PLOS ONE

RESEARCH ARTICLE

The incidence of sore throat and group A

streptococcal pharyngitis in children at high
risk of developing acute rheumatic fever: A
systematic review and meta-analysis
Sarah Pearce1, Asha C. Bowen1,2,3,4,5, Mark E. Engel6, Maya de la Lande3, Dylan
D. Barth ID1,2*
a1111111111 1 Faculty of Health and Medical Sciences, University of Western Australia, Perth, Western Australia,
a1111111111 Australia, 2 Wesfarmers Centre for Vaccines and Infectious Diseases, Telethon Kids Institute, Nedlands,
a1111111111 Western Australia, Australia, 3 Department of Infectious Diseases, Perth Children’s Hospital, Nedlands,
a1111111111 Western Australia, Australia, 4 Menzies School of Health Research, Charles Darwin University, Darwin,
Northern Territory, Australia, 5 Institute for Health Research, University of Notre Dame, Fremantle, Western
a1111111111
Australia, Australia, 6 Department of Medicine, University of Cape Town, Cape Town, Western Cape, South
Africa

* [email protected]

OPEN ACCESS

Citation: Pearce S, Bowen AC, Engel ME, de la Abstract

Lande M, Barth DD (2020) The incidence of sore
throat and group A streptococcal pharyngitis in
children at high risk of developing acute rheumatic
fever: A systematic review and meta-analysis.
Background
PLoS ONE 15(11): e0242107. https://doi.org/ Group A streptococcal (GAS) pharyngitis has traditionally been considered the sole precur-
10.1371/journal.pone.0242107
sor of acute rheumatic fever (ARF). Evidence from Australia, however, suggests that GAS
Editor: Thomas Proft, University of Auckland, NEW skin infections may contribute to the pathogenesis of ARF. A missing piece of evidence is
ZEALAND
the incidence of sore throat and GAS pharyngitis in this setting. We conducted a systematic
Received: August 31, 2020 review and meta-analysis of the incidence of sore throat and GAS pharyngitis in all children
Accepted: October 26, 2020 at risk of developing ARF.
Published: November 18, 2020
Methods
Copyright: © 2020 Pearce et al. This is an open
access article distributed under the terms of the Databases were systematically searched for studies reporting on the incidence of pharyngi-
Creative Commons Attribution License, which tis among children from low to upper-middle income countries, and Indigenous children liv-
permits unrestricted use, distribution, and
ing in high-income countries. Studies were subjected to data extraction by two independent
reproduction in any medium, provided the original
author and source are credited. reviewers. Following an assessment of the methodological quality of the studies, we
extracted incidence rates (IRs) and conducted a meta-analysis. This systematic review is
Data Availability Statement: All relevant data are
within the manuscript and its Supporting registered on PROSPERO (CRD42019113019).
information files.

Funding: The authors received no specific funding Results

for this review. SP received a scholarship from the
From 607 titles identified by the search, 11 articles met the predetermined inclusion criteria;
National Health and Medical Research Council
(NHMRC) funded through the HOT North initiative. ten studies reported IRs while for the remaining study, the incidence was calculated. The
AB is funded by a fellowship from the NHMRC. pooled incidence estimated for sore throat was 82.5 per 100 child-years (95% confidence
Competing interests: The authors have declared interval [CI], 6.5 to 1044.4 per 100 child-years, I2 = 100%) and GAS pharyngitis was 10.8
that no competing interest exists. per 100 child-years (95% CI, 2.3 to 50.0 per 100 child-years, I2 = 99.9%).

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 1 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

Conclusions
The pooled IRs for sore throat in children at risk of developing ARF were higher than rates
reported in developed nations (32.70–40 per 100 child-years) and similar for GAS pharyngi-
tis (12.8–14 per 100 years). The limited Australian data lend support to the need for further
studies to inform the role of GAS pharyngitis in the development of ARF in Australian Indige-
nous children, so as to inform local primary prevention strategies for ARF and Rheumatic
Heart Disease (RHD).

Introduction
Acute pharyngitis is a common presentation to general practitioners [1]. Most cases of sore
throat are caused by viruses; however, 25–30% of these infections are caused by the Group A
Streptococcal (GAS) bacteria [2]. While sore throat is a minor illness, 0.3–3% of patients with
GAS pharyngitis may develop serious complications, [2] such as acute rheumatic fever (ARF)
and its sequela, rheumatic heart disease (RHD) [3]. ARF and RHD are a significant cause of
morbidity and mortality in developing countries and amongst Indigenous populations living
in developed countries [4]. In Australia, the incidence of ARF in Indigenous and Torres Strait
Islander people is >300/1000 person-years, 60-fold greater than in non-Indigenous Austra-
lians (<5/1000 person-years).[5]
Traditionally, GAS pharyngitis is thought to be the sole precipitant of ARF; thus ARF pri-
mary prevention strategies have focused on treating GAS pharyngitis [6]. In Australian Indige-
nous communities, where ARF is endemic, GAS pharyngitis is seemingly rare [7], while the
burden of GAS impetigo is amongst the highest reported in the world [8]. As such, impetigo
has been proposed as an additional driving force behind ARF in these communities. If impe-
tigo is established to be a ‘bona fide’ precipitant of ARF, there would be major implications for
primary prevention strategies, which are currently directed primarily towards the treatment of
GAS pharyngitis [2]. The shift towards including GAS impetigo, (a condition often normal-
ized or neglected by health care providers as a benign condition [9]) in RHD prevention strat-
egies has already begun with an emphasis on healthy skin in the third edition of the 2020
guidelines for the prevention and management of RHD in Australia (https://www.
rhdaustralia.org.au).
The evidence for the skin-ARF theory is based mostly on epidemiological studies and
observations from Northern Australia [6]. Impetigo is highly prevalent in this region, affecting
a median of 45% of Indigenous children at any point in time [8]. Molecular epidemiological
studies also show that the pharyngeal strains usually responsible for ARF (rheumatogenic
strains) are absent, and there is significant overlap between the typically distinct skin and
throat strains [6]. The notion that sore throat is uncommon is largely anecdotal and informed
by reports from Indigenous Health Workers, nurses, and medical officers working in Northern
Australia. Furthermore, there is a paucity of published data on the frequency of Indigenous
children being treated for GAS pharyngitis in remote regions. Only two published studies that
report the incidence of sore throat in this setting, and they yield conflicting results [7, 10].
Given that the skin-ARF hypothesis hinges on the idea that sore throat is uncommon, the lack
of evidence is surprising and must be addressed. A large, longitudinal, prospective cohort
study of GAS pharyngitis and impetigo with episodes of ARF as the primary endpoint would
be ideal, but time, cost and remoteness of the large, sparsely populated region preclude this.

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 2 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

To address this question from a different angle, we have systematically compiled the avail-
able evidence on the incidence of sore throat and GAS pharyngitis in other populations where
ARF is endemic, to compare and contrast with known incidence for Australian Indigenous
children. Incidence has been chosen, rather than prevalence, as it provides data on the number
and frequency of new episodes of pharyngitis, and over time, the increasing risk of ARF from
recurrent episodes [11, 12].

Methods
This systematic review and meta-analysis was conducted in accordance with the Preferred
Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) guidelines (S1 Check-
list) [13] and is registered on PROSPERO (International Prospective Register of systematic
reviews, registration number CRD42019113019, https://www.crd.york.ac.uk/PROSPERO).

Review question
This review asks the following questions. What is the incidence of sore throat and GAS phar-
yngitis in children at high risk of ARF? Is there variation in incidence with respect to geogra-
phy, climate, health care setting or home setting?

Search strategy and data sources

We reviewed similar systematic reviews reporting on the epidemiology of pharyngitis and skin
infections in children from low income countries or Indigenous children living in high-
income areas [8, 14], in order to construct the search terms and avoid overlap with available
evidence. Following this, the search strategy was developed to capture all relevant studies,
using Medical Subject Headings (MeSH) combined with key words (Table 1). PubMed, Ovid
Medline, and Embase databases were searched, then supplemented with grey literature from
searches of the Australian Institute of Health and Welfare (AIHW), Australian Indigenous
HealthInfoNet, Informit, Agency for Healthcare Research and Quality (AHRQ), and The Grey
Literature Report databases. To identify as many studies as possible, no filters were applied to
limit age, language, publication date or study design. Results were also supplemented by
searching the references of the selected papers and their citations through Web of Science. An
update on the search was conducted in PubMed on 11th July 2020.

Table 1. Search strategy.

Incidence [MeSH term] OR burden OR epidemiology OR rate
AND
Pharyngitis [MeSH term] “sore throat” OR pharyngitis OR GAS OR Group A Streptococc� OR streptococc� OR
streptococc� pyogenes OR streptococc� pyogenes pharyngitis OR group A streptococcal pharyngitis OR tonsillitis
OR tonsillopharyngitis OR streptococcus pyogenes [MeSH Terms]
AND
Indigenous OR aborigin� OR torres strait islander OR oceanic ancestry group OR maori OR pacific islander OR
native American OR American Indian OR inuit OR minority group OR first nation OR native people OR
developing countries OR developing nation OR low income countries OR low income nations OR low-middle
income countries OR low-middle income nations OR middle income countries OR middle income nations OR less
developed countries OR less developed nations OR third world countries OR third world nations OR poverty OR
socioeconomic inequality OR living standards OR resource limited OR low resource

MeSH, Medical subject heading

https://doi.org/10.1371/journal.pone.0242107.t001

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 3 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

Study selection
Inclusion criteria
1. Papers reporting the incidence of pharyngitis and/or studies with a follow-up period from
which incidence could be calculated
2. The population under investigation was children from low- to middle-income countries or
Indigenous children living in high-income countries
3. GAS diagnosed based on culture and/or a rapid antigen detection test
Exclusion criteria
1. Prevalence studies of pharyngitis
2. Molecular typing studies
3. Any studies lacking primary data including narrative reviews, opinion pieces, or letters
4. Duplicate publications of the same data; where this occurred the most recent data were
used

Case definitions
The method of case ascertainment for all studies involved a clinical assessment, and a pharyn-
geal swab, and microbiological culture for GAS bacteria. Symptomatic sore throat was defined
simply as a participant complaining of throat pain [15]. GAS pharyngitis was defined as a child
complaining of a sore throat, with confirmation of GAS detection in the pharynx. GAS detec-
tion was either culture confirmation of Streptococcus pyogenes from a throat swab collected
and processed in the laboratory or detection of the presence of the S. pyogenes antigen using a
rapid antigen detection test. Whilst culture confirmation is the gold standard, RADT kits are
used in many jurisdictions for prompt detection of S. pyogenes at the point of clinical care and
have a reported sensitivity of between 80 and 95% [16].

Study records
Selection process. All included titles were compiled, and duplicates removed. Two
reviewers (SP & MD) independently screened the references for eligibility using a numbered
coding system to record the decision for each article. At each step of the process reviewers
would meet and compare their decisions; discrepancies were discussed and where necessary a
third reviewer (DB) resolved any disagreements. The titles and abstracts were screened for rel-
evance, and potential full-text articles downloaded. The full-text articles were reviewed in
depth and the final articles were then selected for inclusion.
Data extraction. The two reviewers independently extracted data from the included
papers and entered them into a data extraction form using three Excel pro forma. The
extracted data were categorised into general information, study characteristics, and incidence
calculations. General information included the study ID, authors, year of publication, aim of
study, start date, end date, and season of participation. The study characteristics included
region, country, climate, income, study setting, geographical setting (rural/urban/peri-urban),
population description, inclusion criteria, exclusion criteria, site of recruitment, selection
methods, age range, ethnicity, pharyngitis definition, and method of ascertainment. Climate
was determined by the Köppen-Geiger Climate Classification system [17] and country income

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 4 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

by the World Bank income classification (at the time the study was conducted). To calculate
the IR for sore throat and GAS pharyngitis, we extracted data on the population denominator
(N), number of new cases (n), incidence density per 100 child-years, and the confidence inter-
vals (CI). Where sample populations included both symptomatic and asymptomatic pharyngi-
tis, we only extracted information from symptomatic participants [18]. Incidence was defined
as the number of new cases (n) over the study period for each child (N) as per the formula
below [19].
Number of new cases during periodðnÞ
Incidence per 100 child‐years ¼ � 365 � 100
ðNumber of participants � days of observationÞðNÞ

The included studies used a variety of methods to report incidence. As such, using the avail-
able data, we recalculated all reported incidences using this formula to adjust for studies where
incidence was not reported or where different units were used, e.g. 1000 child-weeks [7, 10].
Quality assessment. Limited tools are available to assess the methodological quality of
observational studies [20]. A tool developed for prevalence studies by Hoy et al. [21], and
adapted by Werfalli et al. [22], was modified for our prevalence review. Question nine—‘Was
the length of the shortest prevalence period for the parameter of interest appropriate?’ was
changed to ‘Was the follow-up period for the parameter of interest appropriate?’ The two inde-
pendent reviewers used the tool to score the studies as high risk (score<6), moderate risk (6–
8) and low risk (>8) of bias.
Statistical analysis. We calculated pooled incidence estimates for both sore throat and
GAS pharyngitis from studies with available incidence or where we could calculate incidence
using the formula above. Meta-analysis models were applied by using the log IRs of sore throat
and GAS pharyngitis and corresponding standard errors (metan routine: metan logirr solo-
girr), using Stata1 version 16. The random effects model was used, and heterogeneity was
quantified using the I2 statistic. Sensitivity analysis was performed to explain the potential
sources of heterogeneity. The analysis included plotting the pooled IR of studies by sample
size. Subgroup analyses were performed to compare pooled estimates by country, geographical
setting, study setting, climate, and income status according the World Bank Index at the time
the studies were conducted. We were unable to conduct a funnel plot analysis to examine for
publication bias given too few studies included in our mete-analysis. Where meta-analysis was
not possible due to a lack of data or differences in the methods, a narrative review of the results
is provided.

Results
Literature search
Between October 2018 and October 2019, 607 titles were identified by the search strategy (Fig
1). Duplicate removal rendered 424 titles for assessment by two reviewers (SP and MD); a
third reviewer (DB) adjudicated where needed. Twenty-one studies were deemed potentially
eligible, from which 11 were included after a second review by DB. Ten studies directly
reported the IRs [7, 10, 15, 18, 23–28] and we calculated the incidence from a single study [29].
Excluded studies are reported in S1 Table.

Study characteristics
Data from more than 16,814 participants over 46-years were included from the eleven studies
(Table 2) [7, 10, 15, 18, 23–29]. Studies were conducted in eight countries, predominantly
from Asia or Oceania, in a range of income settings. The climate ranged from arid to tropical,

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 5 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

Fig 1. PRISMA flowchart.

https://doi.org/10.1371/journal.pone.0242107.g001

with a mixture of rural and urban households. The studies were conducted in either a school
or health care clinic setting. A summary of the IRs, country income level and climate of
included studies are documented in Table 3.
Study designs were predominantly prospective cohort studies which followed children,
either through school, household or clinic visits (Table 2) [7, 10, 15, 18, 23–29]. A single study
used a multiplier model based on surveillance of sore throat and GAS pharyngitis through 36
pediatric outpatient clinics to calculate incidence [23]. Most studies used convenience sam-
pling except Lennon 2000 [26], Nandi 2001 [25], and Kumar 2009 [29], which each used
quasi-randomisation. Most studies swabbed all children [7, 10, 18, 23, 26, 27, 29] while others
only swabbed those who were symptomatic [15, 24, 25, 28]. Two studies collected venous
blood samples for anti-streptolysin O titres (ASOT) and used an alternative incidence calcula-
tion: incidence = ST/OB/3.5 x 365 x 100 (ST = sore throats, OB = number of observations, 3.5
average duration of sore throat in days) and did not report enough data to allow recalculation
[7, 10]. El Kholy 1972 defined GAS pharyngitis as a child who had a positive GAS pharyngeal
culture two weeks prior to a rise in the ASOT, because there were no clinical cases of pharyngi-
tis [18].

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 6 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

Table 2. The characteristics of included studies.

Study ID Country Age Sample Size Study Design Site of Sampling Inclusion Ethnicity Urban/ Method of
(children) recruitment Method criteria Rural Ascertainment
El-Kholy Egypt 6–12 156 Prospective School Convenience Any child Egyptian Rural Weekly clinical
1972 [18] Cohort Study attending the assessment, monthly
primary school GAS pharyngeal
on a regular swabs, and 6-monthly
basis, present on ASOT. No clinical
the days of observations of
assessment. pharyngitis made so
monthly ASOT were
implemented.
Nicolle Canada 4–14 Rankin Inlet: Prospective School Convenience Children Inuit, Rural 3x weekly visits and
1990 [24] 309 Cohort Study regularly Caucasian & presentations to
St. Therese attending Native nursing station:
Point: 449 primary school Indian clinical assessment;
and children swab + culture if
presenting to the symptomatic
nursing station
with pharyngitis
or impetigo
Lennon New 5–15 Not specified Prospective School Convenience Children at one Not specified Urban Daily swabs by lay
2000 [26] Zealand cohort study of the 25 high but includes workers (mostly
risk schools with Maori/ parents). Any child
consent from Pacific with a positive culture
parents. Islander was referred for
clinical assessment
and treatment.
Nandi 2001 India 5–15 536 Prospective Household Quasi- School aged Indian Periurban Fortnightly clinical
[25] Cohort Study randomisation children living in assessment; swab
Chandigarh + culture if
chosen at symptomatic
random.
McDonald Australia 0–15 1173 adults Prospective Households Convenience Any person Indigenous Rural Monthly clinical
2006 [7] + children Cohort Study with known living in a Australian assessment, swab,
(number of history of household with a culture + ASOT
children not RHD known history of
specified) ARF and/or
RHD.
McDonald Australia 0–15 91 Prospective Households Convenience Any person Indigenous Rural Monthly clinical
2007 [10] Cohort Study with known living in a Australia assessment, swab,
history of household with a culture + ASOT
RHD known history of
ARF and/or
RHD.
Steer 2009 Fiji 7–12 685 Prospective School Convenience Children Indigenous 2 rural, 2 Fortnightly clinical
[15] Urban = 426 Cohort Study between 5–14 Fijian Indo- urban assessment; swab
years old located Fijian Other + culture if
Rural = 241 in one of four symptomatic
schools chosen
for the study.
Kumar India 5–15 4249 Prospective School Quasi- Children aged Indian Rural Fortnightly clinical
2009 [29] Cohort Study randomisation 5–15 attending assessment, swab
school, with + culture
parental consent
Engel 2012 South 5–15 950 Prospective Clinic Convenience Children aged Black Periurban Daily clinical
[28] Africa study 5–15 years African assessments, swabs
presenting to the Mixed and culture
clinic with ancestry
pharyngitis
(Continued )

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 7 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

Table 2. (Continued)

Study ID Country Age Sample Size Study Design Site of Sampling Inclusion Ethnicity Urban/ Method of
(children) recruitment Method criteria Rural Ascertainment
Kumar India 7–11 241 Prospective School Convenience Children Indian Rural Fortnightly clinical
2012 [27] Cohort Study attending village assessment, swab
primary schools + culture
Wu 2016 China 0–14 Not specified Multiplier Paediatric Convenience Children aged Chinese Urban Clinical assessment,
[23] model based outpatient 0–14 years swab + culture.
on paediatric clinics diagnosed with Multiplier model
outpatient pharyngitis or used to calculate
clinic scarlet fever. incidence.

ASOT, Anti-Streptolysin A Titre; NS, Not Stated; ARF, Acute Rheumatic Fever; RHD, Rheumatic Heart Disease

https://doi.org/10.1371/journal.pone.0242107.t002

Meta-analysis
Incidence of sore throat and GAS pharyngitis. The pooled IRs in children aged 5–15
years for sore throats was 82.5 per 100 child-years (95% CI, 6.5 to 1044.4; 4 Studies, n = 6,420;
I2 = 100%) [15, 25, 28, 29] and GAS pharyngitis was 10.6 per 100 child-years (95% CI, 2.3 to
50.0; 6 studies, n = 6,661, I2 = 99.9%) [15, 25–29] (Fig 2).
Subgroup analyses. We further conducted analyses in each of the sore throat and GAS
pharyngitis subgroups, with respect to various parameters. No significant difference in sore
throat incidence was observed according to climate: temperate IR = 51.6 per 100 child-years
(95% CI, 1.1–2529.7; 3 studies, n = 1710) [25, 28, 29] vs tropical IR = 166.4 per 100 child-years
(95% CI, 135.5–204.3; 1 study, n = 685) [15]. Studies conducted in schools had a sore throat IR
of 186.6 per 100 child-years (95% CI, 141.8 to 245.6; 2 studies, n = 1019) [15, 29] compared
with 705.0 per 100 child-years (95% CI, 679.4 to 731.6; 1 study, n = 536) and 0.8 per 100 child-
years (95% CI, 0.8 to 0.9; 1 study, n = 840) in a single study reporting household [25] and clinic

Table 3. Summary of incidence of sore throat and GAS pharyngitis in children at risk of acute rheumatic fever.
Study ID Pharyngitis Incidence (per 100 child- GAS Pharyngitis Incidence (per 100 child- Country, Region Climate Income
years) years)
El-Kholy 1972 [18] Not done 31.2 Egypt, Africa Arid Low-middle
Nicolle 1990 [24] Not done Rankin Inlet: 4.9 St Therese Point: 9.4 Canada, North Cold High
America
Lennon 2000 [26] Not done 50 New Zealand Temperate High
Nandi 2001 [25] 705 95 India, Asia Temperate Low
McDonald 2006 [7] Adults + children: 19 Adults + Children: 4 Australia, Oceania Tropical High
Children: 8 Children: 0
McDonald 2007 Adults + children: 480 Adults + Children: 32 Australia, Oceania Arid High
[10] Children: 457 Children: Not done
Steer 2009 [15] 162.9 14.7 Fiji, Oceania Tropical Upper-
middle
Kumar 2009 [29] 232 5.4 India, Asia Temperate Low
Engel 2012 [28] 0.84 0.18 South Africa Temperate Upper-
middle
Kumar 2012 [27] Not done 16.6 India, Asia Temperate Low
Wu 2016 [23] 29.8 2.68 China, Asia Cold Upper-
middle

GAS, group A streptococcus

https://doi.org/10.1371/journal.pone.0242107.t003

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 8 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

Fig 2. Pooled incidence rates of GAS pharyngitis and sore throat in children at risk of acute rheumatic fever.
https://doi.org/10.1371/journal.pone.0242107.g002

[28] settings, respectively (Fig 3a). Studies conducted in rural settings had a significantly
higher pooled IR of sore throat 208.5 per 100 child-years (95% CI, 167.1 to 260.3; 2 studies,
n = 613) [15, 29] compared to those conducted in urban settings of 150.2 per 100 child-years
(95% CI, 136.2 to 165.71; 1 study, n = 399) per 100 child-years [15] (P<0.05). The pooled IR in
peri-urban settings were found to be 24.3 per 100 child-years (95% CI, 0.03 to 17888.1; 2 stud-
ies, n = 1376) [25, 28] (Fig 3b).
For GAS pharyngitis, schools had a pooled IR of 15.9 per 100 child-years (95% CI, 6.7 to
37.7; 4 studies, 13,260 participants) [15, 26, 27, 29]. This IR was significantly higher than that
in the single study conducted in a clinic setting (IR = 0.2 per 100 child-years (95% CI, 0.2 to
0.2; 1 study, n = 840)) [28] and significantly lower than the IR in studies conducted in in
household settings (IR = 95.0 per 100 child-years (95% CI, 85.9 to 105.1; 1 study, n = 536) [25]
(Fig 3c).
By country, New Zealand had the highest pooled IR of 50 per 100 child-years (95% CI, 48.8
to 51.3; 1 study, n = 12 000) [26] followed by India, 20.6 (95% CI, 3.8 to 112.8; 3 studies,
n = 1111) [25, 27, 29] and Fiji, 14.7 (95% CI, 11.4 to 18.8; 1 study, n = 685) [15]. South Africa
had a significantly lower incidence compared with the other regions with an IR of 0.2 per 100
child-years (95% CI, 0.2 to 0.2; 1 study, n = 840) [28] (P< 0.05).

Narrative summary of studies not included in the meta-analysis

Additional studies, though not amenable to meta-analysis, provide additional data on the inci-
dence of sore throat [7, 10, 18, 23, 24, 26, 27] and GAS pharyngitis [7, 10, 18, 23, 24]. The Aus-
tralian and Canadian studies recruited Indigenous children from high-income countries, the
Chinese study, children from the upper-middle income bracket [30, 31], while the Egyptian
study represents a low-to-middle income country. The studies were prospective cohort studies
conducted amongst households [7, 10] and schools [24]. The remaining study used a multi-
plier model to estimate the incidence based on presentation to pediatric outpatient clinics
[23]. (Table 2).
The incidence of sore throat ranged from 8–457 per 100 child-years (3 studies, n = >1264)
[7, 10, 23]. The highest incidence observed was in arid, Central Australia, whereas the lowest
incidence was found in tropical, Northern Australia. The study from China fell in-between

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 9 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

Fig 3. Subgroup analysis of pooled incidence rates: 3a) Sore throat IRs by study setting, 3b) sore throat IRs by rurality, and 3c) GAS
pharyngitis by study setting. ES, effect size; CI, confidence interval.
https://doi.org/10.1371/journal.pone.0242107.g003

this range. For GAS pharyngitis, the IRs ranged from 0–32 per 100 child-years (5 studies,
n = >2178) [7, 10, 18, 23, 24] Similarly, the highest incidence occurred in arid, Central Austra-
lia and the lowest in tropical, Northern Australia, in which no cases were observed.
In Australian Indigenous children, the incidence of sore throat was reported to be 8 and
457 per 100 child-years in Northern [7] and Central Australia [10] respectively. The incidence
of GAS pharyngitis was 0 per 100 child-years in Northern Australian children [7], and 32 per
100 person-years in Central Australian adults and children (data for children alone was not
reported) [10].

Sensitivity analysis
Given the large heterogeneity, I2 = 100% and I2 = 99.9% for sore throat and GAS pharyngitis
respectively, we assessed the influence of sample size which showed no statistical difference for
studies considered to be adequately powered compared with those inadequately powered
(P>0.05). Country income level according to the world bank classification at the time the
study was conducted was also considered and there was no statistical difference in rates by
income category. We intended to pool the IRs by risk of bias assessment, however, all the
included studies in the meta-analysis had a low risk of bias.
Assessment of risk of bias in included studies. Eight studies had a low risk of bias, [7, 10,
15, 25–29] and three had a moderate risk of bias, [18, 23, 24] (S2 Table; S1 Fig).

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 10 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

Discussion
This systematic review is the first comprehensive synthesis of the incidence of sore throat and
GAS pharyngitis in children at risk of ARF. The main findings are:
1. The pooled incidence of sore throat in children at risk of ARF is 82.5 (95% CI, 6.5 to
1044.4) per 100 child-years.
2. The pooled incidence of GAS pharyngitis in children at risk of ARF is 10.8 (95% CI, 2.3 to
50.0) per 100 child-years.
The pooled IR for sore throat (82.5 per 100 child-years) is higher than estimates reported
elsewhere [32]. A prospective surveillance study conducted in Melbourne reported IRs for sore
throat ranging from 32.70–40 per 100 child-years (urban children, ages 0–18 years) [32]. In an
urban setting at lower risk of ARF, the incidence of sore throat was far lower than what was
found in our study [33]. The pooled incidence of GAS pharyngitis; however, was comparable
to other studies. A prospective surveillance study of GAS pharyngitis in former Czechoslovakia
in 1984 reported an incidence of 7.2 per 100 person-years (urban children and adults) [34].
Two studies in urban families in Melbourne, Australia were also conducted in the early 2000s.
The first study comprising children and adults found an IR of 14 per 100 person-years for
GAS pharyngitis [32]. The second study reported IRs for GAS pharyngitis ranging from 9.2–
12.8 per 100 child-years in children aged 0–18 years [33]. Interestingly, a study conducted in
Philadelphia, USA in the 1950s reported the IR for GAS pharyngitis of 15.5 per 100 child-
years, [18] however, some of the included schools were part of a penicillin treatment program
implying that their findings could be an underestimate.
Our meta-analysis also highlights that there is a stark difference between the incidence of
sore throat and GAS pharyngitis, almost 8-fold lower (82.5 versus 10.8 per 100 child-years,
p<0.05). A recent prevalence study; however, reports that the proportion of GAS causing
pharyngitis is between 20 and 30% [3]. Therefore, one might expect that we would have found
approximately a third the rate of GAS pharyngitis compared to sore throats. The 10-fold lower
rate may be improved in precision with more studies to accurately describe the incidence of
GAS pharyngitis.
We considered the impact of study setting, geographical setting, country, climate and coun-
try income level on the overall incidence estimates of sore throat and GAS pharyngitis in chil-
dren at risk of developing ARF. The IR for both sore throat and GAS pharyngitis was lowest in
studies conducted in the clinic setting and highest in studies conducted in households. These
differences are to be expected; clinic-based studies are likely to see more severe disease rather
than the mild cases that would not warrant a visit to the clinic. The high incidence in the
household setting may be explained by socio-economic risk factors including overcrowding,
given that the study was undertaken in Indian Slums and would therefore be surveying both
mild and severe disease. The school setting is a convenient site for recruitment given the popu-
lation of interest and children with sore throat may not always warrant a visit to the clinic.
Another strength of school-based surveillance is that even though a child with sore throat may
be missed due to absenteeism, depending on how often surveillance takes place, it is possible
to evaluate that case upon return if symptoms are still present. We also compared the pooled
IR by each setting to the overall pooled IR for sore throat and GAS pharyngitis. In the school
setting, the pooled IR for sore throat of 82.5 (95% CI, 6.5 to 1044.4) per 100 child-years was
similar to overall pooled estimate of 186.6 (95% CI, 141.8 to 245.6) per 100 child-years. Simi-
larly, for GAS pharyngitis, the overall pooled IR of 10.8 (95% CI, 2.3 to 50.0) per 100 child-
years was similar to the pooled IR of 15.9 (95% CI, 6.7 to 37.7) per 100 child-years found in the
school setting. This finding suggests that the school settings is probably a more accurate

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 11 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

representation of the incidence among children at risk of ARF compared to studies conducted
in clinic or household settings, which may either under or overestimate the true burden of dis-
ease in this population.
When we considered the impact of geographical setting, studies conducted in rural settings
had a significantly higher IR compared with urban settings and may be explained by socioeco-
nomic factors which influence infection such as household overcrowding, particularly in rural
living indigenous communities. This difference is in contrast to findings in a systematic review
on the prevalence of GAS pharyngitis conducted in Africa, where risk factors such as over-
crowding are more likely to occur in urban settings [35].
A striking feature of this systematic review is the markedly higher IR of both sore throat
and GAS pharyngitis reported by a study conducted in India [25] compared to the study con-
ducted in South Africa [28]. While both studies were conducted in peri-urban communities of
low socio-economic status, South Africa is considered to be an upper-middle income country
with better living conditions compared with India as a low-income country. Another reason
for this disparity could be the recruitment setting in which these studies were conducted. We
also observed a significant difference in IR’s between studies conducted in Northern India.
Both studies by Kumar et al. were conducted in the villages of the Panchkula district with bet-
ter socio-economic conditions compared to the Nandi et al. study conducted in the slums of
Chandigarh.
An interesting observation among the studies not amenable to meta-analysis, was the con-
trast between the results of two studies conducted in remote Indigenous communities [7, 10].
Both were conducted in households with a history of RHD, yet the community in Central Aus-
tralia [10] had markedly higher incidences of sore throat and GAS pharyngitis, while the Top
End of the Northern Territory [7] community had extremely low rates of sore throat and GAS
pharyngitis. Two distinct differences between the communities would be their climates, tropi-
cal compared to arid, and their socioeconomic status, with the Central Australian community
noted to be better resourced and have less household crowding than the Top End community
[10]. The difference in resources would be unlikely to account for the higher IRs in the Central
Australian community, as one would expect rates to be lower with less household crowding
[10]. The tropical climate could be an important factor contributing to the contrasting rates.
The study by Steer 2009 was the only other study conducted in a tropical country, and the IRs
of sore throat and GAS pharyngitis were higher than in Top End Northern Australia, in fact
they were quite close to the IRs found in the Melbourne studies. Without other studies, it is
impossible to infer that there is an effect of the tropical climate. Overall, there is no satisfactory
explanation for the contrast between the results of the two McDonald studies, however,
monthly surveillance visits were conducted and thus could have missed many episodes of sore
throat and GAS pharyngitis, warranting a more robust methods to estimate the true burden in
those settings.
A strength of this review was the broad inclusive search strategy to avoid missing appro-
priate studies. We also recalculated incidence from appropriate studies to ensure that the
same IR were being compared. The risk of bias analysis also highlights a strength, in that all
included studies had either a low or moderate risk of bias, none were at high risk. The robust
meta-analysis method allowed a pooled incidence estimate to be calculated from appropriate
studies. Limitations include difficulty in capturing the ‘children at risk of ARF’ in the search
strategy, even with guidance from previous articles. It was difficult to differentiate viral phar-
yngitis with GAS pharyngeal colonization from true GAS pharyngitis, which may have
affected the results; although this was a limitation across all studies and is an ongoing clinical
challenge in understanding the microbiology of pharyngitis. There was also a variation in
methods used to establish the incidence of GAS pharyngitis and convenience sampling may

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 12 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

have influenced the incidence rates during a particular period of time. Most studies used a
swab and culture to confirm GAS pharyngitis, though some also included serological testing,
which is not the gold standard. The variability in methods may have effected results and the
utility of comparison.
The findings of this study have implications for both clinical practice and research. For cli-
nicians, these results indicate that children from low to upper-middle income countries or
Indigenous children in high-income countries suffer a high burden of sore throat and GAS
pharyngitis. Other GAS infections such as impetigo may also be an important risk factor for
developing ARF and thus these children should be managed with the prevention of ARF in
mind [36]. These results also reinforce the importance of primary prevention programs with
the focus on treating GAS pharyngitis to prevent ARF. A number of questions remain unan-
swered, particularly the disparity between sore throat and GAS incidence in the two McDon-
ald studies from Australia. More research is warranted in the Indigenous Australian
population living in Central and Northern Australia where there are some of the highest
reported rates of ARF in the world [5].

Conclusion
Our results confirm the high incidence of sore throat and GAS pharyngitis infections in chil-
dren at risk of developing ARF, warranting the need for improved prevention and treatment
strategies especially in remote regions and in Indigenous populations. Despite being a com-
mon condition [1] and well evidenced to precede ARF [2], there are few available studies
describing the incidence of sore throat or GAS pharyngitis in children globally. This under-
studied condition makes it difficult to use the available evidence to inform estimates of sore
throat and GAS pharyngitis for Australian Indigenous children at high risk of ARF and thus
emphasizes the need for more research in the context of a high burden of impetigo. Finally,
our findings support the call to accelerate the development of a GAS vaccine as a primary pre-
vention strategy to reduce the burden of GAS diseases in populations at risk of developing
ARF [37].

Supporting information
S1 Checklist. Checklist of the preferred reporting items for systematic reviews and meta-
analyses for the systematic review of sore throat and GAS pharyngitis in children at risk of
acute rheumatic fever.
(DOCX)
S1 Table. List of excluded studies following full text review [38–46].
(PDF)
S2 Table. Summary of risk of bias analysis for studies reporting the incidence of sore
throat and GAS pharyngitis in children at risk of acute rheumatic fever. Quality score: 0–5
high risk of bias (pink), 6–8 moderate risk of bias (yellow), > 8 low risk of bias (green).
(PDF)
S1 Fig. Risk of bias chart for studies reporting the incidence of sore throat and GAS phar-
yngitis in children at risk of acute rheumatic fever. Green corresponds to low risk of bias,
yellow corresponds to moderate risk of bias.
(TIF)

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 13 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

Acknowledgments
The authors would like to acknowledge Dr. Emma Griffiths and Dr. Sally Singleton for their
support and guidance in the early stages of this review.

Author Contributions
Conceptualization: Sarah Pearce, Asha C. Bowen, Dylan D. Barth.
Data curation: Sarah Pearce, Maya de la Lande.
Formal analysis: Sarah Pearce, Mark E. Engel, Dylan D. Barth.
Investigation: Sarah Pearce, Asha C. Bowen, Mark E. Engel.
Methodology: Asha C. Bowen, Dylan D. Barth.
Resources: Asha C. Bowen.
Supervision: Dylan D. Barth.
Validation: Asha C. Bowen.
Writing – original draft: Sarah Pearce.
Writing – review & editing: Sarah Pearce, Asha C. Bowen, Mark E. Engel, Maya de la Lande,
Dylan D. Barth.

References
1. Bryant AE, Stevens DL. Streptococcus pyogenes. In: Bennett J, Dolin R, Blaser M, editors. Mandell,
Douglas, and Bennett’s Principles and Practice of Infectious Diseases, Updated Edition2015. p. 2285–
99.
2. Carapetis J, Bowen A, May P. The inequitable burden of group A streptococcal diseases in Indigenous
Australians: we need to fill evidence gaps and make clinical advances to reduce these diseases of dis-
advantage (Report). MJA. 2016; 205(5).
3. Oliver JR, Pierse N, Stefanogiannis N, Jackson C, Baker MG. Acute rheumatic fever and exposure to
poor housing conditions in New Zealand: A descriptive study. Journal of Paediatrics and Child Health.
2017; 53(4):358–64. https://doi.org/10.1111/jpc.13421 PMID: 28052445
4. Carapetis JR, Steer AC, Mulholland EK, Weber M. The global burden of group A streptococcal dis-
eases. Lancet Infect Dis. 2005; 5:685–94. https://doi.org/10.1016/S1473-3099(05)70267-X PMID:
16253886
5. Parnaby MG, Carapetis JR. Rheumatic fever in Indigenous Australian children. J Paediatr Child Heatlh.
2010; 46(9):257–33. https://doi.org/10.1111/j.1440-1754.2010.01841.x PMID: 20854325
6. McDonald M, Currie BJ, Carapetis JR. Acute rheumatic fever: a chink in the chain that links the heart to
the throat? The Lancet Infect Dis. 2004; 4(4):240–5. https://doi.org/10.1016/S1473-3099(04)00975-2
PMID: 15050943
7. McDonald MI, Towers RJ, Andrews RM, Benger N, Currie BJ, Carapetis JR. Low Rates of Streptococ-
cal Pharyngitis and High Rates of Pyoderma in Australian Aboriginal Communities Where Acute Rheu-
matic Fever Is Hyperendemic. Clin Infect Dis. 2006; 43(6):683–9. https://doi.org/10.1086/506938
PMID: 16912939
8. Bowen A, Mahe A, Hay RJ, Andrews RM, Steer AC, Tong S, et al. The global epidemiology of impetigo:
a systematic review of the population prevalence of impetigo and pyoderma. PLoS One. 2015; 10(8).
https://doi.org/10.1371/journal.pone.0136789 PMID: 26317533
9. Yeoh DK, Anderson A, Cleland G, Bowen AC. Are scabies and impetigo “normalised”? A cross-sec-
tional comparative study of hospitalised children in northern Australia assessing clinical recognition and
treatment of skin infections. PLoS Neglected Tropical Diseases. 2017; 11(7):e0005726. https://doi.org/
10.1371/journal.pntd.0005726 PMID: 28671945
10. McDonald M, Brown A, Edwards T, Hope A, Amu M, Morey F, et al. Apparent Contrasting Rates of
Pharyngitis and Pyoderma in Regions where Rheumatic Heart Disease is Highly Prevalent. Heart Lung
Circ. 2007; 16(4):254–9. https://doi.org/10.1016/j.hlc.2007.02.087. PMID: 17581789

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 14 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

11. Carapetis JR, Beaton A, Cunningham MW, Guilherme L, Karthikeyan G, Mayosi BM, et al. Acute rheu-
matic fever and rheumatic heart disease. Nat Rev Dis Primers. 2016; 2:1–24. https://doi.org/10.1038/
nrdp.2015.84 PMID: 27188830
12. Martin WJ, Steer AC, Smeesters PR, Keeble J, Inouye M, Carapetis J, et al. Post-infectious group A
streptococcal autoimmune syndromes and the heart. Autoimmunity Reviews. 2015; 14(8):710–25.
https://doi.org/10.1016/j.autrev.2015.04.005. PMID: 25891492
13. Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. Preferred reporting items for
systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Systematic reviews.
2015; 4(1):1. https://doi.org/10.1186/2046-4053-4-1 PMID: 25554246
14. May P, Bowen A, Tong S, Steer A, Prince S, Andrews R, et al. Protocol for the systematic review of the
prevention, treatment and public health management of impetigo, scabies and fungal skin infections in
resource-limited settings. Systematic Reviews. 2016; 5:162. https://doi.org/10.1186/s13643-016-0335-0
PMID: 27659511
15. Steer CA, Jenney WJA, Kado FJ, Good EM, Batzloff RM, Magor RG, et al. Prospective Surveillance of
Streptococcal Sore Throat in a Tropical Country. The Pediatric Infectious Disease Journal. 2009; 28
(6):477–82. https://doi.org/10.1097/INF.0b013e318194b2af PMID: 19483515
16. Cohen JF, Bertille N, Cohen R, Chalumeau M. Rapid antigen detection test for group A streptococcus in
children with pharyngitis. Cochrane Database Syst Rev. 2016; 7(7):CD010502.
17. Peel M, Finlayson B, McMahon T. Updated world map of the Koeppen-Geiger climate classification.
Hydrology and Earth System Sciences. 2007; 11(5):1633–44. https://doi.org/10.5194/hess-11-1633-
2007
18. El-Kholy A, Sorour AH, Houser HB, Wannamaker LW, Robins M, Poitras JM, et al. A three-year pro-
spective study of streptococcal infections in a population of rural Egyptian school children. Journal of
medical microbiology. 1973; 6(1):101. https://doi.org/10.1099/00222615-6-1-101 PMID: 4690632
19. Centres for Disease Control and Prevention. Lesson 3: Measures of Risk. Principles of Epidemiology in
Public Health Practice, Third Edition: An Introduction to Applied Epidemiology and Biostatistics Atlanta,
USA2012 [cited 2018 July].
20. Shamliyan T, Kane RL, Dickinson S. A systematic review of tools used to assess the quality of observa-
tional studies that examine incidence or prevalence and risk factors for diseases. Journal of Clinical Epi-
demiology. 2010; 63(10):1061–70. https://doi.org/10.1016/j.jclinepi.2010.04.014 PMID: 20728045
21. Hoy D, Brooks P, Woolf A, Blyth F, March L, Bain C, et al. Assessing risk of bias in prevalence studies:
modification of an existing tool and evidence of interrater agreement. Journal of Clinical Epidemiology.
2012; 65(9):934–9. https://doi.org/10.1016/j.jclinepi.2011.11.014. PMID: 22742910
22. Werfalli M, Musekiwa A, Engel ME, Ross I, Kengne AP, Levitt NS. The prevalence of type 2 diabetes
mellitus among older people in Africa: a systematic review study protocol. BMJ Open. 2014; 4(6).
https://doi.org/10.1136/bmjopen-2013-004747 PMID: 24939807
23. Wu S, Peng X, Yang Z, Ma C, Zhang D, Wang Q, et al. Estimated burden of group a streptococcal phar-
yngitis among children in Beijing, China. BMC Infectious Diseases. 2016; 16(1). https://doi.org/10.1186/
s12879-016-1775-9 PMID: 27566251
24. Nicolle LE, Posti B, Urias B, Law B, Ling N. Group A streptococcal pharyngeal carriage, pharyngitis,
and impetigo in two northern Canadian native communities. Clin Invest Med. 1990; 3(3):99–106.
25. Nandi S, Kumar R, Ray P, Vohra H, Ganguly N. Group A streptococcal sore throat in a periuban popula-
tion of northern India: A one-year prospective study. Bulletin of The World Health Organization. 2001;
79(6):528–33. https://doi.org/10.1590/S0042-96862001000600008 PMID: 11436474
26. Lennon D. Rheumatic fever, a preventable disease? The New Zealand Experience. In: Streptococci
and streptococcal disease: entering the new milennium. Proceedings of the XIV Lancefield International
Symposium on Streptococci and Streptococcal Disease (Aukland, New Zealand). 2000.
27. Kumar R, Chakraborti A, Aggarwal A, Vohra H, Sagar V, Dhanda V, et al. Streptococcus pyogenes
pharyngitis & impetigo in a rural area of Panchkula district in Haryana, India. The Indian Journal of Medi-
cal Research. 2012; 135(1):133–6. https://doi.org/10.4103/0971-5916.93437 PMID: 22382196
28. Engel ME. A Study of Determinants and Prevalence of Rheumatic Heart Disease in Cape Town. PhD
Thesis: University of Cape Town; 2012. https://open.uct.ac.za/bitstream/item/3165/thesis_hsf_2012_
engel_m.pdf?sequence=1
29. Kumar R, Vohra H, Chakraborty A, Sharma YP, Bandhopadhya S, Dhanda V, et al. Epidemiology of
group A streptococcal pharyngitis & impetigo: a cross-sectional & follow up study in a rural community
of northern India. Indian Journal of Medical Research. 2009; 130(6):765. https://doi.org/10.7589/0090-
3558-46.1.316 PMID: 20090140
30. China: Human Development Indicators: United Nations; 2018. United National Development Pro-
gramme: Human Development Reports,:[http://hdr.undp.org/en/countries/profiles/CHN#.

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 15 / 16

PLOS ONE The incidence of sore throat and group A streptococcal pharyngitis

31. Ma J, Mitchell G, Dong G, Zhang W. Inequality in Beijing: A Spatial Multilevel Analysis of Perceived
Environmental Hazard and Self-Rated Health. Ann Am Assoc Geog. 2017; 107(1):109–29. https://doi.
org/10.1080/24694452.2016.1224636
32. Danchin MH, Rogers S, Selvaraj G, Kelpie L, Rankin P, Vorich R, et al. The burden of group A strepto-
coccal pharyngitis in Melbourne families. Indian J Med Res. 2004; 119:Suppl: 144–7.
33. Danchin MH, Rogers S, Kelpie L, Selvaraj G, Curtis N, Carlin JB, et al. Burden of Acute Sore Throat and
Group A Streptococcal Pharyngitis in School-aged Children and Their Families in Australia. Pediatrics.
2007; 120(5):950. https://doi.org/10.1542/peds.2006-3368 PMID: 17974731
34. Rotta J, Duben J, Jedlicka F, Havlickova H, Tumova B, Bruckova M. Prospective study of pharyngitis:
clnical diagnosis and microbiological profile. Zbl Backt 1989; 271:532–42.
35. Barth D, Naicker P, Engel K, Muhamed B, Basera W, Mayosi B, et al. Molecular Epidemiology of Nonin-
vasive and Invasive Group A Streptococcal Infections in Cape Town. mSphere. 18(6):e00907–19.
36. Robertson KA, Volmink JA, Mayosi BM. Antibiotics for the primary prevention of acute rheumatic fever:
a meta-analysis. BMC Cardiovasc Disord. 2005; 5(1):11-. https://doi.org/10.1186/1471-2261-5-11
PMID: 15927077.
37. Vekemans J, Gouvea-Reis F, Kim JH, Excler J-L, Smeesters PR, O’Brien KL, et al. The Path to Group
A Streptococcus Vaccines: World Health Organization Research and Development Technology Road-
map and Preferred Product Characteristics. Clinical Infectious Diseases. 2019; 69(5):877–83. https://
doi.org/10.1093/cid/ciy1143 PMID: 30624673
38. Nicolle LE, Law B, Postl B, Urias B, Ling N, Fast M. Group A streptococcal pharyngeal carriage and
impetigo in two northern native communities. Arctic Med Res. 1988; 47(1):669–71.
39. Sarkar S, Biswas R, Gaur SD, Sen PC, Reddy DCS. A study on sore throat and beta-haemolytic strep-
tococcal pharyngitis among rural school children in Varanasi, with reference to age and season. Indian
J Public Health. 1988; 32(4):191–98.
40. Nicolle LE, Postl B, Urias B, Ling N, Law B. Outcome following therapy of group a streptococcal infection
in schoolchildren in isolated northern communities. Can J Public Health. 1990; 81(6):468–70. PMID:
2282610
41. Tewodros W, Muhe L, Daniel E, Schalén C, Kronvall G. A one-year study of streptococcal infections
and their complications among Ethiopian children. Epidemiol Infect. 1992; 109(2):211–25. https://doi.
org/10.1017/s0950268800050172 PMID: 1397112
42. Majeed HA, al-Doussary L, Moussa MM, Yusuf AR, Suliman AH. Office diagnosis and management of
group A streptococcal pharyngitis employing the rapid antigen detecting test. A 1-year prospective
study of reliability and cost in primary care centres. Ann Trop Paediatr. 1993; 13(1):65–72. https://doi.
org/10.1080/02724936.1993.11747626 PMID: 7681647
43. Smeesters PR, Vergison A, Campos D, de Aguiar E, Miendje Deyi VY, Van Melderen L. Differences
between Belgian and Brazilian group A Streptococcus epidemiologic landscape. PLoS One. 2006;
20(1):e10.
44. Tartof SY, Farrimond F, de Matos JA, Reis JN, Ramos RT, Andrade AN, et al. Inverse association
between Lancefield group G Streptococcus colonization and sore throat in slum and nonslum settings
in Brazil. J Clin Microbiol. 2011; 49(1):409–12. https://doi.org/10.1128/JCM.02095-10 PMID: 21048010
45. Tapia DM, Sow OS, Tamboura MB, Keita PM, Berthe OA, Samake AM, et al. Streptococcal Pharyngitis
in Schoolchildren in Bamako, Mali. The Pediatric Infectious Disease Journal. 2015; 34(5):463–8. https://
doi.org/10.1097/INF.0000000000000608 PMID: 25356953
46. Shay S, Shapiro NL, Bhattacharyya N. Pediatric otolaryngologic conditions: Racial and socioeconomic
disparities in the United States. Laryngoscope. 2017; 127(3):746–52. https://doi.org/10.1002/lary.
26240 PMID: 27599638

PLOS ONE | https://doi.org/10.1371/journal.pone.0242107 November 18, 2020 16 / 16