Research

Original Investigation

Association Between Undervaccination With Diphtheria,

Tetanus Toxoids, and Acellular Pertussis (DTaP) Vaccine and
Risk of Pertussis Infection in Children 3 to 36 Months of Age
Jason M. Glanz, PhD; Komal J. Narwaney, MD, PhD; Sophia R. Newcomer, MPH; Matthew F. Daley, MD;
Simon J. Hambidge, MD, PhD; Ali Rowhani-Rahbar, MD, PhD; Grace M. Lee, MD, MPH; Jennifer C. Nelson, PhD;
Allison L. Naleway, PhD; James D. Nordin, MD, MPH; Marlene M. Lugg, DrPH; Eric S. Weintraub, MPH

IMPORTANCE Undervaccination is an increasing trend that potentially places children and

their communities at an increased risk for serious infectious diseases.
OBJECTIVE To examine the association between undervaccination and pertussis in children 3
to 36 months of age.
DESIGN Matched case-control study with conditional logistic regression analysis.
SETTING Eight managed care organizations of the Vaccine Safety Datalink between 2004 and

2010.
PARTICIPANTS Each laboratory-confirmed case of pertussis (72 patients) was matched to 4
randomly selected controls (for a total of 288 controls). The case patients were matched to
controls by managed care organization site, sex, and age at the index date. The index date
was defined as the date of pertussis diagnosis for the case patients.
EXPOSURE Undervaccination for the diphtheria, tetanus toxoids, and acellular pertussis
(DTaP) vaccine. Undervaccination was defined as the number of doses of DTaP vaccine that
was either missing or delayed by the index date. Case patients and controls could be
undervaccinated by 0, 1, 2, 3, or 4 doses of DTaP vaccine. Children undervaccinated by 0
doses were considered age-appropriately vaccinated by the index date.
MAIN OUTCOME AND MEASURE Pertussis.
RESULTS Of the 72 case patients with pertussis, 12 (16.67%) were hospitalized, and 34
(47.22%) were undervaccinated for DTaP vaccine by the date of pertussis diagnosis. Of the
288 matched controls, 64 (22.22%) were undervaccinated for DTaP vaccine.
Undervaccination was strongly associated with pertussis. Children undervaccinated for 3 or 4
doses of DTaP vaccine were 18.56 (95% CI, 4.92-69.95) and 28.38 (95% CI, 3.19-252.63)
times more likely, respectively, to have received a diagnosis of pertussis than children who
were age-appropriately vaccinated.
CONCLUSIONS AND RELEVANCE Undervaccination with DTaP vaccine increases the risk of
pertussis among children 3 to 36 months of age.

Author Affiliations: Author

affiliations are listed at the end of this
article.

JAMA Pediatr. 2013;167(11):1060-1064. doi:10.1001/jamapediatrics.2013.2353

Published online September 9, 2013.
1060

Downloaded From: http://archpedi.jamanetwork.com/ on 11/11/2014

Corresponding Author: Jason M.

Glanz, PhD, Institute for Health
Research, Kaiser Permanente
Colorado, PO Box 378066, Denver,
CO 80237 ([email protected]).
jamapediatrics.com

Risk of Pertussis in Children by Undervaccination

Original Investigation Research

hildren who are not age-appropriately vaccinated according to the recommendations of the Advisory Committee on Immunization Practices are considered
undervaccinated.1 A recent longitudinal study showed that the
rate of undervaccination increased significantly among a cohort of insured children born between 2004 and 2008 in the
United States.2 There are numerous potential reasons for undervaccination, including parental choice, missed opportunities, barriers to health care, and medical contraindication to
vaccination.3-6 Amid the recent pertussis epidemics across the
United States, undervaccination is a concerning trend that potentially places children at increased risk for serious infection. For the present study, we examined the association between undervaccination and the risk of pertussis in children
3 to 36 months of age.

Methods
Setting and Study Population
We conducted a matched case-control study that was nested
within a large cohort of children born between 2004 and 2008
(n = 323 247). These children were members of 8 managed care
organizations (MCOs) that comprise the Vaccine Safety
Datalink, a project funded by the Centers for Disease Control
and Prevention that links electronic administrative databases in order to conduct epidemiological studies of vaccine
safety.7 These databases contain information on demographics; vaccination history; and pharmacy, laboratory, and medical encounters in the outpatient, emergency department, and
inpatient settings. Each MCO sites institutional review board
approved the study.
Children in the study cohort had to be continuously enrolled in the MCO from 2 to 12 months of age and were then
followed up until 36 months of age unless they disenrolled from
their MCO. For each child, person-time follow-up accrued during periods of active membership enrollment.

Cases of Pertussis
In this large cohort, we identified potential cases of Bordetella pertussis infection using automated laboratory databases at each participating MCO. The automated pertussis laboratory results were primarily recorded as text and were not
stored in a standardized data format across all sites. For this
reason, a trained medical records abstractorblinded to the
childs vaccination statusmanually reviewed all of the laboratory results to confirm positive laboratory diagnoses of pertussis by a polymerase chain reaction test or by B pertussis cul-

ture. The date of a positive laboratory result for B pertussis

represented the index date for the case-control analysis. All
cases were diagnosed between 3 and 36 months of age.

Controls
Each case patient with pertussis was matched to 4 diseasefree controls by MCO site, sex, and age at the index date (7
days). The controls were selected from the cohort of children
enrolled in the MCO health plans between 2004 and 2010. Eligible controls did not have a record of pertussis prior to the index date.

Vaccination Status
For case patients and their matched controls, vaccination
status for diphtheria, tetanus toxoids, and acellular pertussis
(DTaP) vaccine was ascertained retrospectively from the
index date. The DTaP vaccine is recommended at 2, 4, 6,
15-18, and 48-83 months of age.8 Because the children in the
present study were followed up through 36 months of age,
we analyzed the first 4 doses of DTaP vaccine. Children were
either age-appropriately vaccinated or undervaccinated at
the index date. Undervaccination was defined as the number
of doses of DTaP vaccine that was either missing or delayed
by the index date (Table 1). The calculation for determining
undervaccination was based on a published algorithm that
measures the difference between when a vaccine dose was
actually administered and when the dose should have been
administered according to the Advisory Committee on
Immunization Practices schedule. The algorithm accounts
for age, grace periods, catch-up schedules, vaccine shortages, combination vaccines, and changes in vaccination
policy.2 For our study, if a child received a diagnosis of pertussis at 5 months of age on January 15, 2008, he or she
would be matched to 4 disease-free controls of the same sex
at the same MCO who were also 5 months of age between
January 7 and 22, 2008. For this matched stratum, ageappropriately vaccinated children are those who received 2
doses on time (with 0 doses missing or delayed), and undervaccinated children are those who received either 1 dose on
time (with 1 dose missing or delayed) or 0 doses on time
(with 2 doses missing or delayed). In the analysis, these 3
groups of children would be classified as being undervaccinated by 0, 1, and 2 doses of DTaP vaccine, respectively.

Analysis
The final case-control population was analyzed with conditional logistic regression to estimate odds ratios and 95% CIs.
In the regression models, the outcome variable was pertussis

Table 1. Parameters for Determining the DTaP Vaccination Status of the Study Populationa

DTaP Vaccine,
No. of Doses

Recommended
Age per ACIP, mo

Minimum
Acceptable
Age,b d

Minimum Acceptable
Interval Between
Doses,b d

Age of Child
When Count for
Undervaccination
Initiated, d

38

66

24

154

94

24

215

361

179

580

15-18

jamapediatrics.com

Downloaded From: http://archpedi.jamanetwork.com/ on 11/11/2014

93

Abbreviations: ACIP, Advisory

Committee on Immunization
Practices; DTaP, diphtheria, tetanus
toxoids, and acellular pertussis.
a

Adapted from Luman et al1 and

Glanz et al.2

Allowing for a 4-day grace period.

JAMA Pediatrics November 2013 Volume 167, Number 11

1061

Research Original Investigation

Risk of Pertussis in Children by Undervaccination

Table 2. Demographic Characteristics and DTaP Vaccination Status

of the Study Population

Table 3. Estimates of the Risk of Laboratory-Confirmed Pertussis

for Those Undervaccinated vs Those Age-Appropriately Vaccinateda
Comparison
of DTaP
Vaccine Doses
Undervaccinated by

Children, No. (%)

Cases
(n = 72)

Controls
(n = 288)

Sex

OR (95% CI)

1 vs 0

2.25 (0.97-5.24)

P Value
.06

Female

36 (50.00)

144 (50.00)

2 vs 0

3.41 (0.89-13.05)

.07

Male

36 (50.00)

144 (50.00)

3 vs 0

18.56 (4.92-69.95)

<.001

14.27 (10.12)

14.27 (10.07)

4 vs 0

28.38 (3.19-252.63)

.002

4.36 (2.23-8.55)

<.001

Age, mean (SD), mo

Age group, y

1, 2, 3, or 4 vs 0

<1

39 (54.17)

156 (54.17)

1-2

18 (25.00)

72 (25.00)

Abbreviations: DTaP, diphtheria, tetanus toxoids, and acellular pertussis;

OR, odds ratio.

>2-3

15 (20.83)

60 (20.83)

34 (47.22)

64 (22.22)

38 (52.78)

224 (77.78)

12 (16.67)

47 (16.32)

4 (5.56)

8 (2.78)

12 (16.67)

6 (2.08)

6 (8.33)

3 (1.04)

Undervaccinated for DTaP vaccine

No. of DTaP vaccine doses
undervaccinated by

Abbreviation: DTaP, diphtheria, tetanus toxoids, and acellular pertussis.

case status, and the predictor variable was undervaccination

status. Case patients and controls were undervaccinated by 0,
1, 2, 3, or 4 doses of DTaP vaccine at the index date. Children
undervaccinated by 0 doses (age-appropriately vaccinated) represented the referent group. To examine a dose-response relationship, we compared the risk of pertussis for children undervaccinated by 1, 2, 3, or 4 doses of DTaP vaccine with the
risk of pertussis for the referent group. The matched odds ratios from the conditional logistic regression models were used
to calculate the attributable risk percent of pertussis in the total
population.9

Results

To examine a dose-response relationship, we compared the risk of pertussis

for children undervaccinated by 1, 2, 3, or 4 doses of DTaP vaccine with the risk
of pertussis for the referent group (0 doses missed).

same sex distribution as the case patients. Sixty-four (22.22%)

of the controls were undervaccinated, and 4 of these 6 controls (6.25%) had an ICD-9-CM code for vaccine refusal.

Risk of Pertussis
Undervaccination was strongly associated with laboratoryconfirmed pertussis in children 3 to 36 months of age (Table 3).
There was an apparent dose-response relationship in which the
risk of pertussis increased as the magnitude of undervaccination with DTaP vaccine increased. Although not statistically significant, children undervaccinated by 1 or 2 doses of DTaP vaccine were 2.25 (95% CI, 0.97-5.24) and 3.41 (95% CI, 0.89-13.05)
times more likely, respectively, to have received a diagnosis of
pertussis than children who were age-appropriately vaccinated. Children undervaccinated by 3 or 4 doses were 18.56 (95%
CI, 4.92-69.95) and 28.38 (95% CI, 3.19-252.63) times more likely,
respectively, to have received a diagnosis of pertussis than children who were age-appropriately vaccinated. The attributable
risk percent in the entire population was 36.39% (95% CI, 19.65%49.66%), suggesting that 36.39% of all of cases in the population were attributed to undervaccination.

Cases and Controls

In the automated laboratory databases, there were 1522 children between 3 and 36 months of age with a laboratory test
result (positive or negative) for pertussis between 2004 and
2010. A manual review of the laboratory results revealed that
72 of these 1522 children (4.73%) received a laboratoryconfirmed diagnosis of pertussis. The mean age of the case patients was 14.27 months (median age, 11.06 months), and 50%
were female patients (Table 2). Of these 72 case patients, 12
(16.67%) were hospitalized. At the index date, 38 (52.78%) were
age-appropriately vaccinated with DTaP vaccine, and 34
(47.22%) were undervaccinated. Of the 34 undervaccinated
children, 10 (29.41%) had an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) code for
parental vaccine refusal/delay (codes V64.05 and V64.06), suggesting that the parents had intentionally refused or delayed
vaccine doses for personal, nonmedical reasons.
Four controls were matched to each case patient (for a total
of 288 controls). The controls had the same mean age and the
1062

JAMA Pediatrics November 2013 Volume 167, Number 11

Downloaded From: http://archpedi.jamanetwork.com/ on 11/11/2014

Discussion
In our case-control study of infants and young children from
8 MCOs across the United States, we found a significant increased risk for pertussis in undervaccinated children between 2004 and 2010. We used automated MCO health care
databases and medical record reviews to identify laboratoryconfirmed cases of pertussis and automated immunization records to ascertain vaccination status. Not surprisingly, we found
that the risk of pertussis greatly increased as undervaccination with DTaP vaccine increased. Our data also suggest that
approximately 36% of cases of pertussis in children 3 to 36
months of age could have been prevented with on-time vaccination with DTaP vaccine.
Our study has limitations. Most notably, we believe that
our analyses may not accurately estimate the true risks among
children undervaccinated by either 1 or 2 doses of DTaP vaccine, owing to the heterogeneity of the children defined to be
jamapediatrics.com

Risk of Pertussis in Children by Undervaccination

Original Investigation Research

in these strata. For example, the stratum of children undervaccinated by 1 dose of DTaP vaccine included children 21
months of age who received 3 doses of DTaP vaccine on time
and children 3 months of age who received 0 doses of DTaP
vaccine on time. Similarly, the stratum of children undervaccinated by 2 doses of DTaP vaccine included children 24 months
of age who received 2 doses on time, as well as children 5
months of age who received 0 doses on time. In these scenarios, it is reasonable to assume that the younger children who
never received any doses on time would be at greater risk than
the older children who received some doses on time. Although it is biologically plausible that there is an interaction
between age and dose, there were too few cases in the first 2
strata of undervaccination to adequately examine the interaction in the analysis.
There are several potential confounding factors that may
be associated with both vaccination status and exposure to pertussis, including geography, household size, presence of adolescents in the household, day-care attendance, and Hispanic
ethnicity.10-12 Our analyses may have been limited because the
Vaccine Safety Datalink databases do not routinely capture
most of these variables. However, in our analysis, we were able
to match case patients to controls on MCO site, which helped
to control for geographic variation in vaccination status and
pertussis.
Lastly, our results may have also been influenced by a diagnostic bias. It is possible that physicians are more likely to
conduct pertussis tests on undervaccinated children than ageappropriately vaccinated children when they present to the
clinic for acute infections. This, in turn, would lead to an overestimate of the association between undervaccination and pertussis. However, it has also been shown that undervacci-

ARTICLE INFORMATION
Accepted for Publication: March 14, 2013.
Published Online: September 9, 2013.
doi:10.1001/jamapediatrics.2013.2353.
Author Affiliations: Institute for Health Research,
Kaiser Permanente Colorado, Denver (Glanz,
Narwaney, Newcomer, Daley, Hambidge);
Department of Epidemiology, Colorado School of
Public Health, Aurora (Glanz, Hambidge);
Department of Pediatrics, University of Colorado,
Aurora (Daley, Hambidge); Community Health
Services, Denver Health, Colorado (Hambidge);
Kaiser Permanente Vaccine Study Center, Oakland,
California (Rowhani-Rahbar); Center for Child
Health Care Studies, Department of Population
Medicine, Harvard Pilgrim Health Care Institute and
Harvard Medical School, Boston, Massachusetts
(Lee); Group Health Cooperative, Seattle,
Washington (Nelson); Kaiser Permanente
Northwest, Portland, Oregon (Naleway);
HealthPartners Research Foundation, Minneapolis,
Minnesota (Nordin); Department of Research and
Evaluation, Southern California Kaiser Permanente,
Pasadena (Lugg); Immunization Safety Office,
Division of Healthcare Quality Promotion, Centers
for Disease Control and Prevention, Atlanta,
Georgia (Weintraub).
Author Contributions: Dr Glanz had full access to
all of the data in the study and takes responsibility

nated children have lower outpatient visit rates than ageappropriately vaccinated children.2 Therefore, although more
pertussis cases would be identified in the undervaccinated
group because of increased testing, some cases in the undervaccinated group may have been missed because the children in this group use the medical system less frequently than
age-appropriately vaccinated children. These potential biases would distort the odds ratio estimates in the opposite direction, and it is unlikely that they explain the large association between undervaccination and pertussis that we found
in the present study.
The United States is currently experiencing the largest
outbreak of pertussis in 50 years.10,13 There are numerous
possible reasons for this outbreak, including an increased
awareness of the disease and increased testing, more widespread use of polymerase chain reaction testing, and the fact
that DTaP vaccines may be less potent and provide protection for less time than the old DTP vaccines.13-15 Our data
suggest that undervaccination, whether due to parental
refusal of vaccines or other barriers to health care, is an
important contributing factor, especially given the documented increasing rates of undervaccination in 2 recently
published studies.2,16
Undervaccination with DTaP vaccine places infants and
young children at increased risk for pertussis. Although not
supported by our data, it is also possible that undervaccination indirectly threatens the health of surrounding populations that are at high risk for serious complications from pertussis, such as infants who are too young to be vaccinated.17
We believe that our study supports on-time vaccination with
DTaP vaccine, as recommended by the Advisory Committee
on Immunization Practices.8

for the integrity of the data and the accuracy of the

data analysis.
Study concept and design: Glanz, Narwaney,
Newcomer, Daley, Hambidge, Naleway, Nordin.
Acquisition of data: Glanz, Newcomer, Hambidge,
Lee, Nelson, Naleway, Lugg, Weintraub.
Analysis and interpretation of data: Glanz,
Narwaney, Newcomer, Daley, Hambidge,
Rowhani-Rahbar, Lee, Nelson, Nordin, Lugg,
Weintraub.
Drafting of the manuscript: Glanz, Daley, Nordin.
Critical revision of the manuscript for important
intellectual content: All authors.
Statistical analysis: Glanz, Narwaney, Newcomer,
Rowhani-Rahbar, Nelson, Lugg, Weintraub.
Obtained funding: Glanz, Hambidge.
Administrative, technical, or material support:
Glanz, Newcomer, Lugg, Weintraub.
Study supervision: Glanz, Newcomer, Daley,
Hambidge, Lugg, Weintraub.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was funded through
a subcontract with America's Health Insurance
Plans under contract 200-2002-00732 from the
Centers for Disease Control and Prevention.
Disclaimer: Although the Centers for Disease
Control and Prevention played a role in the design
and conduct of the study, collection, management,
analysis, and interpretation of the data, as well as
preparation, review, and approval of the

jamapediatrics.com

Downloaded From: http://archpedi.jamanetwork.com/ on 11/11/2014

manuscript, the findings and conclusions in this

report are those of the authors and do not
necessarily represent the official position of the
Centers for Disease Control and Prevention.
Previous Presentation: Preliminary results from
this study were presented at the Institute of
Medicine committees meeting on Assessment of
Studies of Health Outcomes Related to the
Childhood Immunization Schedule; May 29, 2012;
Washington, DC.
Additional Contributions: We sincerely thank the
data management and medical record abstraction
staff at each of the participating sites for their
valuable work in creating the Vaccine Safety
Datalink data sets and reviewing the medical
records.
REFERENCES
1. Luman ET, McCauley MM, Stokley S, Chu SY,
Pickering LK. Timeliness of childhood
immunizations. Pediatrics. 2002;110(5):935-939.
2. Glanz JM, Newcomer SR, Narwaney KJ, et al. A
population-based cohort of undervaccination in 8
managed care organizations across the United
States. JAMA Pediatr. 2013;167(3):274-281.
3. Dempsey AF, Schaffer S, Singer D, Butchart A,
Davis M, Freed GL. Alternative vaccination schedule
preferences among parents of young children.
Pediatrics. 2011;128(5):848-856.

JAMA Pediatrics November 2013 Volume 167, Number 11

1063

Research Original Investigation

4. Freed GL, Clark SJ, Butchart AT, Singer DC, Davis

MM. Parental vaccine safety concerns in 2009.
Pediatrics. 2010;125(4):654-659.
5. Agency for Healthcare Research and Quality.
2009 National Healthcare Disparities Report.
http://www.ahrq.gov/research/findings/nhqrdr
/nhdr09/nhdr09.pdf. Published March 2010.
Accessed August 13, 2013.
6. Centers for Disease Control and Prevention.
Guide to vaccine contraindications and precautions.
http://www.cdc.gov/vaccines/recs/vac-admin
/downloads/contraindications-guide-508.pdf.
Accessed August 13, 2013.
7. Baggs J, Gee J, Lewis E, et al. The Vaccine Safety
Datalink: a model for monitoring immunization
safety. Pediatrics. 2011;127(suppl 1):S45-S53.
8. Recommended immunization schedules for
persons aged 0 through 18 YearsUnited States,
2012 [published correction appears in MMWR Morb

1064

Risk of Pertussis in Children by Undervaccination

Mortal Wkly Rep. 2012;61(8):147]. MMWR Morb

Mortal Wkly Rep. 2012;61(5):1-4.
9. Greenland S. Variance estimators for
attributable fraction estimates consistent in both
large strata and sparse data. Stat Med.
1987;6(6):701-708.
10. Centers for Disease Control and Prevention
(CDC). Pertussis epidemicWashington, 2012.
MMWR Morb Mortal Wkly Rep. 2012;61(28):517-522.
11. Wendelboe AM, Hudgens MG, Poole C, Van Rie
A. Estimating the role of casual contact from the
community in transmission of Bordetella pertussis
to young infants. Emerg Themes Epidemiol.
2007;4:15.
12. Wendelboe AM, Njamkepo E, Bourillon A, et al;
Infant Pertussis Study Group. Transmission of
Bordetella pertussis to young infants. Pediatr Infect
Dis J. 2007;26(4):293-299.

JAMA Pediatrics November 2013 Volume 167, Number 11

Downloaded From: http://archpedi.jamanetwork.com/ on 11/11/2014

13. Cherry JD. Epidemic pertussis in 2012the

resurgence of a vaccine-preventable disease. N Engl
J Med. 2012;367(9):785-787.
14. Cherry JD. Why do pertussis vaccines fail?
Pediatrics. 2012;129(5):968-970.
15. Klein NP, Bartlett J, Rowhani-Rahbar A, Fireman
B, Baxter R. Waning protection after fifth dose of
acellular pertussis vaccine in children. N Engl J Med.
2012;367(11):1012-1019.
16. Robison SG, Groom H, Young C. Frequency of
alternative immunization schedule use in a
metropolitan area. Pediatrics. 2012;130(1):32-38.
17. Grizas AP, Camenga D, Vzquez M. Cocooning:
a concept to protect young children from infectious
diseases. Curr Opin Pediatr. 2012;24(1):92-97.

jamapediatrics.com