Journal of Adolescent Health 52 (2013) 427e432

www.jahonline.org

Original article

Factors Predicting Completion of the Human Papillomavirus Vaccine Series

Rachel Gold, Ph.D., M.P.H. *, Allison Naleway, Ph.D., and Karen Riedlinger, M.P.H.
The Center for Health Research, Kaiser Permanente Northwest, 3800 N Interstate Avenue, Portland, Oregon

Article history: Received May 16, 2012; Accepted September 14, 2012
Keywords: Human papillomavirus vaccine; HPV4; Vaccine series completion; Vaccine adherence; Adolescent females; Cervical cancer

A B S T R A C T
IMPLICATIONS AND
CONTRIBUTION
Purpose: This study identified factors associated with completion of the three dose quadrivalent
human papillomavirus vaccine (HPV4) series by female adolescents.
Persons who receive only
Methods: Between February and September 2008, we prospectively surveyed 11- to 26-year-old part of the three-dose hu-
female members of an integrated managed care organization shortly after their first HPV4 dose to man papillomavirus vac-
identify factors that predicted series completion. We used regression analyses to assess whether cine series may not be fully
self-reported experiences at the index visit, knowledge/attitudes about HPV and HPV4, and protected against human
medical record data on adverse events, demographic characteristics, care-utilization frequency, papilloma virus, yet comp-
and visit characteristics, were associated with vaccine series completion within one year of the first letion rates remain sub-
HPV4 dose. optimal. This prospective
Results: Of 899 survey respondents (27% of 3347 survey recipients), 786 (87%) maintained study identified patient-
continuous enrollment in the health plan in the year following the first HPV4 dose. Fifty percent and provider-level factors
(n ¼ 393) completed the vaccine series within that year. In multivariate analyses of survey predicting completion of
respondents, only respondents’ ability to correctly identify the number of shots required for series the series. We suggest
completion was significantly associated with series completion. Reported bruising was associated strategies that providers
with decreased likelihood of completion, and the clinician stating that future shots were required can use to promote series
was associated with increased likelihood, but both were of borderline significance. Females ages completion.
16e20 had the lowest series completion.
Conclusions: Improving HPV4 completion will require targeted efforts. Our results suggest that
providers may help by stressing the need for additional doses of vaccine, and confirming that
patients understand this information. Special attention should be given to females ages 16e20.
Future randomized trials should assess the effect on vaccine completion of these simple, low-
cost interventions.
Ó 2013 Society for Adolescent Health and Medicine. All rights reserved.

The quadrivalent human papillomavirus vaccine (HPV4) is
recommended for females ages 11e12 years and for catch-up
vaccination of 13e26 year olds [1]. Clinical trials have demon-
strated that HPV4 is highly effective in preventing infection with
HPV types 16 and 18, which are associated with 70% of cervical
cancer cases worldwide, and HPV types 6 and 11, which are
associated with 90% of genital wart cases. The vaccine is a three-
* Address correspondence to: Rachel Gold, Ph.D., M.P.H., The Center for Health
Research, Kaiser Permanente Northwest, 3800 N Interstate Avenue, Portland, OR
97227.
E-mail address: rachel.gold@kpchr.org (R. Gold).
dose series administered over the course of six months, with
recommendations for receipt of the second and third doses at
two and six months after the first dose. Because of this relatively
specific and condensed schedule, and because adolescents’
health careeseeking behaviors may not be conducive to
completion of a multidose immunization [2e9], adherence to the
vaccine series is of concern.
Since approval of HPV4 in 2006, studies on HPV4 series
completion have reported a range of completion rates varying
from 28% to 70% [10e14]. Clearly, many adolescents either do not
complete the series in the timeframe recommended or within
a year after beginning the series (the follow-up time used in most

1054-139X/$ e see front matter Ó 2013 Society for Adolescent Health and Medicine. All rights reserved.
http://dx.doi.org/10.1016/j.jadohealth.2012.09.009
428 R. Gold et al. / Journal of Adolescent Health 52 (2013) 427e432
published studies). Persons who receive a partial series may not
be fully protected against HPV [10,11], with implications both for
health outcomes and for potentially wasted vaccine and financial
expenditure by individuals or insurers. Therefore, a first step in
addressing these low completion rates is to explore which factors
predict completion of the series among those who start it. A few
studies have begun to explore this question, using data from
electronic medical records (EMR) or chart review. These studies
identified age, type of provider administering the first dose, some
measures of socioeconomic status (SES), and race/ethnicity as
associated with HPV4 series completion [10e14].
We conducted a prospective survey of members of an inte-
grated managed care organization to evaluate whether socio-
demographic factors, vaccinees’ knowledge about HPV and
HPV4, and self-reported experiences at the visit where the first
HPV4 dose was administered, predicted series completion.
Although the studies described used data available in the
medical chart, to our knowledge ours is the first study to survey
females at the time of HPV4 vaccine initiation and then follow
them to assess factors that predict completion of the series.
Methods
Study population and data sources
Between February and September 2008, we identified all
females ages 11e26 who received their first HPV4 dose at Kaiser
Permanente Northwest. In this prospective cohort study, we
contacted these potential survey respondents within a week of
their first HPV4 dose via a series of mailings. A detailed report of
the survey methods was previously published [15]. To summa-
rize: we mailed 11- to 26-year-old females a postcard inviting
participation in a web version of the survey, followed by paper
versions of the survey sent one and two weeks later; parents of
females 11e17 years of age were sent a passive permission letter
before the first mailing to their daughter. Completed surveys
were generally returned to us within one to two weeks of our
initial mailing. We obtained data on survey respondents and
nonrespondents from the EMR, including receipt of HPV4 doses
and visit and demographic characteristics. The EMR includes
data from Oregon’s statewide immunization information system,
so vaccines administered outside the health plan were captured.
We followed everyone for one year after the date of their first
HPV4 dose, with completion of the vaccine series defined as
receiving all three HPV4 doses in that time.
We measured SES in two ways: (1) Medicaid receipt in the last
year and (2) the Diez-Roux Index score, a measure of neighbor-
hood SES, based on census tract of residence at the time of the
first HPV4 dose [16,17]. Neighborhood SES was dichotomized
into “high” and “low” SES census tracts based on quintiles of the
distribution of Index values. We created a measure of care utili-
zation frequency by extracting the number of health plan visits in
the year before the first HPV4 dose. Consistent with our health
plan data for this period, race/ethnicity was unknown or missing
for about half of the study population, so we did not include this
measurement in our analyses. Visit characteristics included
department of the provider who authorized the vaccination, and
number of other vaccines concomitantly administered with
HPV4 at the index visit.
The survey consisted of multiple-choice and Likert-scale
questions about experiences at the visit, challenges to making
or keeping the index appointment, knowledge and attitudes
about HPV and HPV4, potential barriers to vaccination, and acute
adverse events occurring after the first HPV4 dose. Although
many of the data elements collected in the survey fall under the
Health Belief Model’s framework [18], other elements, more
specific to vaccine receipt, were constructed based on previous
research on factors associated with adolescents’ compliance with
recommended vaccinations [2,3,5,7,14]. Our analyses do not
include survey data on difficulties experienced in making or
getting to the index appointment because respondents
uniformly reported having no such problems.
Analyses
We describe completion of the vaccine series among all
females who were invited to participate in the survey, comparing
those who participated to those who did not. All further analyses
were limited to survey respondents who were continuously
enrolled in the health plan in the year after the first HPV4 dose.
Within this population, we evaluated the extent to which
demographic factors, visit characteristics, care utilization in the
year prior to the first HPV4 dose, and survey responses were
associated with completion of the vaccine series (receipt of all
three HPV4 doses in the year after the first dose), using chi-
square tests to assess differences in series completion.
We conducted factor analyses to assess the need to create
composite measures for survey responses related to: (1) infor-
mation given by the provider about HPV; (2) information given
by the provider about HPV4; (3) the respondent’s knowledge
about HPV; and (4) perceived risk of cervical cancer. As indicated
by the factor analyses, we created a measure summarizing
respondents’ knowledge about HPV (number of correct
responses to six true/false questions such as “HPV is a rare
infection”), and perceived risk (sum of Likert-scale responses
asking agreement with statements that subjects themselves
were too healthy or too young to get cervical cancer).
We conducted log-binomial regression analyses of the asso-
ciations between each factor of interest and completion of the
full HPV4 series. All analyses were adjusted for age based on an
a priori assumption that factors of importance might differ
between age groups [15]. Age was categorized as 11e15, 16e20,
or 21e26 years old at the time of the first HPV4 dose, based on
preliminary analyses of series completion by age. Because this
was a cohort study, and HPV4 series completion occurred in
>10% of the survey population, we calculated adjusted relative
risks using PROC GENMOD [19,20]. Any factors that were
significant in the bivariate analyses, as well as age, were included
in a final multivariate logistic regression model. In a subanalysis
including only those survey respondents who reported pain at
the first HPV4 dose (n ¼ 610; 78% of the study population), we
examined the relationship between extent of pain and subse-
quent series completion. All analyses were conducted in SAS
version 9.2. The study was approved by the Institutional Review
Board at Kaiser Permanente Northwest.
Results
Of 3,490 survey invitations sent to females who initiated the
HPV4 vaccine series from February to September 2008, 134 (4%)
were returned as undeliverable. Of the remaining 3,356 survey
invitees, there were 899 (27%) respondents. Completion of the
vaccine series was significantly higher among the 899 survey
invitees who responded to the survey (50%) than among the
 R. Gold et al. / Journal of Adolescent Health 52 (2013) 427e432 429

Table 1
Association between sociodemographic and visit characteristics and quadrivalent human papillomavirus vaccine (HPV4) series completion among survey respondents
(N ¼ 786)

Variable Categories Na Percent who completed Relative risk (RR)

the series

Total 786 50
Age (years) 11e15 418 55 1.37 (1.14, 1.65)
16e20 228 40 1 (reference)
21e26 140 52 1.31 (1.04, 1.64)
RRs adjusted for age
Neighborhood SES Low 413 47 1 (reference)
High 359 53 1.14 (.99, 1.31)
Received Medicaid in last year? No 771 50 1 (reference)
Yes 15 53 1.05 (.65, 1.68)
Number of health plan visits in last year 0 170 46 1 (reference)
1 165 51 1.11 (.89, 1.38)
2 or more 451 51 1.13 (.94, 1.36)
Provider specialty Pediatrics 462 52 1 (reference)
Internal medicine 77 47 1.03 (.76, 1.39)
Ob-gyn 78 53 1.13 (.82, 1.54)
Family practice 168 46 .97 (.78, 1.20)
Number of other vaccines at index visit No other vaccines 267 53 1 (reference)
1 or more 519 49 .85 (.73, .99)

Bold text indicates statistical significance or borderline significance.

a
Those missing data in a given category are not included.

2,547 who did not (34%); p < .0001. Females 13e17 years of age
were less likely to respond to the survey than were those ages
11e12 or 18e26 years.
Of the 899 survey respondents, 786 (87%) were continuously
enrolled in the Kaiser Permanente Northwest health plan in the
year after their first HPV4 dose; this was our study population for all
analyses unless otherwise noted. Half of these females completed
the three-dose HPV4 series within a year of the first HPV4 dose
(Table 1). In chi-square testing, series completion differed signifi-
cantly by age, with the lowest completion in 16e20 year olds (40%)
compared to 11e15 year olds (55%) and 21e26 year olds (52%). After
adjusting for age, series completion among females who received
concomitant vaccines at the index visit was significantly lower
compared with those who received HPV4 alone (RR ¼ .85, 95%
confidence interval ¼ .73e.99). No other significant differences
in series completion were associated with demographic charac-
teristics, prior care utilization, or visit characteristics.
According to the survey responses, several factors were
significantly associated with series completion after adjusting for
age (Table 2). Females who reported that HPV vaccination was
the main reason they came to the clinic on the day of the first
HPV4 dose were significantly more likely to complete the series.
Those reporting that their provider discussed the need to return
to the clinic for more HPV4 doses, and those who could correctly
identify the total number of doses in the HPV4 series, also were
more likely to complete the series.
Those reporting having ever seen an advertisement for HPV4
were significantly less likely to complete the series. We explored
whether persons who initiated the HPV4 series after seeing the
advertisement might have lost motivation when they learned
that several shots were required, but we found no significant
correlation between having seen the advertisement and report-
ing that the main reason for the index visit was to get HPV4
(results not shown).
Those who reported experiencing bruising or swelling at the
time of the first HPV4 dose were significantly less likely to
complete the series compared with those who did not. However,
those reporting pain at the first HPV4 dose did not have signifi-
cantly lower series completion in our study population
compared to those who did not report pain (Table 2). Of the 610
(78%) survey respondents who reported any pain at the first
HPV4 dose, 605 provided information on the level of pain (very
mild [16% of those reporting pain], mild [36%], moderate [37%],
severe [9%], very severe [2%]); reported level of pain was not
significantly associated with vaccine series completion in this
group. Reported syncope and dizziness at the time of the first
HPV4 dose were not associated with series completion.
In multivariate models including age and all variables that
were significant in bivariate analyses, females ages 16e20 at the
first HPV4 dose were significantly less likely to complete the
series (Table 3). Ability to correctly identify the number of shots
in the vaccine series remained a significant predictor of series
completion, and having seen an HPV4 advertisement remained
negatively associated with series completion. Whether the
clinician noted the need to return for more shots trended
towards significance (RR ¼ 1.33, 95% CI ¼ 1.00e1.77), as did
reported bruising at the time of the first HPV4 dose (RR ¼ .88,
95% CI ¼ .70e1.00).
Discussion
To our knowledge, this is the first study to survey HPV4
recipients to prospectively evaluate demographic factors, self-
reported knowledge about HPV and HPV4, and self-reported
experiences at the time of the first receipt of HPV4, and to
assess their association with completion of the three-dose HPV4
series, in an integrated managed care setting. We found that
patients age 16e20 years at the time of the first HPV4 dose and
having seen an HPV4 advertisement were associated with lower
vaccine series completion; ability to identify the number of shots
in the series was associated with greater series completion.
Those whose clinician noted the need to return for more shots
trended toward higher completion, and those reporting bruising
from the first HPV4 dose toward lower completion.
Some of these findings were consistent with previous
studies of the association between HPV4 series completion
and factors such as patient age, provider specialty, provider
recommendation of vaccination [13,14], and parents’ reports of
430 R. Gold et al. / Journal of Adolescent Health 52 (2013) 427e432

Table 2
Association between survey responses, and quadrivalent human papillomavirus vaccine (HPV4) series completion among survey respondents (N ¼ 786)

Variable Categories Na Percent completed Bivariate RRs

the series (all adjusted for age)

Experienced adverse reaction?

Reported bruising or swelling after index shot No 598 52 1 (reference)
Yes 188 43 .82 (.68, .98)
Reported pain after index shot No 176 47 1 (reference)
Yes 610 51 1.31 (.94, 1.81)
Severity of pain (among those reporting pain) Moderate/severe/very severe 287 49 1 (reference)
Very mild or mild 318 53 1.12 (.96, 1.31)
Reported syncope/dizziness at index shot No 669 49 1 (reference)
Yes 116 53 1.05 (.88, 1.27)
Reported that provider discussed .
HPV infection No 269 52 1 (reference)
Yes 511 49 .96 (.84, 1.11)
Benefits of HPV4 No 189 50 1 (reference)
Yes 593 50 1.06 (.90, 1.25)
Possible side effects No 301 52 1 (reference)
Yes 481 49 .95 (.83, 1.09)
Genital warts No 628 50 1 (reference)
Yes 153 50 1.02 (.86, 1.22)
Pap tests No 565 52 1 (reference)
Yes 215 44 .91 (.75, 1.09)
Coming back for more HPV4 shots No 105 35 1 (reference)
Yes 674 52 1.55 (1.18, 2.03)
Knowledge about HPV and HPV4
Correct answers to 6 questions about HPV, 0e1 171 51 1 (reference)
HPV vaccine 2e4 360 51 1.01 (.85, 1.20)
5e6 249 47 .99 (.80, 1.22)
Correctly identified total number of HPV4 No 135 37 1 (reference)
shots in the series Yes 640 53 1.49 (1.18, 1.88)
Risk beliefs
I am too healthy/young to be at cervical Disagree/strongly disagree 68 53 1 (reference)
cancer risk Agree/strongly agree/neither 704 50 .99 (.78, 1.26)
agree nor disagree
Believe that cervical cancer is easily cured Disagree/strongly disagree 608 50 1 (reference)
Agree/strongly agree/neither 167 49 1.02 (.86, 1.20)
agree nor disagree
Family history of cervical cancer No 688 51 1 (reference)
Yes 87 43 .82 (.63, 1.06)
Reason for visit
Ever saw advertisement for the vaccine No 208 60 1 (reference)
Yes 563 47 .81 (.70, .94)
Reason came to clinic on day of index shot HPV4 150 59 1 (reference)
Check up/other 632 48 .83 (.71, .97)
Worried/scared about getting the index shot No 426 50 1 (reference)
Yes 353 50 .93 (.80, 1.08)

Bold text indicates statistical significance or borderline significance.

a
Missing values not displayed; each category totals 786 when persons missing data are included, except the severity of pain categories which would equal 610 when
persons missing data are included.

their children’s experience [21]. However, some of our findings
diverged from the findings in these earlier studies. In another
integrated care population (N ¼ 29,598), Chao et al [14] found
that completion rates were lower in younger girls 9e17 years of
age (42%) compared with adults 18e26 years of age (47%), with
the lowest completion rates occurring at age 17. We found that
females ages 16e20 were significantly less likely to complete the
series than those ages 11e12 or 18e26. This age group may be
less likely to complete than younger females who are brought in
by their parents and older females who are motivated to get
vaccinated on their own. In contrast, neither Neubrand et al [13]
nor Widdice et al [10] found that age was associated with series
completion. Chao et al [14] found that persons with a pediatri-
cian as their primary care provider were less likely to complete
the vaccine series; we found no relationship between provider
department and completion. The differences between our results
and those of Chao and colleagues may be due to their larger
sample size and consequent ability to detect smaller differences.
In our study, females who reported that their provider had
mentioned the need to return for more doses, and those who
were able to correctly identify the number of shots in the HPV4
series, were more likely to complete the series. These two factors
were significantly (p < .0001) correlated with each other;
however, both remained statistically significant or were of
borderline significance when included in the multivariate model,
suggesting that providers mentioning the need for further doses
and knowledge of the number of doses in the HPV4 series by
vaccinees affect series completion independently. This finding
suggests that providers should inform their patients that addi-
tional vaccines are needed and verify that the patient knows
exactly how many more doses are needed. Such clinic-based
education approaches have been proven effective at improving
vaccination coverage among pediatric patients in some settings;
other evidence-based approaches, such as client reminders and
recalls, have also been shown to be effective and should be
considered [22].
 R. Gold et al. / Journal of Adolescent Health 52 (2013) 427e432 431

Table 3 a previous study in an integrated care population. Chao et al report

Multivariate regressions of factors associated with completion of the quadriva-
lent human papillomavirus vaccine (HPV4) series
Adjusteda RR þ 95% CI
Entire study population
(N ¼ 786)
that higher neighborhood educational attainment and Medicaid
receipt (among younger persons only) were associated with series
completion [14]. Other studies have not consistently reported series
completion associated with SES or Medicaid enrollment [10,13].

Age
11e15 years 1.41 (1.16, 1.70)
16e20 years 1.00 (reference)
21e26 years 1.28 (1.02, 1.61)
Provider noted need for more shots
Yes 1.33 (1.00, 1.77)
No 1.00 (reference)
Could identify number shots in the vaccine series
Yes 1.38 (1.08, 1.76)
No 1.00 (reference)
Saw the ad for the vaccine
Yes .77 (.67, .89)
No 1.00 (reference)
HPV vaccination was the main reason for visit
No .88 (.75, 1.02)
Yes 1.00 (reference)
Received other vaccines at index visit
Yes 1.00 (.86, 1.17)
No 1.00 (reference)
Reported bruising or swelling after index shot
Yes .84 (.70, 1.00)
No 1.00 (reference)
Bold text indicates statistical significance or borderline significance.
a
Adjusted for all other variables shown here.
Females who reported that HPV vaccination was the main
reason they came to the clinic on the day of the first HPV4 dose
were more likely to complete the series, in age-adjusted
regression analyses. Although this lost statistical significance in
the multivariate model, the finding is similar to that of Neubrand
et al [13], who found that persons whose primary reason for the
visit was to get an HPV4 (as determined through chart abstrac-
tion) were more likely to complete the series. Additionally, we
found that females who received HPV4 only at their index visit
were more likely to complete the series than those who received
multiple concomitant vaccines at their index visit. This suggests
that providers should emphasize the importance of vaccine
series completion, especially when administering multiple
vaccines at the same visit as the first HPV4 dose.
In all analyses, females who had ever seen an advertisement
for HPV4 vaccine were significantly less likely to complete the
series. We assessed whether females motivated to get an HPV4
vaccine because they saw an advertisement might not know that
the vaccine required more than one shot, and therefore might
have lost motivation to return for additional shots; our analyses
did not support this hypothesis.
We found no relationship between self-reported local reac-
tions, pain, or syncope and dizziness following the first HPV4
dose, and completion of the vaccine series in adjusted models.
We also found no relationship between series completion and
degree of pain experienced among the 610 females reporting
pain at the first HPV4 dose. Reported bruising trended toward
association with decreased series completion, but overall our
findings are congruent with those of Reiter et al, who found that
parents’ reports of their adolescent’s pain at first HPV4 dose was
not associated with series completion [21].
We also found no statistically significant relationships between
neighborhood SES or Medicaid receipt and completion of the HPV4
series, though our results trended in the same direction as
Limitations
Our study population consisted of females responding to
a survey with a 27% response rate. Although not ideal, this rate is
consistent with other survey research in this age group [23e25].
Females completing the survey were more likely than those who
did not complete the survey to be higher SES, ages 11e12 or
18e26, and to have received no other vaccinations at the index
encounter. Thus, our study population is not perfectly repre-
sentative of the health plan population; survey respondents may
have been better informed and more engaged in their health care
than those who did not participate.
We selected factors to examine the association with vaccine
series completion based on previous findings on barriers to
adolescent vaccination [2,3,5e7,14,26e35]. Other important
factors may not have been considered, such as provider system
characteristics, other aspects of the providerepatient relationship,
and the role of parents. Adverse events occurring more than seven
days after the first HPV4 dose would not have been identified in
our EMR review; adverse events occurring after the survey
was completed would not have been reported. We were also
unable to evaluate the association between race/ethnicity and
HPV4 series completion because of incomplete data. Neubrand
et al [13] and Widdice et al [10] found that white persons were
significantly more likely to complete the series; Chao et al [14] also
reported significant differences in series completion associated
with race/ethnicity, with these data available for 59% of their
population. We had race data on only 49% of our study population,
and these data indicated that 81% of the group was white, a much
less diverse population than the population studied by Chao et al
[14]. Last, these data were collected in 2008, when the HPV4
vaccine was newly available; factors behind vaccine series
completion may have changed since the time of data collection.
Despite these limitations, our study provides important
information to help identify which patient and provider factors
predict completion of the HPV4 series in a stably insured
managed health plan population. The 28% to 70% HPV4 series
completion rates reported nationally [10e14] (50% among our
survey respondents) needs improvement, and our results
suggest some relatively easy to implement and low-cost strate-
gies that providers can use to promote completion of the HPV4
series. Notably, providers should tell their patients at the time of
the first HPV4 dose that two more shots are needed, and explain
the intervals at which they need to return to complete the series.
Providers should also assess whether their patients and/or
parents understood that information and whether patients can
correctly identify the number of shots needed to complete the
series. Future research should include randomized trials to assess
whether, and how much, this and other simple interventions
such as reminder calls can improve HPV4 vaccine completion.
References
[1] Markowitz LE, Dunne EF, Saraiya M, et al. Quadrivalent human papillo-
mavirus vaccine: Recommendations of the Advisory Committee on
Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 2007;56:
1e24.
432 R. Gold et al. / Journal of Adolescent Health 52 (2013) 427e432
[2] Dempsey AF, Davis MM. Overcoming barriers to adherence to HPV vacci-
nation recommendations. Am J Manag Care 2006;12:S484e91.
[3] Humiston SG, Rosenthal SL. Challenges to vaccinating adolescents: Vaccine
implementation issues. Pediatr Infect Dis J 2005;24:S134e40.
[4] Lee GM, Lorick SA, Pfoh E, et al. Adolescent immunizations: Missed
opportunities for prevention. Pediatrics 2008;122:711e7.
[5] Lehmann C, Benson PA. Vaccine adherence in adolescents. Clin Pediatr
(Phila) 2009;48:801e11.
[6] Middleman AB. New adolescent vaccination recommendations and how to
make them "stick". Curr Opin Pediatr 2007;19:411e6.
[7] Oster NV, Phillips-Tangum CA, Averhoff F, et al. Barriers to adolescent
immunization: A survey of family physicians and pediatricians. J Am Board
Fam Pract 2005;18:13e9.
[8] Slonim AB, Roberto AJ, Downing CR, et al. Adolescents’ knowledge, beliefs,
and behaviors regarding hepatitis B: Insights and implications for
programs targeting vaccine-preventable diseases. J Adolesc Health 2005;
36:178e86.
[9] Whitehead SJ, Cui KX, De AK, et al. Identifying risk factors for under-
immunization by using geocoding matched to census tracts: A statewide
assessment of children in Hawaii. Pediatrics 2007;120:e535e42.
[10] Widdice LE, Bernstein DI, Leonard AC, et al. Adherence to the HPV vaccine
dosing intervals and factors associated with completion of 3 doses. Pedi-
atrics 2011;127:77e84.
[11] Tan W, Viera AJ, Rowe-West B, et al. The HPV vaccine: Are dosing
recommendations being followed? Vaccine 2011;29:2548e54.
[12] Centers for Disease Control and Prevention. National, state, and local area
vaccination coverage among adolescents aged 13-17 years e United States,
2009. MMWR Morb Mortal Wkly Rep 2010;59:1018e23.
[13] Neubrand TP, Breitkopf CR, Rupp R, et al. Factors associated with
completion of the human papillomavirus vaccine series. Clin Pediatr (Phila)
2009;48:966e9.
[14] Chao C, Velicer C, Slezak JM, et al. Correlates for completion of 3-dose
regimen of HPV vaccine in female members of a managed care organiza-
tion. Mayo Clin Proc 2009;84:864e70.
[15] Naleway AL, Gold R, Drew L, et al. Reported adverse events in young
women following quadrivalent human papillomavirus vaccination.
J Women’s Health 2012;21:425e32.
[16] Chichlowska KL, Rose KM, Diez-Roux AV, et al. Individual and neighbor-
hood socioeconomic status characteristics and prevalence of metabolic
syndrome: The Atherosclerosis Risk in Communities (ARIC) Study. Psy-
chosom Med 2008;70:986e92.
[17] Nordstrom CK, Diez Roux AV, Jackson SA, et al. The association of personal
and neighborhood socioeconomic indicators with subclinical cardiovas-
cular disease in an elderly cohort. The cardiovascular health study. Soc Sci
Med 2004;59:2139e47.
[18] Janz NK, Becker MH. The Health Belief Model: A decade later. Health Educ
Behav 1984;11:1e47.
[19] Lindquist K. How can I estimate relative risk in SAS using proc
GENMOD for common outcomes in cohort studies? http://www.ats.ucla.
edu/stat/sas/faq/relative_risk.htm [serial online] 2011; Accessed March
13, 2011.
[20] Kleinman LC, Norton EC. What’s the risk? A simple approach for estimating
adjusted risk measures from nonlinear models including logistic regres-
sion. Health Serv Res 2009;44:288e302.
[21] Reiter PL, Brewer NT, Gottlieb SL, et al. How much will it hurt? HPV vaccine
side effects and influence on completion of the three-dose regimen.
Vaccine 2009;27:6840e4.
[22] Briss PA, Rodewald LE, Hinman AR, et al. Reviews of evidence regarding
interventions to improve vaccination coverage in children, adolescents,
and adults. The Task Force on Community Preventive Services. Am J Prev
Med 2000;18(1 Suppl):97e140.
[23] Sax LJ, Gilmartin SK, Bryant AN. Assessing response rates and nonresponse
bias in web and paper surveys. Research in Higher Education 2003;44:
409e32.
[24] Sax LJ, Gilmartin SK, Lee JJ, et al. Using web surveys to reach community
college students: An analysis of response rates and response bias. Commun
Coll J Res Practice 2008;32:712e29.
[25] Shiu TH, Fan X. Comparing response rates from web and mail surveys: A
meta-analysis. Field Methods 2010;20:249e71.
[26] Song JY, Park CW, Jeong HW, et al. Effect of a hospital campaign for
influenza vaccination of healthcare workers. Infect Control Hosp Epidemiol
2006;27:612e7.
[27] Lin WC, Ball C. Factors affecting the decision of nursing students in
Taiwan to be vaccinated against hepatitis B infection. J Adv Nurs 1997;25:
709e18.
[28] Weingarten S, Riedinger M, Bolton LB, et al. Barriers to influenza vaccine
acceptance. A survey of physicians and nurses. Am J Infect Control 1989;
17:202e7.
[29] Main KM, Jorgensen JT, Hertel NT, et al. Automatic needle insertion
diminishes pain during growth hormone injection. Acta Paediatr 1995;84:
331e4.
[30] France JL, France CR, Himawan LK. A path analysis of intention to redonate
among experienced blood donors: An extension of the theory of planned
behavior. Transfusion 2007;47:1006e13.
[31] France CR, France JL, Roussos M, et al. Mild reactions to blood donation
predict a decreased likelihood of donor return. Transfus Apher Sci 2004;30:
17e22.
[32] Bonk VA, France CR, Taylor BK. Distraction reduces self-reported physio-
logical reactions to blood donation in novice donors with a blunting coping
style. Psychosom Med 2001;63:447e52.
[33] Cable R, Trouern-Trend JJ, Badon SJ. The effect of syncopal reactions on
subsequent blood donations. Transfusion 1999;39:114S.
[34] Kahn JA, Rosenthal SL, Hamann T, Bernstein DI. Attitudes about human
papillomavirus vaccine in young women. Int J STD AIDS 2003;14:
300e6.
[35] Davis K, Dickman ED, Ferris D, Dias JK. Human papillomavirus vaccine
acceptability among parents of 10- to 15-year-old adolescents. J Low Genit
Tract Dis 2004;8:188e94.