 11.2
least one year of exposure
Infertility - men 15.8
Primary infertility 6.8
Secondary infertility 9.0

Fledderjoha- Ghana Cohort 1 350 15 - 49 Self-reported Women reporting difficulties Married or in union Period No Self-assessed
nn and infertility measure conceiving (takes a long time women duration difficulties
Johnson (Direct) to get pregnant when they conceiving:
(2016) want to and/or can no longer Unadjusted 65.0
become pregnant) Adjusted 20.0

Unadjusted: Does not

account for contraceptive
use or fertility desires

Adjusted: Women who desire

a child and are not currently
using modern contraception

Constructed Woman without a birth after Women in union Period 12 Infertility - 69.0
infertility measure 12/24/60/84 months 24 Unadjusted 64.0
(Indirect) 60 35.0
Unadjusted: Does not 80 24.0
account for contraceptive
use or fertility desires 12 Adjusted 17.0
24 15.0
Adjusted: Women who desire 60 7.0
a child and are not currently 80 4.0
using modern contraception

Miller-Fellows Kenya Cross- 160 15 - 45 Self-reported Women who were in a Women who had ever Lifetime 60 or 12 Subfertility 44.0 (37.0 – 52.0)
et al. (2017) sectional infertility measure sexual union and not using been married or in Primary infertility 2.0 (0.3 – 5.0)
(Direct) contraception for at least 5 a co-residing sexual Secondary infertility 42.0 (35.0 – 51.0)
years and did not have a live union, were currently
birth and/or women who pregnant, and/or had
reported a period of over one given birth to at least
year without a pregnancy one child.
with regular, unprotected
sexual intercourse

Barden- Malawi Cohort Women: 678 Women: Self-reported Women/men who said they Women/men who had Lifetime No Perceived infertility:
O'Fallon Men: 362 15 - 34 infertility measure ever experienced difficulty ever been pregnant duration Women 19.6
(2005) Men: 20 - 44 (Direct) becoming pregnant or tried to become Men 19.6
pregnant

Executive summary Introduction Methods Results Discussion Annexes 44

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Fledderjoha- Malawi Cross- 116 15 - 25 Self-reported Women who reported that All women Lifetime No Impaired fertility 12.8
nn et al. sectional infertility measure they had a lot or some duration
(2017) (Direct) difficulty conceiving and/or
carrying a pregnancy

Rao et al. Malawi Cross- 915 15 - 39 Self-reported Women who had ever tried All women Lifetime 24 Infertility 20.0
(2018) sectional infertility measure and failed to conceive a
(Direct) pregnancy for 2 years or
longer

Polis et al. Nigeria Cross- 6 340 18 - 44 Current Duration Couples not yet pregnant by Women at risk of Period 12 Infertility 31.1(27.9 - 34.7)
(2017) sectional Design 12/24/36 months (estimated) pregnancy (18–44 Primary infertility 17.4 (12.9 - 23.8)
years, married or Secondary infertility 34.1 (30.3 - 39.3)
cohabitating, sexually
active within the past 4 24 Infertility 17.7 (15.7 - 20)
weeks and not currently Primary infertility 10.0 (7.0 - 14.3)
using contraception Secondary infertility 19.2 (17.1 - 22.1)
and had not been
sterilized) 36 Infertility 11.5 (10.2 - 13.0)
Primary infertility 6.8 (4.6 - 10)
Secondary infertility 12.3 (11.0 - 14.1)

Ekudayo Nigeria Cross- 16 922 15 - 49 Constructed Women who reported 0 Married women or Period 24 Infertility 3.1
(2020) sectional infertility measure as the total number of women living with a
(Indirect) children ever born, who partner who reported
were not pregnant at the at least 1 child or
time of the survey and who were presently
were not presently using pregnant (unclear if
contraception contraceptive users
were included)

Bello et al. South Africa Cross- 482 18 - 49 Retrospective Women with a TTP greater Women who had Period 6 TTP > 6 months 50.0
(2010) sectional Time-to- than 6/12/24 months for the been pregnant and 12 TTP > 12 months 32.0
Pregnancy (TTP) most recent pregnancy or planned their most 24 TTP > 24 months 17.0
Design current pregnancy attempt recent pregnancies and
(if pregnancy has not been women who had never
achieved) been pregnant but were
trying

Pick and South Africa Cross- 298 15+ Self-reported Women not using any form Women who were Lifetime No Infertility 19.0
Obermeyer sectional infertility measure of contraception at the time not using any form of duration
(1996) (Direct) of the survey who reported contraception at the
difficulty in becoming time of the survey
pregnant

Executive summary Introduction Methods Results Discussion Annexes 45

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Women with no pregnancy Women who were Period 60 Infertility 10.0

in the preceding five years not using any form of
who reported difficulty in contraception at the
becoming pregnant and time of the survey
were not using contraception
at the time of the survey

Larsen (2005) United Republic Cross- 993 20 - 44 Retrospective Women who have been Women in first union Period 24 Infertility 12.1 (9.4 - 14.8)
of Tanzania sectional Time-to- having intercourse without (married or consensual Primary infertility 2.5 (1.5 - 3.5)
Pregnancy (TTP) using contraception or trying union) for at least two Secondary infertility 9.6 (7.3 - 11.9)
Design in any way to delay or avoid years
getting pregnant for at least
two years without conceiving

Women who have tried to Women in first union Period 24 Infertility 6.9 (5.2 - 8.6)
conceive for at least two (married or consensual Primary infertility 1.8 (0.9 - 2.7)
years union) for at least two Secondary infertility 5.0 (3.5 - 6.5)
years

1 120 20 - 44 Self-reported Women who report ever Women in first union Lifetime No Infertility 10.3 (8.4 - 12.2)
infertility measure having problems getting (married or consensual duration Primary infertility 2.9 (1.9 - 3.9)
(Direct) pregnant union) Secondary infertility 7.4 (6.0 - 8.8)

720 20 - 44 Constructed Women who have had Women in first union Period 60 Infertility 11.5 (9.2 - 13.7)
infertility measure no births at least 5 years (married or consensual Secondary infertility 11.1 (9.8 - 12.6)
(Indirect) subsequent to last birth or union) for at least five
marriage, if childless. years

Women who have had Women in first union Period 60 Infertility 5.5 (3.9 - 7.1)
no births at least 5 years (married or consensual Secondary infertility 4.8 (3.9 - 5.7)
subsequent to last birth or union) for at least five
marriage, if childless despite years
confirming that she wants
a(nother) child

Women married at least five/ Women in first union Lifetime 60 Primary infertility 3.5 (2.3 - 4.7)
seven years without ever (married or consensual 84 1.9 (0.8 - 3.0)
having a child union) for at least five/
seven years

Bernhard et United Republic Cross- 530 > 15 Self-reported Women with unsuccessful All women Lifetime 6 Primary infertility 1.5
al. (2000) of Tanzania sectional infertility measure attempt to conceive for more Secondary infertility 5.1
(Direct) than 6 months

Klouman et United Republic Cross- 636 15 - 44 Undetermined Women who were unable All women Period 12 Infertility 10.3
al. (2005) of Tanzania sectional to become pregnant within Primary infertility 3.1
a year of living with their Secondary infertility 7.2
partners

Executive summary Introduction Methods Results Discussion Annexes 46

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Larsen (2003) Cameroon, Cross- Cameroon: 15 - 49 Constructed Childless women married for Women married for at Period 60 Primary infertility 3.1 - 6.9
Central African sectional 3091 infertility measure at least five years least five years (current (range)
Republic, CAR: 3783 (Indirect) contraceptive users
Gabon, and Chad: 5068 classified as fertile)
Chad Gabon: 3205

Childless women married for Women married for Period 60 Primary infertility 3.2 - 7.0
at least five years at least five years (range)
(contraceptive use not
taken into account)

Cameroon: 20 - 44 Ever-married, parous women Ever-married women Period 60 Secondary infertility 18.9 - 26.3
2819 who have had no live births who had at least one (range)
CAR: 3219 during the last 5 years before child and had been
Chad: 4418 the interview observed at least 5
Gabon: 3134 years subsequent to the
birth date of their first
child (contraceptive
users classified as
fertile)

Ever-married, parous women Ever-married women Period 60 Secondary infertility 19.1 - 29.4
who have had no live births who had at least one (range)
during the last 5 years before child and had been
the interview observed at least 5
years subsequent to the
birth date of their first
child (contraceptive use
not taken into account)

Larsen (2000) Multiple Cross- Range: 1361 - 20 - 44 Constructed Women who have had no Ever-married women Period 60/84 Infertility: Range 7.0 - 28.0
countries (22 sectional 5869 infertility measure livebirths during the last (contraceptive users 60 Secondary: Range 7.0 - 25.0
countries with (Indirect) 5 (secondary)/7 (primary) classified as fertile)
data in or after years before censoring
1990) (i.e., the month of survey
or the month of last sexual
intercourse, whichever
came first)

Women who have had no Ever-married women Period 60/84 Infertility: Range 7.0 - 29.0
livebirths during the last (contraceptive use not 84 Primary: Range 1.0 - 6.0
5 (secondary)/7 (primary) taken into account) 60 Secondary: Range 7.0 - 26.0
years before censoring
(i.e., the month of survey
or the month of last sexual
intercourse, whichever came
first)

Executive summary Introduction Methods Results Discussion Annexes 47

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Ericksen and Multiple Cross- Range: 20 - 41 Constructed Women who had been All women exposed to Period 60 Infertility: Range 8.9 - 16.6
Brunette countries sectional 1499 - 6206 infertility measure married or sexually conception 84 12.0 - 20.8
(1996) (12 countries (Indirect) experienced without a birth
with data in or for at least 5/7years
after 1990)

Woodall and Ethiopia, Kenya, Cross- 17 547 15 - 49 Constructed Women with no live birth Women married or Period 60 Infertility 35.0
Kramer (2018) United Republic sectional infertility measure within the last 5 years who in union for at least Primary infertility 3.0
of Tanzania, and (Indirect) have been married or in 5 years, not using a
Uganda union for at least five years, contraceptive, and not
are not using contraception, presently pregnant
and are not currently
pregnant

REGION OF THE AMERICAS

Bushnik et al. Canada Cross- 3 225 900 18-44 Constructed Couples who did not become Couples not using any Period 12 Infertility 15.7 (14.2 - 17.4)
(2012) sectional infertility measure pregnant after exposure to form of birth control Primary infertility 20.6 (17.8 - 23.7)
(Indirect) the risk of conception during within the past 12
the previous 12 months months

3 176 900 Couples who did not become Couples who did not Period 12 Infertility 14 (12.6 - 15.6)
pregnant after exposure to use any form of birth Primary infertility 18.7 (15.9 - 21.7)
the risk of conception during control within the
the previous 12 months past 12 months and
reported having sexual
intercourse in the past
12 months

Couples who did not become Couples who did Period 12 Infertility 11.5 (10.2 - 12.9)
pregnant after exposure to not use any form of Primary infertility 10.2 (8.3 - 12.5)
the risk of conception during birth control within
the previous 12 months the past 12 months,
reported having
sexual intercourse in
the past 12 months,
and reported ever
having tried to become
pregnant with their
current partner

Risch et al. Canada Case- 564 35 - 79 Self-reported Women who ever had All women (control Lifetime No Secondary infertility 7.6
(1994) control infertility measure an interval of time when group) duration
(Direct) pregnancy was attempted
without success

Executive summary Introduction Methods Results Discussion Annexes 48

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Dulberg and Canada Cross- 1 413 18 - 44 Constructed Women and their husband/ Women who had been Period 12 Infertility 8.5 (7.0 - 9.9)
Stephens sectional (12-month) infertility measure partner who did not use married or cohabitating 24 7.0 (5.6 - 8.4)
(1993) 1 350 (Indirect) any contraceptive method for at least 12/24
(24-month) (non-surgical or surgical) months prior to the
and was not pregnant during survey
12/24 months prior to the
interview.

Balakrishnan Canada Cross- 7 765 16 - 49 Self-reported Women who believe she or All women in union Period No Perceived infertility 3.19
and Maxim sectional infertility measure her partner are incapable of duration Primary infertility 5.22
(1993) (Direct) having children Secondary infertility 2.79

9 267 Constructed Women who reported that All women in union Period 12 Inferred infertility 6.66
infertility measure neither they nor their partner Primary infertility 11.23
(Indirect) were sterilized, had not used Secondary infertility 5.31
any form of contraception in
the year prior to the survey,
and were not currently
pregnant or post-partum
mothers

7 765 Women classified as All women in union Period 12 or no Aggregate infertility 7.74
perceived infertile (direct duration Primary infertility 11.68
question) and/or inferred Secondary infertility 6.72
infertile (indirect questions)

Fuentes and Chile Cross- 365 15 - 45 Self-reported Women married for one Women at risk of Period 12 Infertility 10.14
Devoto (1994) sectional infertility measure year failing to achieve pregnancy in their first (prevalence) (9.15 - 11.05)
(Direct) pregnancy after one or more year of marriage (i.e.,
years of unprotected sexual having unprotected
intercourse intercourse).

270 Women married for eight Women at risk of Period 12 Infertility

years failing to achieve pregnancy in their (prevalence) 7.04
pregnancy after one or more eighth year of marriage Primary infertility 3.33 (2.74 - 3.72)
years of unprotected sexual (i.e., having unprotected Secondary infertility 3.71 (3.28 - 4.14)
intercourse intercourse).

474 Women married for one Married women Period 12 Infertility 7.8
year failing to achieve (frequency)
pregnancy after one or more
years of unprotected sexual
intercourse

Women married for eight Married women Period 12 Infertility 4.01

years failing to achieve (frequency)
pregnancy after one or more
years of unprotected sexual
intercourse

Executive summary Introduction Methods Results Discussion Annexes 49

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Women suffering from Married women Lifetime 12 Infertility 25.74 (+/- 3.9)
infertility at some point
in their lives disregarding
whether they are currently
infertile or not

Priestley Jamaica Cross- 8 180 15 - 49 Constructed Sexually experienced non- Sexually experienced Period 24 Impaired fecundity 31
(2012) sectional infertility measure contracepting women who women Primary 12.2
(Indirect) are non-surgically sterile, Secondary 36.1
subfecund (women who
think pregnancy is difficult),
or have a long interval
without contraception (more
than 24 months without the
use of contraception)

Jacobson et United States of Cross- 1 014 22 - 45 Retrospective Women who did not Women who were Lifetime 6 Infertility 42.6
al. (2018) America sectional Time-to- get pregnant after at risk of becoming 12 35.3
Pregnancy (TTP) 6/12/24 months of regular pregnant 24 23.5
Design (>3 times per month)
unprotected sex

Women who did not Women who were Lifetime 6 Infertility 25.5
get pregnant after at risk of becoming 12 19.7
6/12/24 months of regular pregnant 24 11.2
(>3 times per month)
unprotected sex while
actively trying to become
pregnant

Women who did not get Women who were Lifetime 12 Infertility 35.9
pregnant after 12 months of at risk of becoming (6 ≥35yrs)
regular (>3 times per month) pregnant
unprotected intercourse for
those <35 years old or after 6
months for those ≥35 years
old

Women who did not get Women who were Lifetime 12 Infertility 20.5
pregnant after 12 months of at risk of becoming (6 ≥35yrs)
regular (>3 times per month) pregnant
unprotected intercourse for
those <35 years old or after 6
months for those ≥35 years
old while actively trying to
become pregnant

Executive summary Introduction Methods Results Discussion Annexes 50

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Crawford et United States of Cross- Florida: 1 285 18 - 50 Self-reported Those who ever tried to get All adults Lifetime 12 Infertility
al. (2015) America sectional MA: 1 302 infertility measure pregnant for a year or longer Florida 9.7 (7.6–11.8)
Michigan: (Direct) and were unable to do so Massachusetts 6.0 (4.6–7.5)
3 360 Michigan 4.2 (3.5–5.0)

Florida: 736 Those who ever tried to get All adults who ever Lifetime 12 Florida 25.3
MA: 1 246 pregnant for a year or longer tried to get pregnant Massachusetts 9.9
Michigan: and were unable to do so Michigan 5.8
2 742

Thoma et al. United States of Cross- 277 15-44 Current Duration Women who want to become Women at risk of Period 12 TTP > 12 months 15.5 (8.6 - 27.5)
(2013) America sectional Design pregnant with a TTP greater pregnancy at time of Primary 24.30 (12.4-43.5)
than 12 months (estimated) interview (not using a
method of contraception
nor pregnant but were
sexually active at the
time of interview AND
responded "Yes" to
the question ‘‘Is the
reason you are not
using a method of birth
control now because
you, yourself, want to
become pregnant as
soon as possible?’’)

222 Women who want to become Women at risk of Period 12 TTP > 12 months 12.6 (7.6-21.4)
pregnant with a TTP greater pregnancy at time of Primary 18.3 (11.0-30.9)
than 12 months (estimated) interview (not using a
method of contraception
nor pregnant but were
sexually active at the
time of interview AND
responded "Yes" to the
question ‘‘Is the reason
you are not using a
method of birth control
now because you,
yourself, want to become
pregnant as soon as
possible?’’) AND did not
report use of current
infertility treatment

Executive summary Introduction Methods Results Discussion Annexes 51

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

3 812 Constructed Married or cohabiting All married or Period 12 Infertility 7.0 (6.2 - 7.8)
infertility measure respondents who had been cohabiting respondents Primary infertility 13.2 (11.2 - 15.2)
(Indirect) in a continuous relationship Secondary infertility 5.3
for 12 months or more with
no use of contraception, but
sexually active every month
for the past 12 months, and
did not have a pregnancy

Not reported Married or cohabiting All married or Period 12 Infertility 6.6 (5.8-7.5)
respondents who had been cohabiting respondents Primary infertility 11.9 (9.9-13.9)
in a continuous relationship who did not report
for 12 months or more with current use of infertility
no use of contraception, but treatment.
sexually active every month
for the past 12 months, and
did not have a pregnancy
AND did not report current
use of infertility treatment

Louis et al. United States of Cross- 157 15-45 Current Duration Men who are in a Men at risk of Period 12 Infertility 12.0 (7.0 - 23.2)
(2013) America sectional Design relationship and trying to pregnancy at time of Primary infertility 14.0 (6.0 - 25.6)
become pregnant with a TTP interview (sexually
> 12 months (estimated) active in the past year
with a female partner
and currently trying to
get pregnant)

Not reported Men who are in a Men at risk of Period 12 Infertility 9.4 (5.2-17.2)
relationship and trying to pregnancy at time of
become pregnant with a TTP interview (sexually
> 12 months (estimated) active in the past year
with a female partner
and currently trying
to get pregnant) and
who reported not
having sought infertility
treatment during their
current pregnancy
attempt.

Executive summary Introduction Methods Results Discussion Annexes 52

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Chandra et al. United States of Cross- 61 755 15-44 Hybrid* Women with impaired All women Period No Impaired fecundity 10.9 (SE = 0.4)
(2013) America sectional (Weighted fecundity (i.e. non-surgically duration Primary infertility 11.2 (SE = 0.7)
population Primary: sterile (self or partner), or 36 Secondary infertility 10.6 (SE = 0.6)
size) Self-reported subfecund (self or partner),
infertility measure and/or long interval (36
12 279 (Direct) or more months) without
(unweighted conception)
sample size) Secondary:
Constructed
infertility measure
(Indirect)

25 605 Women with impaired Married women Period No Impaired fecundity 12.1 (SE = 0.8)
(Weighted fecundity (i.e. non-surgically duration Primary infertility 21.2 (SE = 2.0)
population sterile (self or partner), or 36 Secondary infertility 9.9 (SE = 0.8)
size) subfecund (self or partner),
and/or long interval (36
3 971 or more months) without
(unweighted conception)
sample size)

62 128 Self-reported Men who are nonsurgically All men Period No Infertility 9.4 (SE = 0.5)
(Weighted infertility measure sterile (self or partner) or duration
population (Direct) subfertile
size)

5 422 Constructed Women and their partner Married women Period 12 Infertility 6.0 (SE = 0.5)
(unweighted infertility measure who, during the previous Primary infertility 14.0 (SE = 1.6)
sample size) (Indirect) 12 months or longer, were Secondary infertility 4.0 (SE = 0.5)
continuously married, were
sexually active each month,
had not used contraception,
and had not become
pregnant.

Boulet et al. United States of Cross- 8 691 18-50 Self-reported Women who reported All women Lifetime No Perceived infertility 13.20 (11.3-15.2)
(2016) America sectional infertility measure difficulty becoming or duration
(Direct) staying pregnant

Nelson et al. United States of Cross- 291 35 - 47 Self-reported Women who ever tried to get All women Lifetime 12 Infertility 20.0
(2011) America sectional infertility measure pregnant for over one year
(Direct) without being able to

Executive summary Introduction Methods Results Discussion Annexes 53

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

McQuillan et United States of Cross- 580 25 - 50 Self-reported Women who have (a) tried All women Lifetime 12 Subfecundity 35.0
al. (2003) America sectional infertility measure for longer than 12 months (medically defined
(Direct) to conceive any of their infertility)
pregnancies, (b) sought
medical help to conceive
any of their pregnancies,
(c) ever tried to get pregnant
for more than 12 months
without success, and/or (d)
ever had regular unprotected
intercourse for more than a
year without pregnancy.

Weiss et al. United States of Cross- 1 989 Seattle and Self-reported Women who report difficulty All women Lifetime No Infertility 19.0
(1998) America sectional New Jersey: infertility measure in either becoming pregnant duration Primary infertility 31.0
20 - 44 (Direct) or maintaining a pregnancy.

Atlanta:
20 - 54

Merritt et al. United States of Case- 2 100 Not reported Self-reported Women who had tried to All women (control Lifetime No Infertility 20.8
(2013) America control infertility measure become pregnant without group) duration
(Direct) success or had seen a doctor
about having difficulties in
getting pregnant or carrying
a pregnancy to term.

Jacob et al. United States of Cross- 580 25 - 50 Self-reported Women who experienced All women Lifetime 12 Infertility 28.0
(2007) America sectional infertility measure 12 months of regular
(Direct) unprotected intercourse
without conception at some
time in their lives

Gleason et al. United States of Cohort 1 652 29 - 35 Self-reported Women who reported Women who ever tried Lifetime 12 Infertility 24.1
(2020) America infertility measure regular sexual intercourse to get pregnant
(Direct) over at least 12 months
without the use of
contraception and without
conceiving a child and ever
tried to get pregnant

Anyalechi et United States of Cross- 2 628 18-49 Self-reported Women who had ever Women who ever had Lifetime 12 Infertility 13.80 (12.3-15.3)
al. (2019) America sectional infertility measure had sexual intercourse intercourse with a male
(Direct) with a male partner and partner
attempted to get pregnant
for 12 months without
becoming pregnant

Executive summary Introduction Methods Results Discussion Annexes 54

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Cairncross et United States of Cross- 2 809 42 - 52 Self-reported Women unable to achieve Women who had ever Lifetime 12 or Infertility 24.7
al. (2020) America sectional infertility measure a clinical pregnancy for a attempted to conceive fertility
(Direct) period of > 12 months of medication
trying to conceive or who for >
used fertility medications for 1 month
> 1 month

SOUTH-EAST ASIA REGION

Zargar et al. India Cross- 10 063 15 - 44 Self-reported Couples who conceived Couples married for Period 12 Primary Infertility 15.07
(1997) sectional infertility measure more than one year after one year or more
(Direct) marriage or had not yet
conceived at the time of the
survey despite unprotected
sexual intercourse for more
than one year after marriage

Couples who had not yet Couples married for Period 12 Unresolved Primary 4.66
conceived at the time of the one year or more Infertility
survey despite unprotected
sexual intercourse for more
than one year after marriage

Unisa (1999) India Cross- Not reported 20 - 49 Constructed Women who have been Women who have been Period 36 Childlessness (3 or 5.0
sectional infertility measure married for three or more married for three or more years)
(Indirect) years without a live birth more years

Katole and India Cross- 570 15 - 49 Self-reported Women at risk of becoming All married women Period 24 Primary infertility 8.9
Saoji (2019) sectional infertility measure pregnant (not pregnant,
(Direct) sexually active, not using
contraception, and not
lactating) who report
trying unsuccessfully for a
pregnancy for 2 years or more

Udgiri and India Cross- 693 (Rural) 20 - 49 Self-reported Women at risk of becoming Women at risk of Period 24 Infertility
Patil (2019) sectional 419 (Urban) infertility measure pregnant who report becoming pregnant Rural 7.6
(Direct) trying unsuccessfully for Urban 8.8
a pregnancy for 2 years or
more Primary infertility
Rural 5.3
Urban 5.7

Secondary infertility
Rural 2.3
Urban 3.1

Executive summary Introduction Methods Results Discussion Annexes 55

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Singh and India Cross- 44 415 20 - 34 Self-reported Women who ever had a All women married for Lifetime 24 Infertility 10.7
Shukla (2015) sectional infertility measure problem getting pregnant at least two years Primary infertility 8.4
(Direct) despite cohabitation and
exposure to pregnancy for
two or more years

Women who never had a All women married for Period 24 Primary infertility 2.6
live birth up to the interview at least two years
date, and reported problems
conceiving for the first time
(failure to conceive despite
two years of cohabitation
and exposure to pregnancy)

21 583 Women who never had a Women married for at Period 24 Primary infertility 2.3
live birth up to the interview least two years and not
date, and reported problems using contraception
conceiving for the first time
(failure to conceive despite
two years of cohabitation
and exposure to pregnancy)

Purkayastha India Cross- 499 627 15 - 49 Constructed Currently married women Currently married Period 60 Primary infertility 1.79
(2020) sectional infertility measure who are married for five women
(Indirect) years or more, not currently
pregnant, never used
contraceptives, have no
terminated pregnancies, and
have zero children ever born

Chauhan et al. India Cross- 1 167 15+ Undetermined Women who failed to Ever married women Period 12 Secondary infertility 1.7
(2015) sectional conceive following a
previous pregnancy
or abortion despite
cohabitation and exposure
to pregnancy in absence
of contraception for one or
more years

Samarakoon Sri Lanka Cross- 2 000 Women: Undetermined Women who have never All married women Period 12 Primary infertility 4.05 (3.2 - 4.9)
et al. (2007) sectional 15 - 48 conceived in spite of
Men: 17 - 53 cohabitation and exposure
to pregnancy for a period of
12 months

Executive summary Introduction Methods Results Discussion Annexes 56

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

1 907 Women who have previously All married women who Period 24 Secondary infertility 16 (14.39 -
conceived but have have had a pregnancy 17.60)
been unable to conceive
subsequently despite
cohabitation and exposure
to pregnancy for a period
of 24 months (exposure to
pregnancy was from the end
of the period of lactation
amenorrhea for women
who breast fed the previous
infant).

EUROPEAN REGION

Bach et al. Denmark Cross- 1 372 Not reported Retrospective Nulliparous women with a Nulliparous women Period 12 Primary infertility 21
(2015) sectional Time-to- TTP greater than 12 months who planned or
Pregnancy (TTP) or infertility treatment prior partly planned their
Design to the studied pregnancy pregnancy and gave
birth to a singleton

Guldbrandsen Denmark Cohort 73 107 Not reported Retrospective Pregnant women with Pregnant women with a Period 6 Subfecundity 32.0
et al. (2014) Time-to- planned pregnancies and planned pregnancy 12 16.0
Pregnancy (TTP) a TTP greater than 6/12
Design months

Hollegaard et Denmark Cross- 2 927 18 + Retrospective Pregnant women with a TTP Pregnant women Period 12 Subfertility 17.3
al. (2007) sectional Time-to- greater than 12 months
Pregnancy (TTP)
Design

Kjaer Denmark Case- 247 18 - 49 Undetermined Women who failed to All women (controls) Lifetime 12 Infertility 12.0 (8.0 - 17.0)
Pedersen control conceive after trying for
et al (1994) more than 12 months
Translation

182 Women who failed to Women who had Lifetime 12 Infertility 17.0 (12.0 - 23.0)
conceive after trying for attempted and/or
more than 12 months achieved pregnancy
(controls)

Hærvig et al Denmark Cross- 2 140 50 - 51 Self-reported Men who ever tried to All men Lifetime 12 Infertility 17.9
(2018) sectional infertility measure achieve a pregnancy,
(Direct) without success during the
first 12 months

Executive summary Introduction Methods Results Discussion Annexes 57

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Kirkegaard et Denmark Cross- 9 507 Not reported Retrospective Women with spontaneous Women with Period 12 Subfertility 9.9
al. (2014) sectional Time-to- planned pregnancy with TTP spontaneous planned 24 3.1
Pregnancy (TTP) of 12/24 or more months pregnancies
Design

Raatikainen Finland Cross- 17 114 Not reported Retrospective Pregnant women who Pregnant women not Period 36 TTP >36 months 2.2
et al. (2010) sectional Time-to- reported a TTP > 36 months using contraception at
Pregnancy (TTP) the time of pregnancy
Design

Terävä et al. Finland Cross- 4 371 25 - 64 Self-reported Women who experienced a All women Lifetime 12 Subfertility 16.0
(2008) sectional infertility measure time period when they had
(Direct) tried to become pregnant,
but had not conceived or
conception took more than
12 months

Klemetti et al Finland Cross- Women: 1198 30 - 44 Self-reported Women/men who made All women/men Lifetime 12 Infertility - women 20.0
(2010) sectional Men: 1093 infertility measure unsuccessful attempts to Men 9.0
(Direct) conceive a child over a period
of 12 months or longer

Taponen et al. Finland Case- 60 31 Self-reported Women who reported that Women not using oral Lifetime No Infertility 10.0
(2004) control infertility measure Infertility has ever (earlier contraceptives or IUD duration
(Direct) or at this moment) been a devices (control group)
problem

Slama et al. France Cross- 69 18-44 Current Duration Couples not yet pregnant Women at risk of Period 12 TTP > 12 months 34.0 (15.0 - 54.0)
(2006) sectional Design after 12/24 months of pregnancy (18 to 44 24 TTP > 24 months 16.0 (4.0 - 29.0)
unprotected intercourse years at interview,
(estimated) declared not to be
pregnant, currently had
a male partner, had
had sexual intercourse
within the last 2
months, did not use any
birth control method
(including sterilization
of either partner), and
had not given birth to a
live- or stillborn baby in
the 3 months before the
interview)

Executive summary Introduction Methods Results Discussion Annexes 58

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

53 Estimated proportion of Women at risk of Period 6 TTP > 6 months 47.0 (28.0 - 66.0)
couples not yet pregnant pregnancy (as defined 12 TTP > 12 months 26.0(10.0 - 42.0)
after 6/12/24 months who above) and reported 24 TTP > 24 months 10.0 (2.0 - 18.0)
were trying to become they were currently
pregnant willing to become
pregnant or that
they had stopped
contraception to
become pregnant,
even if they declared
that they did not desire
to become pregnant
currently

Slama et al. France Cross- 867 (overall) 18-44 Current Duration Couples with no detected Women at risk of Period 6 Infertility 46.0 (36.0-35.0)
(2012) sectional 360 (primary) Design pregnancy during the pregnancy defined as Primary infertility 47.0 (26.0 - 68.0)
first 6/12/24 months of not using any birth
unprotected intercourse control method, had a 12 Infertility 24.0 (19.0 - 30.0)
(estimated) male partner and had Primary infertility 26.0 (15.0 - 36.0)
been sexually active in
the previous month. 24 Infertility 11.0 (8.0 - 14.0)
Primary infertility 11.0 (7.0 - 16.0)

Couples who declared that Women at risk of Period 6 Infertility 45.0 (34.0 - 55.0)
they had stopped using birth pregnancy defined as 12 23.0 (18.0 - 28.0)
control methods in order not using any birth 24 10.0 (8.0- 12.0)
to obtain a pregnancy and control method, had a
had no detected pregnancy male partner and had
conceived during the been sexually active
first 6/12/24 months of in the previous month
unprotected intercourse AND declared that they
(estimated) had stopped using
birth control methods
in order to obtain a
pregnancy

Executive summary Introduction Methods Results Discussion Annexes 59

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Couples who declared that Women at risk of Period 6 Infertility 43.0 (34.0 - 53.0)
they had stopped using birth pregnancy defined as 12 20.0 (16.0 - 25.0)
control methods in order not using any birth 24 8.0 (6.0 - 10.0)
to obtain a pregnancy and control method, had a
had no detected pregnancy male partner and had
conceived during the been sexually active
first 6/12/24 months of in the previous month
unprotected intercourse AND declared that they
(estimated) had stopped using
birth control methods
in order to obtain a
pregnancy AND who
did not use infertility
treatments over the
CDUI period

Eustache et France Cross- 390 Male partner: Retrospective Pregnant couples who took Pregnant couples Period 12 TTP > 12 months 5.0 (3.2–7.3)
al. (2004) sectional 20–45 Time-to- more than 12 months to whose male partner did
Pregnancy (TTP) conceive and whose male not provide a semen
Design partner did not provide a sample
semen sample

Muller et al. France Cross- Total: 894 Men: 20 - 45 Retrospective Pregnant women with a TTP Pregnant women not Period 12 TTP > 12 mo: Range 5.0 - 11.0
(2006) sectional Range: Time-to- greater than 12 months using contraception at
178 - 273 Pregnancy (TTP) the start of pregnancy
Design

Ajrouche et al France Case- 1 167 Not reported Self-reported Women who took more Women with a child Period 12 Difficulty becoming 18.0
(2014) control infertility measure than a year to conceive the (control group) pregnant
(Direct) index child and/or needed
to consult a doctor and/or
needed for the mother or
father to undergo fertility
treatment

Kuppers- Germany Cross- 1 216 25 - 45 Retrospective Women who have ever in Women who have Lifetime 12 Subfecundity 31.8 (29.4 - 34.6)
Chinnow and sectional Time-to- their lifetime experienced ever been at risk of
Karmaus Pregnancy (TTP) a time of unprotected becoming pregnant
(1997) Design intercourse (TUI) (with (i.e., had unprotected
Translation or without the onset of sexual intercourse)
pregnancy) of more than
12 months

Executive summary Introduction Methods Results Discussion Annexes 60

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Datta et al. United Kingdom Cross- Women: 16 - 74 Self-reported Women/men who ever had All women/men Lifetime 12 Infertility - women 12.5 (11.7–13.3)
(2016) sectional Unweighted: infertility measure a time, lasting 12 months who reported Men 10.1 (9.2–11.1)
8 066 (Direct) or longer, when they and having experience of
Weighted: their partner were trying for heterosexual sex
7 052 a pregnancy, but it didn’t
happen
Men:
Unweighted:
5 553
Weighted:
6 811

Joffe (2000) United Kingdom Cross- 1 540 16-59 Retrospective Individuals with time to All individuals whose Period 12 TTP >12 months 15.0
sectional Time-to- pregnancy greater than 12 first pregnancy was (Primary)
Pregnancy (TTP) months for first pregnancy a birth and was not
Design due to a contraceptive
failure

Gyorffy et al. Hungary Cross- 1 069 24+ Self-reported Participants whose time-to- All women (control Lifetime 12 TTP > 12 months 9.8
(2014) sectional infertility measure pregnancy had been longer group)
(Direct) than one year in case of any
of their pregnancies.

Van der Netherlands Cross- 243 25 - 40 Self-reported Male respondents who All men Period 12 Subfertility 8.6
Avoort et al. (Kingdom of sectional infertility measure reported a lack of conception
(2003) the) (Direct) after at least 12 months of
unprotected intercourse

137 Male respondents who Men at risk for fertility Period 12 Subfertility 15.3
reported a lack of conception problems
after at least 12 months of
unprotected intercourse

Hoenderboom Netherlands Cohort 2 377 16 - 39 Retrospective Women with an attempted Women who had ever Period 12 Primary infertility 16.7
et al. (2020) (Kingdom of Time-to- time to first pregnancy of > attempted to conceive (Time to first
the) Pregnancy (TTP) 12 months planned pregnancy)
Design

Sundby and Norway Cross- 4 034 40 - 42 Self-reported Women who tried to become All Women Lifetime 12 Infertility 10.3
Schei (1996) sectional infertility measure pregnant for more than a
(Direct) year without succeeding

Women who tried to become Married women not Lifetime 12 Permanent Infertility 2.6
pregnant for more than a using contraception,
year without succeeding postmenarchal and
and had never given birth to premenopausal
a child

Executive summary Introduction Methods Results Discussion Annexes 61

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Women who tried to become Married women not Lifetime 12 Subfertility 7.7
pregnant for more than a using contraception,
year without succeeding and postmenarchal and
had given birth to at least premenopausal
one child

Rostad et al. Norway Cross- 4 951 50 - 59 Self-reported Women who have ever tried All women Lifetime 12 Infertility 12.7
(2013) sectional infertility measure for more than a year to get
(Direct) pregnant regardless of any
subsequent birth

Nguyen et al. Norway Cross- 26 303 18 - 39 Retrospective Couples who planned Couples who planned Period 12 Infertility 12.0
(2007) sectional Time-to- their pregnancy and that their pregnancy
Pregnancy (TTP) took more than 12 months
Design to achieve pregnancy or
received infertility treatment

Magnus et al. Norway Cohort 64 064 27 - 62 Hybrid* Women with a planned Women who had a Period 12 Time to pregnancy 12.3
(2021) Primary pregnancy and a TTP > 12 planned pregnancy
Retrospective or reported use of assisted
Time-to- reproductive technologies
Pregnancy (TTP)
Design

Secondary:
Self-reported
infertility measure
(Direct)

Soares et Portugal Cross- 1 540 18+ Self-reported Women who reported ever All women Lifetime 12 Infertility 11.9 (10.4-13.7)
al. (2011) sectional infertility measure trying to get pregnant for
Translation (Direct) more than a year without
success

Philippov et Russian Cross- 2 000 18 - 45 Undetermined Women who had not Married women Lifetime 12 Infertility 16.7
al. (1998) Federation sectional conceived after 12 months Primary infertility 3.8
or more of unprotected Secondary infertility 12.9
intercourse

Bhattacharya United Kingdom Cross- 4 066 31-50 Self-reported Women who had Women whose fertility Lifetime 12 Infertility 17.5 (16.3–18.6)
et al. (2009) sectional (12-month) infertility measure unsuccessfully attempted had been tested Primary infertility 10.5 (9.5–11.4)
4 049 (Direct) conception for 12/24 months Secondary infertility 5.3 (4.7–6.0)
(24-month) or longer Experienced both 1.7 (1.3–2.0)

24 Infertility 9.1 (8.2–10.0)

Primary infertility 5.9 (5.2–6.6)
Secondary infertility 2.9 (2.4–3.4)
Experienced both 0.3 (0.2–0.5)

Executive summary Introduction Methods Results Discussion Annexes 62

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Women who had Women whose fertility Lifetime 12 Infertility 19.3 (18.1–20.5)
unsuccessfully attempted had been tested Primary infertility 9.8 (8.9–10.7)
conception for 12/24 months Secondary infertility 7.0 (6.2–7.8)
or longer and/or had Experienced both 2.5 (2.0–2.9)
sought medical help with
conception 24 Infertility 11.8 (10.8–12.8)
Primary infertility 5.7 (5.0–6.4)
Secondary infertility 5.2 (4.5–5.9)
Experienced both 0.9 (0.6–1.1)

Cabrera-Leon Spain Cross- 443 30 - 49 Retrospective Women who did not achieve All women Lifetime 12 Infertility
et al. (2015) sectional Time-to- pregnancy after having had Huelva City 17.79
Pregnancy (TTP) sexual intercourse with Spain 17.58
Design vaginal penetration and no (17.57–17.59)
contraception for one year Primary infertility
or more. Huelva City 6.14
Spain 6.12 (6.12–6.12)

Secondary infertility
Huelva City 11.64
Spain 11.33
(11.32–11.37)

Women with biological All women Lifetime Subfertility

children who spent more 6 Huelva City 21.00
than 6/12/24 months trying Spain 19.98
to become pregnant with (19.97–20.0)
any of their biological 12 Huelva City 11.62
children Spain 11.21
(11.2–11.22)
24 Huelva City 4.59
Spain 4.36 (4.35–4.37)

Self-reported Women who perceived All women Lifetime No Subjective Infertility:

infertility measure having or having had duration Huelva City 9.41
(Direct) difficulty in getting pregnant Spain 8.22 (8.21–8.23)

Akre et al. Sweden Cross- 401 653 20 + Retrospective Primiparous women who did Primiparous women Period 12 Primary subfertility 8.3
(1999) sectional Time-to- not become pregnant after
Pregnancy (TTP) more than one year
Design

Wulff et al. Sweden Cross- 534 25-44 Self-reported Women who experienced a All women Lifetime 12 Infertility ever 24.3
(1997) sectional infertility measure period of infertility (inability
(Direct) to conceive within 12 months
of unprotected intercourse)
at some point in life

Executive summary Introduction Methods Results Discussion Annexes 63

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Couples who were unable All women Period 12 Primary infertility 6.2
to get pregnant (again) after Secondary infertility 2.8
having tried for 12 months

Hallen (2011) Sweden Cross- 201 18 - 55 Self-reported Men reporting a period of 1 All men (control group) Period 12 Involuntary 7.0 (3.4–10.5)
sectional infertility measure or more years of involuntary childlessness
(Direct) childlessness during the last
five years

Björvang et al Sweden Cross- 818 Not reported Retrospective Women with planned Pregnant women with Period 12 Infertility 9.7
(2020) sectional Time-to- pregnancies who had a TTP planned pregnancies
Pregnancy (TTP) greater than 12 months
Design

Brunetti et Switzerland Cross- 216 29 Retrospective Women exposed to Women exposed to Lifetime 24 Primary infertility 2.8
al. (1994) sectional Time-to- fertilisation (i.e. women fertilisation before (unresolved)
Translation Pregnancy (TTP) wishing to have children, their 28th birthday
Design with a stable partner and (i.e. women wishing
a regular sex life) for 24 or to have children, with
more months who have a stable partner, a
never conceived by the age regular sex life, and
of 29 presence of 24 months
of unprotectedness)

212 Women aged 29 who waited Women who have Lifetime 12 Unintentionally 10.0
at least once for more than conceived delayed motherhood
a year before obtaining a (hypofertility;
conception resolved subfertility)

Gokler et al. Türkiye Cross- 570 18 - 49 Self-reported Women who have the All married women Period 12 Infertility 12.8
(2014) sectional infertility measure inability to become pregnant Primary (% of total) 38.4
(Direct) despite regular sexual Secondary (% of 61.6
intercourse during the last total)
year

Sarac and Koc Türkiye Cross- 5 947 15 - 49 Constructed Women who have been Women who have been Period 60 Primary infertility 1.8
(2018) sectional infertility measure married for at least five married for at least five
(Indirect) years, have not used any years
birth control methods during
that time, and have not given
birth

Executive summary Introduction Methods Results Discussion Annexes 64

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

6 835 Women who have been Women who have been Period 12 Infertility 8.1
married for at least one married for at least one
year, have not used any year
contraception during the
last year and who have not
become pregnant in the last
year

5 860 18 - 44 Women who were at risk Women who are at Period 12 Infertility 8.6
of pregnancy in the first 12 risk of exposure to
months of the total 5-year pregnancy
period

Albayrak and Türkiye Cross- 2 400 15 - 49 Undetermined Women who had never been Married women Period 12 Primary infertility 6.3
Günay (2007) sectional able to conceive, although (Childless women)
they had been married at
least 12 months, were living
with their husband and had
a desire for a baby

Gunnell and United Kingdom Cross- 2 377 36 - 50 Retrospective Women who failed to All women Lifetime 12 Infertility 26.4 (24.6 - 28.2)
Ewings (1994) sectional Time-to- become pregnant after Primary infertility 16.1 (14.6 - 17.6)
Pregnancy (TTP) 12/24 months of regular Secondary infertility 15.8 (14.3 - 17.3)
Design unprotected intercourse
24 Infertility 12.9
Primary infertility 7.4
Secondary infertility 6.6

Women with primary All women Lifetime 12 Involuntary 3.0

unresolved infertility Childlessness
and women who became
pregnant but remained
involuntarily childless

Buckett United Kingdom Cross- 728 45 - 55 Retrospective Women who tried to All women Lifetime 12 Infertility 17.3 (14.6 - 20.0)
and Bentick sectional Time-to- conceive for more than Primary infertility 10.6 (8.4-12.8)
(1997) Pregnancy (TTP) 12/24 months Secondary infertility 6.7 (4.9 - 8.5)
Design
24 Infertility 12 (9.6 - 14.4)

Women with primary or All women Lifetime 12 Unresolved 4.3 (2.8 - 5.8)
secondary infertility who involuntary
never conceived infertility

Oakley et al. United Kingdom Cross- 7 702 18-55 Self-reported Women who reported having Women who had Lifetime No Self-reported 19.5 (18.6-20.4)
(2010) sectional infertility measure problems getting pregnant become pregnant duration infertility
(Direct) or ever tried to get
pregnant

Executive summary Introduction Methods Results Discussion Annexes 65

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Women with at least Women who had Lifetime 12 TTP > 12 months 16.0 (15.1-16.9)
1 planned pregnancy with a become pregnant
TTP > 12 months or ever tried to get
pregnant and reported
the pregnancy was
planned

6 584 40 - 55 Women with no pregnancy Women who had Lifetime No Primary unresolved 2.4 (2.0-2.8)
despite trying become pregnant duration infertility
or ever tried to get
pregnant

Women with no live birth Women who had Lifetime No Primary unresolved 4.3 (3.8-4.8)
despite trying become pregnant duration infertility
or ever tried to get
pregnant

Bolumar et al. Denmark, Cross- 3 187 25 - 44 Retrospective Women who had planned Women who had Period 12 Primary infertility 12.0
(1997) Germany, Italy, sectional Time-to- their pregnancies and stopped using
Poland, and Pregnancy (TTP) reported a TTP greater than birth control and
Spain Design 12 months for their first had planned their
pregnancy pregnancy

Karmaus and Denmark, Cross- 932 25 - 44 Retrospective Women with a time of Time of unprotected Period 12 Primary
Juul (1999) Germany, Italy, sectional Time-to- unprotected intercourse intercourse (TUI) for Subfecundity:
Poland, and Pregnancy (TTP) (TUI) greater than 12/24 first TUI with a starting Total 23.4
Spain Design months for first TUI with date less than 5 years Range 14.8 - 33.3
a starting date less than 5 before the interview, 24
years before the interview among women at risk Total 17.6
Range 10.1 - 26.8

Women planning their Time of unprotected Period Primary

pregnancies with a time of intercourse (TUI) for Subfecundity:
unprotected intercourse first TUI with a starting 12 Total 18.7
(TUI) greater than 12/24 date less than 5 years Range 12.1 - 30.2
months for first TUI with before the interview,
a starting date less than 5 among women at 24 Total 13.6
years before the interview risk and planning Range 8.1 - 24.1
pregnancy

Toft et al. Greenland, Cross- Warsaw: 376 18+ Retrospective Married women with a TTP Married women not Period 12 TTP > 12 months
(2005) Poland and sectional Kharkiv: 307 (Greenland) Time-to- greater than 12 months using contraception Warsaw 19.0
Ukraine Greenland: Pregnancy (TTP) Kharkiv 27.0
520 Not reported Design Greenland 15.0
for other sites

Executive summary Introduction Methods Results Discussion Annexes 66

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Jensen et al. Denmark, Cross- Range: Male partner: Retrospective Pregnant couples who took Pregnant couples Period 6 TTP > 6/12 mo: 16.9 - 20.9
(2001) Finland, France, sectional 191 - 302 20–45 Time-to- more than 6/12 months to whose male partner 12 Range 7.5 - 10.1
and United Pregnancy (TTP) conceive and whose male provided a semen
Kingdom Design partner provided a semen sample
sample

EASTERN MEDITERRANEAN REGION

Hassan (1997) Egypt Cross- 20 002 < 50 years Constructed Women who have been Married women Period 12 Infertility 12.0
sectional infertility measure exposed to or at risk of Primary infertility 4.3
(Indirect) pregnancy for successive Secondary infertility 7.7
12 months or more without
conceiving

Kazemijaliseh Iran (Islamic Cross- 1067 18 - 57 Retrospective Women failing to achieve Married women willing Lifetime 12 Primary infertility 17.3
et al. (2015) Republic of) sectional Time-to- a clinical pregnancy after to become pregnant
Pregnancy (TTP) 12 months or more of
Design regular unprotected sexual
intercourse

Nasrabad et Iran (Islamic Cross- 90 141 15 - 49 Constructed Couples of reproductive Ever-married women Period 12 Primary infertility 2.3
al. (2013) Republic of) sectional infertility measure age who are having sexual
(Indirect) intercourse without
contraception and are
unable to establish a
pregnancy within one year

Sexually active women who Ever-married women Period 60 Primary infertility 2.6
are not using a contraception
but are unable to have a live
birth for five or more years

Esmaeilzadeh Iran (Islamic Cross- 1 081 20 - 45 Self-reported Women who have Women who attempted Lifetime 12 Infertility 15.5 (13.5 - 17.5)
et al. (2012) Republic of) sectional infertility measure experienced a delay conception Primary infertility 12.2
(Direct) in conception for least Secondary infertility 1.9
12 months of unprotected Experienced both 1.5
intercourse at some time in
their life

Women who were currently Women who attempted Period 12 Primary Infertility 4.3 (2.3 - 6.3)
experiencing a delay conception
in conception for least
12 months of unprotected
intercourse and had not
previously given birth to
a child

Executive summary Introduction Methods Results Discussion Annexes 67

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Mirzaei et al. Iran (Islamic Cross- 2 611 20 - 49 Self-reported Women who have failed to All married, divorced, or Period 12 Infertility 5.2(4.3 - 6.1)
(2018) Republic of) sectional infertility measure achieve clinical pregnancy widowed women Primary infertility 2.68 (2.4 - 3.8)
(Direct) after 12 months or more Secondary infertility 2.15 (1.89 - 3.4)
unprotected coitus

Safarinejad Iran (Islamic Cross- 11 441 15-50 Undetermined Women who did not Women who ever Lifetime 24 Infertility 8.0 (3.2 - 15.0)
(2008) Republic of) sectional conceive despite cohabitated for at least Primary infertility 4.6 (3.6 - 5.2)
cohabitation and exposure two years Secondary infertility 3.4 (2.4 - 5.1)
to pregnancy for two years

Vahidi et al. Iran (Islamic Cross- 10 662 19-49 Constructed Ever-married women Ever-married women Lifetime 12 Primary infertility 24.9 (23.5-26.2)
(2009) Republic of) sectional infertility measure who have experienced of
(Indirect) infertility (no pregnancy)
despite one year of
unprotected intercourse

10 873 Women who meet the Ever-married women Period 12 Primary infertility 3.4 (3.0-3.8.0)
definition of lifetime primary
infertility and have not
conceived up to the study
time

Ahmadi Asr Iran (Islamic Cross- 3 183 Not reported Retrospective Women with no conception Women married for at Lifetime 12 Infertility 3.27
Badr et al. Republic of) sectional Time-to- during marriage after at least one year Primary infertility 2.04
(2006) Pregnancy (TTP) least 12 months' period of Secondary infertility 1.23
Design intercourse without using
contraception

2 623 15-49 Women with no conception Women ages 15 - 49 Lifetime 12 Infertility 3.35
during marriage after at years married for at Primary infertility 2.05
least 12 months' period of least one year Secondary infertility 1.30
intercourse without using
contraception

Akhondi et al. Iran (Islamic Cross- 17 178 20-40 Constructed Women who are sexually Married women Lifetime 12 Primary infertility 20.2 (SE = 0.2)
(2019) Republic of) sectional infertility measure active, do not use any 24 12.5
(Indirect) contraception, and do 36 10.3
not have a live birth after 48 9.6
12/24/36/48/60 months 60 9.20

Dovom et al. Iran (Islamic Cross- 888 18 - 49 Retrospective Women who have not Women who had Period 12 Primary Infertility 6.4 (4.8 - 8.0)
(2014) Republic of) sectional Time-to- become pregnant after at attempted to get
Pregnancy (TTP) least 1 year of unprotected pregnant for at least
Design intercourse one year, not use a
contraceptive method,
and have regular
unprotected sexual
intercourse

Executive summary Introduction Methods Results Discussion Annexes 68

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

681 Women who failed to achieve Women who had Lifetime 12 Secondary Infertility 7.8 (6.0 - 9.6)
a second clinical pregnancy attempted to get
after 12 months or more of pregnant for the
regular unprotected sexual second time for at least
intercourse whether or not one year, not use a
having the second child contraceptive method,
and have regular
unprotected sexual
intercourse

888 Women with any delay of Women married for at Lifetime 12 Infertility 21.2 (18.4 - 23.8)
more than one year to get least one year who have
pregnant during their life ever had a willingness
regardless of whether or not for pregnancy
they have a child now

Hosseini et al. Iran (Islamic Cross- 2 296 18 - 49 Self-reported Couples with lack of Married women at risk Period 12 Primary infertility 3.2
(2012) Republic of) sectional infertility measure pregnancy after one year of of pregnancy Secondary infertility 1.7
(Direct) continuous unprotected sex
during the ovulation point of
menstrual cycle

Sharif et al. Iran (Islamic Cross- 1 469 18 - 45 Self-reported Couples who had not All women of Period 12 Infertility 15.24 (14.7-15.4)
(2020) Republic of) sectional infertility measure achieved pregnancy in the childbearing age and
(Direct) past 12 months in union

WESTERN PACIFIC REGION

Herbert et al. Australia Cohort 13 715 45 - 50 Self-reported Women who have tried Women born between Lifetime 12 Infertility 11.0
(2009a) infertility measure unsuccessfully to get 1946 - 1951
(Direct) pregnant for 12 months or
more, have been diagnosed
as infertile by a doctor (self
or partner), and/or had
treatment for infertility in
lifetime (self or partner)

Herbert et al. Australia Cohort 1 031 28 - 33 Self-reported Women who had tried to Women who had tried Lifetime 12 Infertility 17.3
(2009b) infertility measure conceive for 12 or more to conceive or had been
(Direct) months unsuccessfully pregnant

Mena et al. Australia Cohort 6 130 Time 1: Self-reported Women who reported Women who ever tried Period 12 Infertility 15.4 (14.5 - 16.4)
(2020) 22 - 27 infertility measure ever having problems to get pregnant (Cumulative
Final time: (Direct) with infertility (tried incidence)
37 - 42 unsuccessfully to get
pregnant for 12 months
or more) with a current or
previous partner

Executive summary Introduction Methods Results Discussion Annexes 69

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Damone et al. Australia Cross- 8 612 28 - 33 Self-reported Couples who ever had All women in union Lifetime 12 Infertility 11.1
(2019) sectional infertility measure problems with fertility (tried (currently or previously)
(Direct) unsuccessfully for 12 months
or more to get pregnant)

Zhou et al. China Cross- 17 275 20 - 49 Retrospective Women who wanted Women exposed to the Period 12 Infertility 15.5
(2018) sectional Time-to- to become pregnant in risk of pregnancy (not Primary infertility 9.5
Pregnancy (TTP) the previous year, who using contraception Secondary infertility 6
Design had unprotected sexual and had not lived
intercourse at least once a separated longer than
month, and who were trying three months)
to achieve pregnancy longer
than 12 months

10 742 Women who wanted Women attempting to Period 12 Infertility 25

to become pregnant in become pregnant (not Primary infertility 15.3
the previous year, who using contraception, Secondary infertility 9.7
had unprotected sexual had not lived separated
intercourse at least once a longer than three
month, and who were trying months, and willing to
to achieve pregnancy longer become pregnant)
than 12 months

Xingping et China Cross- 5 325 844 < 49 Undetermined Women who failed to achieve Married women of Period 12 Infertility 1.57
al. (2006) sectional pregnancy after one year of childbearing age Primary (% of total) 82.4
Translation regular sexual intercourse Secondary (% of total) 17.6
without contraception

Zhang et al. China Cross- 12 342 27 - 57 Self-reported Married couples who All married couples Period 12 Infertility 4.2
(2014) sectional infertility measure failed to achieve a clinical who had regular Primary infertility 3.1
(Direct) pregnancy after 12 months of unprotected Secondary infertility 1.1
regular unprotected sexual intercourse for at least
intercourse 12 months prior to the
date of interview

Yang et China Cross- 5 631 20 - 49 Undetermined Couples who had a desire All couples Period 24 Infertility 1.72
al. (2011) sectional to have children, had Primary (% of total) 58.76
Translation normal cohabitation for Secondary (% of 41.24
two years, had regular total)
sexual intercourse, and
were not pregnant without
contraception

Wu et al. China Cross- 274 Reproductive Undetermined Married women who did not Married women Period 24 Infertility 1.1
(2004) sectional age use contraception within two
Translation years after marriage and did
not become pregnant

Executive summary Introduction Methods Results Discussion Annexes 70

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Cai et al. China Cross- 1 835 20 - 49 Self-reported Married women with an Married women Period 12 Infertility 15.2
(2011) sectional infertility measure absence of pregnancy after Primary infertility 7.5
Translation (Direct) 1 year of normal sexual Secondary infertility 7.7
intercourse without the use
of contraception under the
conditions of exposure to
pregnancy

Zhang and China Cross- 2 187 20-49 Undetermined Couples who had a desire Married women willing Period 12 Primary Infertility 13.08
Zhang (2013) sectional to have children, normal to have a child and not Secondary infertility 35.25
Translation sexual activity for more than using contraception
one year and have not used during a specified one-
contraception, but still fail to year period
conceive

Song (2013) China Cross- 3 110 37 - 38 Retrospective Women with no child after All married women Period 24 Sterility 14.24
sectional Time-to- 24/84 months of marriage 84 1.67
Pregnancy (TTP)
Design

Wang et al. China Cohort 700 20 - 40 Prospective Time- Couples in a committed Couples in a committed Period 12 Infertility 28.0
(2018) to-Pregnancy relationship and planning to relationship planning to
(TTP) Design conceive with a TTP greater conceive
than 12 months

Yang et al. China Cross- 7 025 18 - 49 Retrospective Couples whose waiting Couples who had been Period 12 Infertility 11.4
(2017) sectional Time-to- time to pregnancy was 12 or married for more than
Pregnancy (TTP) more months for the first 12 months
Design pregnancy or couples who
had never been pregnant
and time trying to conceive
was 12 or more months

Huang and China Cross- 18 893 Women: Undetermined Married couples who Couples married in Period 12 Infertility 13.3
Tang (2013) sectional 18 - 49 had regular sexual 2007 who did not use
Translation intercourse and did not take contraceptive measures
contraceptive measures, within one year after
but had not conceived after marriage
cohabitation for more than
12 months

Meng et al. China Cohort 1 627 Not reported Prospective Time- Newly married couples Newly married couples Period 12 Infertility 13.6 (11.9 - 15.3)
(2015) (12-month) to-Pregnancy who failed to achieve a exposed to the risk of Primary infertility 14.0 (12.2 - 15.8)
936 (TTP) Design clinical pregnancy after pregnancy Secondary infertility 11.2 (7.2 - 15.2)
(24-month) 12/24 months or more of
regular unprotected sexual 24 Infertility 8.5 (6.7 - 10.3)
intercourse Primary infertility 8.7 (6.7 - 10.7)
Secondary infertility 7.9 (3.6 - 12.2)

Executive summary Introduction Methods Results Discussion Annexes 71

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Hu et al. China Cohort 820 24-46 Prospective Time- Couples who engaged Couples who engaged Period 12 Infertility 26.2
(2020) to-Pregnancy in regular unprotected in regular unprotected
(TTP) Design intercourse and had a TTP intercourse and
greater than 12 months achieved pregnancy
during follow-up

He et al. China Cross- 12 364 18 - 49 Undetermined Women who failed to All married and Lifetime 12 Infertility 10.11
(2020) sectional achieve a clinical pregnancy cohabitating women
after one year or more
of unprotected sexual
intercourse, despite having a
desire to get pregnant

2 486 Women who failed to Married and Lifetime 12 Infertility 20.92

achieve a clinical pregnancy cohabitating women
after one year or more not using contraception
of unprotected sexual
intercourse, despite having a
desire to get pregnant

Chen et al. China Cross- 6 906 > 21 Retrospective Couples with TTP greater Married couples who Period 12 Primary Infertility 11.97
(2015) sectional Time-to- than or equal to12 months ever tried for pregnancy
Pregnancy (TTP) or being unable to conceive
Design after trying for at least
12 months for their first
pregnancy

Righarts et al. New Zealand Cross- 974 25 - 50 Hybrid* Women who ever tried to All women who had Lifetime 12 Infertility 21.7 (19.1 - 24.2)
(2015) sectional conceive for 12/24 months ever conceived or had 24 12.8 (10.7 - 15.2)
Primary: or more tried to conceive
Self-reported
infertility measure
(Direct)

Secondary:
Constructed
infertility measure
(Indirect)

Women who ever tried to All women who had Lifetime 12 Infertility 25.3 (22.6 - 28.1)
conceive for 12 months or ever conceived or had
more and/or sought help to tried to conceive
conceive

476 Women ages 40 years or Women ages 40 years Lifetime 12 Primary unresolved 1.9 (0.9 - 3.6)
more with no previous or more who had ever infertility
births, whose infertility was tried to become or had
not resolved with a live birth been pregnant.

Executive summary Introduction Methods Results Discussion Annexes 72

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

van Roode et New Zealand Cohort Men: 386 32 and 38 Self-reported Men/women who, with a Men/women who ever Lifetime 12 Infertility - men 17.9 (14.2 - 22.1)
al. (2015) Women: 396 infertility measure partner, had ever tried for reported or attempted women 25.0 (20.8 - 29.6)
(Direct) 12 months or more to get pregnancy
pregnant without success

Men/women who, with a Men/women who ever Lifetime 12 Infertility - men 21.8 (17.7-26.2)
partner, had ever tried for reported or attempted women 26.0 (21.8-30.6)
12 months or more to get pregnancy
pregnant without success OR
who sought medical help to
get pregnant

Not reported Men/women who, with a All men/women Lifetime 12 Infertility - men 18.2 (14.8 - 22.1)
partner, had ever tried for women 22.5 (18.7 - 26.6)
12 months or more to get
pregnant without success OR
who sought medical help to
get pregnant

Righarts et al. New Zealand Cross- Men: 3 744 16 - 74 Self-reported Men/women who ever had All men/women who Lifetime 12 Infertility- men 8.2 (7.1-9.4)
(2021) sectional Women: infertility measure a time, lasting 12 months had heterosexual women 12.5 (11.3-13.8)
5 222 (Direct) or longer, when they or intercourse
a partner were trying for
a pregnancy but it didn't
happen

Fertility- Men/women who ever had Fertility-tested women Lifetime 12 Infertility 15.4 (14.0-16.9)
tested a time, lasting 12 months (ever conceived or
women: or longer, when they or tried unsuccessfully to
3 792 a partner were trying for conceive for 12 months
a pregnancy but it didn't or longer)
happen

Men: 3 744 Men/women who ever had All men/women who Lifetime 12 and/ Infertility- men 11.4 (10.1-12.8)
Women: a time, lasting 12 months or had heterosexual or no women 13.4 (12.2-14.7)
5 222 longer, when they or a partner intercourse duration
were trying for a pregnancy (self-
but it didn’t happen OR they perceived)
believed they or their partner
were infertile.

Fertility- Men/women who ever had Fertility-tested women Lifetime 12 and/ Infertility 16.3 (14.8-17.8)
tested a time, lasting 12 months or (ever conceived or or no
women: longer, when they or a partner tried unsuccessfully to duration
3 792 were trying for a pregnancy conceive for 12 months (self-
but it didn't happen OR they or longer) perceived)
believed they or their partner
were infertile.

Executive summary Introduction Methods Results Discussion Annexes 73

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Kreisel et al. Palau Cross- 315 >17 Self-reported Women who have tried Women who reported Lifetime 12 Infertility 39.7 (34.2 - 45.3)
(2020) sectional infertility measure unsuccessfully to become ever trying to become
(Direct) pregnant for 12 or more pregnant
months

Passey et al Papua New Cross- 201 15 - 45 Self-reported Women who reported that All women Period 24 Infertility 29.6
(1998) Guinea sectional infertility measure they wanted more children,
(Direct) were trying to conceive,
and had had unprotected
intercourse for 2 or more
years

MULTIPLE REGIONS

Rutstein and Multiple Cross- Less 25 - 49 Constructed Women who have been Ever-married women Period 60 Infertility 25.7
Shah (2004) sectional Developed infertility measure married for the past five Range 16.0 - 30.0
Countries (Indirect) years, who ever had sexual
(China intercourse, who did not use Primary infertility 2.5
excluded): contraception during the Range 1.5 - 2.8
939 796.7 past five years, and who did
not have any births in the Secondary infertility 23.8
past five years Range 13.6 - 28.2

Keiding et al. Benin, Nigeria, Cross- Range: 18 - 44 Current Duration Nulliparous women not Women at risk of Period 12 Primary Infertility: 24.0 - 85.0
(2021) Senegal, United sectional 211 - 1 183 Design yet pregnant by 12 months conception at the time Range
Republic of (estimated) of the survey (18 and 44
Tanzania, *Initiation of attempt years, only one partner,
Indonesia, estimated by date of currently married or
Philippines, cohabitation with partner living with a partner,
Dominican had sex in the last 4
Republic, and weeks, never had a live
Colombia birth, menstruating and
not currently pregnant,
not menopausal,
had not had a
hysterectomy, and was
not contracepting at the
time of interview)

Mascarenhas Global Cross- Range: 20-44, Constructed Women who desire a child Women in both infertile Period 60 Primary infertility:
et al. (2012a) sectional 337 - 62 785 > 30 infertility measure and have been in a union and fertile unions, World 1.9 (1.8 - 2.2)
(countries (Indirect) for at least five years, during where women in a Range 0.8 - 4.0
with data on which they have not used fertile union have
women in any contraceptives, and who successfully had at least
union only) have not had a live birth one live birth and have
been in the union for at
least five years at the
time of the survey

Executive summary Introduction Methods Results Discussion Annexes 74

Infertility prevalence estimates, 1990–2021

AUTHORS GEOGRAPHIC TYPE OF ANALYTIC AGE RANGE METHODOLOGIC NUMERATOR DENOMINATOR PERIOD OR DURATION RATIO MEASURED ESTIMATE %
(YEAR) LOCATION STUDY SAMPLE APPROACH LIFETIME (MONTHS) (95% CI)
SIZE MEASURE

Women who desire a child All women (calculated Period 60 Primary infertility:
and have been in a union as the product of the World 1.5 (0.3 - 1.7)
for at least five years, during prevalence of infertility Range 0.5 - 3.0
which they have not used among child-seeking
any contraceptives, and who women and the
have not had a live birth proportion who are
exposed to the risk of
pregnancy)

China Women who desire a child Women in both infertile Period 60 Secondary infertility:
excluded and have been in a union and fertile unions, World 10.2 (9.2 - 11.4)
for at least five years since where women in a Range 3.8 - 22.2
their last birth, during which fertile union have
they have not used any successfully had at least
contraceptives, and who one live birth and, at
have not had another live the time of the survey,
birth have been in the union
for at least five years
following their first birth

Women who desire a child All women (calculated Period 60 Secondary infertility:
and have been in a union as the product of the World 2.9 (2.6 - 3.2)
for at least five years since prevalence of infertility Range 0.8 - 10.5
their last birth, during which among child-seeking
they have not used any women and the
contraceptives, and who proportion who are
have not had another live exposed to the risk of
birth. pregnancy)

Mascarenhas 26 countries Cross- Range: 20 - 49 Constructed Women that have been in a Women in union for Period 60 Primary: Range 0.8 - 3.4
et al. (2012b) sectional 532 - 71 095 infertility measure union for at least five years at least five years Secondary: Range 8.7 - 32.6
(primary) (Indirect) (secondary infertility: since (secondary infertility:
the partner’s last live birth) following their first live
Range: without a live birth, during birth) at the time of the
190 - 22 740 which neither partner used survey
(secondary) contraception, and where
the female partner expresses
a desire for a(nother) child.

Taylor et al. United Kingdom Cross- Melbourne: Not reported Retrospective Women who failed to Pregnant women Period 12 Subfecundity
(1999) and Australia sectional 929 Time-to- conceive current pregnancy Melbourne 20.2
Manchester: Pregnancy (TTP) within 12 months of Manchester 14.5
960 Design unprotected intercourse

* Hybrid approach includes studies that combined two approaches to generate a single infertility estimate

Executive summary Introduction Methods Results Discussion Annexes 75

Annex 4. Pooled lifetime and period
infertility prevalence estimates, by
methodological approach
Pooled lifetime infertility prevalence Pooled period infertility prevalence estimates,
estimates, by methodological approach by methodological approach
invlogit(ES) Weight invlogit(ES) Weight
Study with 95% CI (%) Study with 95% CI (%)
Retrospective TTP Prospective TTP
Ahmadi Asr Badr (2006) 0.03 [0.03, 0.04] 2.55 Hu (2020) 0.26 [0.23, 0.29] 1.99
Brunetti (1994) 0.10 [0.07, 0.15] 2.27 Meng (2015) 0.14 [0.12, 0.15] 2.00
Buckett (1997) 0.17 [0.15, 0.20] 2.56 Wang (2018) 0.28 [0.25, 0.31] 1.99
Cabrera-Leon (2015) 0.18 [0.18, 0.18] 2.63 Heterogeneity: τ2 = 0.24, I2 = 97.49%, H2 = 39.82 0.22 [0.14, 0.33]
Test of θi = θj: Q(2) = 84.10, p = 0.00
Dovom (2014) 0.21 [0.19, 0.24] 2.58
Gunnell (1994) 0.26 [0.25, 0.28] 2.62 Retrospective TTP
Jacobson (2018) 0.35 [0.32, 0.38] 2.60 Bello (2010) 0.32 [0.28, 0.36] 1.98
Kuppers-Chinnow (1997) 0.32 [0.29, 0.34] 2.60 Björvang (2020) 0.10 [0.08, 0.12] 1.96
Somé (2016) 0.10 [0.08, 0.14] 2.46 Eustache (2004) 0.05 [0.03, 0.07] 1.81
Heterogeneity: τ2 = 0.72, I2 = 99.59%, H2 = 242.99 0.17 [0.10, 0.26] Guldbrandsen (2014) 0.16 [0.16, 0.16] 2.02
Test of θi = θj: Q(8) = 803.31, p = 0.00 Hoenderboom (2020) 0.17 [0.15, 0.18] 2.01
Hollegaard (2007) 0.17 [0.16, 0.19] 2.01
Self-report Jensen (2001) 0.10 [0.07, 0.15] 1.82
Anyalechi (2019) 0.14 [0.12, 0.15] 2.60 Jensen (2001) 0.08 [0.05, 0.12] 1.73
Bhattacharya (2009) 0.19 [0.18, 0.21] 2.62 Jensen (2001) 0.10 [0.06, 0.15] 1.77
Cairncross (2020) 0.25 [0.23, 0.26] 2.62 Jensen (2001) 0.08 [0.05, 0.12] 1.82
Kirkegaard (2014) 0.10 [0.09, 0.11] 2.02
Crawford (2015) 0.25 [0.22, 0.29] 2.58
Magnus (2021) 0.12 [0.12, 0.13] 2.02
Crawford (2015) 0.06 [0.05, 0.07] 2.58 Muller (2006) 0.11 [0.08, 0.15] 1.85
Crawford (2015) 0.10 [0.08, 0.12] 2.56 Muller (2006) 0.09 [0.06, 0.13] 1.82
Datta (2016) 0.13 [0.12, 0.13] 2.62 Muller (2006) 0.05 [0.03, 0.09] 1.57
Esmaeilzadeh (2012) 0.16 [0.14, 0.18] 2.59 Muller (2006) 0.08 [0.05, 0.13] 1.70
Feuntes (1994) 0.26 [0.19, 0.34] 2.34 Nguyen (2007) 0.12 [0.12, 0.12] 2.02
Geelhoed (2002) 0.16 [0.14, 0.18] 2.58 Taylor (1999) 0.20 [0.18, 0.23] 1.99
Gleason (2020) 0.24 [0.22, 0.26] 2.61 Taylor (1999) 0.14 [0.12, 0.17] 1.98
Gyorffy (2014) 0.10 [0.08, 0.12] 2.55 Toft (2005) 0.15 [0.12, 0.18] 1.95
Herbert (2009b) 0.17 [0.15, 0.20] 2.58 Toft (2005) 0.19 [0.15, 0.23] 1.94
Hærvig (2018) 0.18 [0.16, 0.20] 2.61 Toft (2005) 0.27 [0.22, 0.32] 1.94
Klemetti (2010) 0.20 [0.18, 0.22] 2.59 Yang (2017) 0.11 [0.11, 0.12] 2.02
Zhou (2018) 0.25 [0.24, 0.26] 2.02
Kreisel (2020) 0.40 [0.34, 0.45] 2.53
Heterogeneity: τ2 = 0.28, I2 = 99.54%, H2 = 218.89 0.13 [0.11, 0.16]
McQuillan (2003) 0.35 [0.31, 0.39] 2.57 Test of θi = θj: Q(23) = 1942.07, p = 0.00
Nelson (2011) 0.20 [0.16, 0.25] 2.47
Oakley (2010) 0.16 [0.15, 0.17] 2.62 Current duration
Righarts (2015) 0.25 [0.23, 0.28] 2.59 Polis (2017) 0.31 [0.28, 0.35] 1.99
Righarts (2021) 0.15 [0.14, 0.17] 2.61 Slama (2006) 0.34 [0.18, 0.55] 1.36
Rostad (2013) 0.13 [0.12, 0.14] 2.62 Slama (2012) 0.24 [0.19, 0.30] 1.91
Soares (2011) 0.12 [0.10, 0.14] 2.58 Thoma (2013) 0.16 [0.08, 0.27] 1.52
Sundby (1996) 0.10 [0.09, 0.11] 2.61 Heterogeneity: τ2 = 0.08, I2 = 68.93%, H2 = 3.22 0.26 [0.20, 0.34]
Terävä (2008) 0.16 [0.15, 0.17] 2.62 Test of θi = θj: Q(3) = 9.32, p = 0.03
Wulff (1997) 0.24 [0.21, 0.28] 2.55
van Roode (2015) 0.26 [0.22, 0.31] 2.53 Self-report
Ajrouche (2014) 0.18 [0.16, 0.20] 2.00
Heterogeneity: τ2 = 0.26, I2 = 98.59%, H2 = 70.78 0.18 [0.15, 0.21]
Cai (2011) 0.15 [0.14, 0.17] 2.00
Test of θi = θj: Q(26) = 1177.91, p = 0.00 Feuntes (1994) 0.10 [0.09, 0.11] 2.01
Gokler (2014) 0.13 [0.10, 0.16] 1.95
Undetermined Hallen (2011) 0.07 [0.04, 0.11] 1.70
He (2020) 0.21 [0.20, 0.22] 2.63 Mena (2020) 0.15 [0.14, 0.16] 2.02
Khaer Pedersen (1994) 0.17 [0.12, 0.23] 2.35 Mirzaei (2018) 0.05 [0.04, 0.06] 1.98
Philippov (1998) 0.17 [0.15, 0.18] 2.61 Sharif (2020) 0.15 [0.15, 0.16] 2.02
Heterogeneity: τ2 = 0.03, I2 = 87.10%, H2 = 7.75 0.19 [0.16, 0.22] Sundby (1998) 0.09 [0.09, 0.11] 2.00
Test of θi = θj: Q(2) = 19.89, p = 0.00 Van der Avoort (2003) 0.15 [0.10, 0.22] 1.78
Walraven (2001) 0.10 [0.08, 0.12] 1.98
Overall 0.18 [0.15, 0.20] Zhang (2014) 0.04 [0.04, 0.05] 2.01
Heterogeneity: τ2 = 0.33, I2 = 99.48%, H2 = 193.20 Heterogeneity: τ2 = 0.26, I2 = 99.00%, H2 = 99.92 0.11 [0.08, 0.14]
Test of θi = θj: Q(11) = 1166.76, p = 0.00
Test of θi = θj: Q(38) = 2117.16, p = 0.00
Test of group differences: Qb(2) = 0.25, p = 0.88 Constructed
0.02 0.05 0.12 0.27 0.50 Balakrishnan (1993) 0.07 [0.06, 0.07] 2.02
Random-effects REML model Bushnik (2012) 0.16 [0.14, 0.17] 2.00
Dulberg (1993) 0.09 [0.07, 0.10] 1.98
Fledderjohann (2016) 0.17 [0.15, 0.19] 2.00
Hassan (1997) 0.12 [0.12, 0.12] 2.02
Sarac (2018) 0.09 [0.08, 0.09] 2.01
Heterogeneity: τ2 = 0.17, I2 = 98.78%, H2 = 81.75 0.11 [0.08, 0.15]
Test of θi = θj: Q(5) = 327.25, p = 0.00

Undetermined
Huang (2013) 0.13 [0.13, 0.14] 2.02
Kouman (2005) 0.10 [0.08, 0.13] 1.94
Xingping (2006) 0.02 [0.02, 0.13] 2.02
Heterogeneity: τ2 = 1.49, I2 = 99.97%, H2 = 3269.47 0.06 [0.02, 0.21]
Test of θi = θj: Q(2) = 10955.53, p = 0.00

Overall 0.13 [0.11, 0.15]

Heterogeneity: τ2 = 0.40, I2 = 99.81%, H2 = 521.01
Test of θi = θj: Q(51) = 152792.99, p = 0.00
Test of group differences: Qb(5) = 31.16, p = 0.00
0.02 0.05 0.12 0.27 0.50
Random-effects REML model

Executive summary Introduction Methods Results Discussion Annexes 76

Annex 5. Certainty
of estimates

Table A5.1 Certainty rating for 12-month infertility estimates (lifetime)

№ of Quality assessment Effect Certainty

studies
Study design Risk of Inconsistency Indirectness Imprecision Other Rate
bias considerations (95% CI)

Infertility (12 months cross section; assessed with: Prevalence per 100 patients)

37 Observational Not Seriousb Not seriousc Not seriousd Publication Prevalence ⨁⨁⨁◯
seriousa bias: not 18 (15 to MODERATE
detected 20 per
100 persons)

a
77.4% of studies were at overall low risk of bias. 21.1% were at moderate risk of bias. Only 1.5% were at high risk of bias.
b
Considerable heterogeneity was observed. This was based on visual inspection of point estimates and 95% CI reported on the forest plots.
Individual study estimates ranged from 2% to 32%. None of the hypothesized subgroup analyses (geographic region or measurement approach)
explained the observed heterogeneity.
c
All individual studies addressed a question that was very similar, if not the same, as the review question in these estimates.
d
Although a clear threshold for judging imprecision was not set, the width of the confidence interval was judged as sufficiently narrow. For the
given assessment, the confidence interval was judged as sufficiently narrow and may not lead to completely different decisions. That is, whether
the upper or lower bound of the 95% CI represents the truth, the reactions to the estimates will probably be the same.

Table A5.2 Certainty rating for pooled 12-month infertility estimates (period prevalence)

№ of Quality assessment Effect Certainty

studies
Study design Risk of Inconsistency Indirectness Imprecision Other Rate
bias considerations (95% CI)

Infertility (12 months cross section; assessed with: Prevalence per 100 patients)

43 Observational Not Seriousb Not seriousc Not seriousd Publication prevalence ⨁⨁⨁◯
seriousa bias: not 13 (11 to MODERATE
detected 15 per
100 persons)

a
77.4% of studies were at overall low risk of bias. 21.1% were at moderate risk of bias. Only 1.5% were at high risk of bias.
b
Considerable heterogeneity was observed. This was based on visual inspection of point estimates and 95% CI reported on the forest plots.
Individual study estimates ranged from 2% to 32%. None of the hypothesized subgroup analyses (geographic region or measurement approach)
explained the observed heterogeneity.
c
All individual studies addressed a question that was very similar, if not the same, as the review question addressed in this review.
d
Although a clear threshold for judging imprecision was not set, the width of the confidence interval was judged as being sufficiently narrow.
For the given assessment, the confidence interval was judged as sufficiently narrow and may not lead to completely different decisions. That is,
whether the upper or lower bound of the 95% CI represents the truth, the reactions to the estimates will likely be the same.

Executive summary Introduction Methods Results Discussion Annexes 77

Infertility prevalence estimates, 1990–2021

Table A5.3 Certainty rating for primary 12-month infertility estimates (lifetime)

№ of Quality assessment Effect Certainty

studies
Study design Risk of Inconsistency Indirectness Imprecision Other Rate
bias considerations (95% CI)

Infertility (12 months cross section; assessed with: Prevalence per 100 patients)

12 Observational Not Seriousb Not seriousc Not seriousd Publication Prevalence ⨁⨁⨁◯
seriousa bias: not 10 (6 to MODERATE
detected 14 per
100 persons)

a
77.4% of studies were at overall low risk of bias. 21.1% were at moderate risk of bias. Only 1.5% were at high risk of bias.
b
Considerable heterogeneity was observed. This was based on visual inspection of point estimates and 95% CI reported on the forest plots.
Individual study estimates ranged from 2% to 25%. None of the hypothesized subgroup analyses (geographic region or measurement approach)
explained the observed heterogeneity.
c
All individual studies addressed a question that was very similar, if not the same, as the review question addressed in these estimates.
d
Although a clear threshold for judging imprecision, the width of the confidence interval was judged as sufficiently narrow. For the given
assessment, the confidence interval was judged as sufficiently narrow and may not lead to completely different decisions. That is, whether the
upper or lower bound of the 95% CI represents the truth, the reactions to the estimates will likely be the same.

Table A5.4 Certainty rating for primary 12-month infertility estimates (period prevalence)

№ of Quality assessment Effect Certainty

studies
Study design Risk of Inconsistency Indirectness Imprecision Other Rate
bias considerations (95% CI)

Infertility (12 months cross section; assessed with: Prevalence per 100 patients)

33 Observational Not Seriousb Not seriousc Not seriousd None Prevalence ⨁⨁⨁◯
seriousa 9 (7 to 12 per MODERATE
100 persons)

a
77.4% of studies were at overall low risk of bias. 21.1% were at moderate risk of bias. Only 1.5% were at high risk of bias.
b
Considerable heterogeneity was observed. This was based on visual inspection of point estimates and 95% CI reported on the forest plots.
Individual study estimates ranged from 1% to 38%. None of the hypothesized subgroup analyses (geographic region or measurement approach)
explained the observed heterogeneity.
c
All individual studies addressed a question that was very similar, if not the same, as the review question addressed in these estimates.
d
Although a clear threshold for judging imprecision was not set, the confidence interval was judged as sufficiently narrow. For the given
assessment, the confidence interval was judged as sufficiently narrow and may not lead to completely different decisions. That is, whether the
upper or lower bound of the 95% CI represents the truth, the reactions to the estimates will probably be the same.

Table A5.5 Certainty rating for secondary 12-month infertility estimates (lifetime)

№ of Quality assessment Effect Certainty

studies
Study design Risk of Inconsistency Indirectness Imprecision Other Rate
bias considerations (95% CI)

Infertility (12 months cross section; assessed with: Incidence per 100 patients)

10 Observational Not Seriousb Not seriousc Not seriousd None Prevalence ⨁⨁⨁◯
seriousa 7 (4 to 11 per MODERATE
100 persons)

a
77.4% of studies were at overall low risk of bias. 21.1% were at moderate risk of bias. Only 1.5% were at high risk of bias.
b
Considerable heterogeneity was observed. This was based on visual inspection of point estimates and 95% CI reported on the forest plots.
Individual study estimates ranged from 1% to 16%. None of the hypothesized subgroup analyses (geographic region or measurement approach)
explained the observed heterogeneity.
c
All individual studies addressed a question that was very similar, if not the same, as the review question addressed in these estimates.
d
Although a set any clear threshold for judging imprecision was not set, the width of the confidence interval was judged as sufficiently narrow.
For the given assessment, the confidence interval was judged as sufficiently narrow and may not lead to completely different decisions. That is,
whether the upper or lower bound of the 95% CI represents the truth, the reactions to the estimates will probably be the same.

Executive summary Introduction Methods Results Discussion Annexes 78

Annex 5. Certainty of estimates

Table A5.6 Certainty rating for secondary 12-month infertility estimates (period prevalence)

№ of Quality assessment Effect Certainty

studies
Study design Risk of Inconsistency Indirectness Imprecision Other Rate
bias considerations (95% CI)

Infertility (12 months cross section; assessed with: Prevalence per 100 patients)

17 Observational Not Seriousb Not seriousc Not seriousd None Prevalence ⨁⨁⨁◯
seriousa 5 (3 to 9 per MODERATE
100 persons)

a
77.4% of studies were at overall low risk of bias. 21.1% were at moderate risk of bias. Only 1.5% were at high risk of bias.
b
We observed considerable heterogeneity was observed. This was based on visual inspection of point estimates and 95% CI reported on the forest plots.
Individual study estimates ranged from 0% to 35%. None of the hypothesized subgroup analyses (geographic region or measurement approach)
explained the observed heterogeneity.
c
All individual studies addressed a question that was very similar, if not the same, as the review question addressed in these estimates
d
Although a clear threshold for judging imprecision was not set, the width of the confidence interval was judged as sufficiently being narrow.
For the given assessment, the confidence interval was judged as sufficiently narrow and may not lead to completely different decisions. That is,
whether the upper or lower bound of the 95% CI represents the truth, the reactions to the estimates will probably be the same.

Executive summary Introduction Methods Results Discussion Annexes 79

For further information please contact:

Department of Sexual and Reproductive

Health and Research
World Health Organization

Avenue Appia 20
CH-1211, Geneva 27
Switzerland

email: [email protected]
https://www.who.int/health-topics/infertility