- Email: [email protected]
Ethnic Groups: Data From a National Survey
Journal of Adolescent Health xxx (2019) 1e8
www.jahonline.org Original article
Declines in Vaccine-Type Human Papillomavirus Prevalence in Females Across Racial/Ethnic Groups: Data From a National Survey Nancy M. McClung, Ph.D., R.N. a, b, *, Rayleen M. Lewis, M.P.H. a, c, Julia W. Gargano, Ph.D. a, Troy Querec, Ph.D. d, Elizabeth R. Unger, Ph.D., M.D. d, and Lauri E. Markowitz, M.D. a a
National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention (CDC), Atlanta, Georgia Epidemic Intelligence Service, Centers for Disease Control and Prevention, Atlanta, Georgia c Synergy America, Inc., Duluth, Georgia d National Center for Emerging and Zoonotic Infectious Diseases, Centers for Disease Control and Prevention (CDC), Atlanta, Georgia b
Article history: Received March 15, 2019; Accepted July 8, 2019 Keywords: Human papillomavirus; HPV vaccine; Public health
A B S T R A C T
Purpose: To monitor human papillomavirus (HPV) vaccine impact in the U.S., we evaluated quadrivalent vaccine (4vHPV)-type prevalence among females aged 14e34 years in the prevaccine (2003e2006) and vaccine (2013e2016) eras overall and by race/ethnicity in the National Health and Nutrition Examination Survey. Methods: We analyzed HPV DNA prevalence in self-collected cervicovaginal specimens, demographic characteristics, sexual behavior, and self-reported/parent-reported vaccination status. We compared prevaccine to vaccine era 4vHPV-type prevalence, using unadjusted and adjusted prevalence ratios (PR and aPR) and 95% confidence intervals (CIs). PRs were calculated by race/ ethnicity (non-Hispanic white [NHW], non-Hispanic black [NHB], and Mexican American [MA]). Overall aPRs were adjusted for race/ethnicity, lifetime sex partners, and poverty. Results: Overall, 4,674 females had HPV typing results; 3,915 reported NHW, NHB, or MA race/ ethnicity. Vaccination coverage of 1 dose was 53.9% among 14- to 19-year-olds (NHW 52.6%, NHB 58.1%, and MA 59.5%) and 51.5% among 20- to 24-year-olds (NHW 58.8%, NHB 45.0%, MA 33.8%). Among 14- to 19-year-olds, 4vHPV-type prevalence decreased overall (11.5% to 1.8%; aPR ¼ .14 [CI: .08e.24]) and in NHW (PR ¼ .14 [CI: .06e.29]), NHB (PR ¼ .26 [CI: .12e.54]), and MA (PR ¼ .13 [CI: .03e.53]). In 20- to 24-year-olds, 4vHPV-type prevalence decreased overall (18.5% to 5.3%; aPR ¼ .29 [CI: .15e.56]) and in NHW (PR ¼ .27 [CI: .11e.67]) and NHB (PR ¼ .38 [CI: .18e.80]). No significant declines were observed in older age groups. Conclusions: Within 10 years of vaccine introduction, 4vHPV-type prevalence declined 86% among 14- to 19-year-olds, with declines observed in NHW, NHB, and MA females, and 71% among 20- to 24-year-olds, with declines in NHW and NHB females. These extraordinary declines should lead to substantial reductions in HPV-associated cancers. Ó 2019 Published by Elsevier Inc. on behalf of Society for Adolescent Health and Medicine.
IMPLICATIONS AND CONTRIBUTION
This study provides updated human papillomavirus vaccine-type prevalence estimates through 2016 in females and reports new findings by race/ethnicity. Since vaccine introduction, a decline in prevalence was observed in all reported race/ethnicities among 14- to 19-yearolds, with an 86% decline overall, and a 71% decline among 20- to 24-yearolds.
Conflicts of interest: The authors have no conflicts of interest to disclose. Disclaimer: 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. * Address correspondence to: Nancy M. McClung, Ph.D., R.N., Centers for Disease Control and Prevention, 1600 Clifton Rd. NE, MS A34, Atlanta, GA 30329. E-mail address: [email protected] (N.M. McClung). 1054-139X/Ó 2019 Published by Elsevier Inc. on behalf of Society for Adolescent Health and Medicine. https://doi.org/10.1016/j.jadohealth.2019.07.003
2
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
In the U.S., routine vaccination against human papillomavirus (HPV) has been recommended since 2006 for females and 2011 for males [1]. Routine vaccination is recommended at age 11 or 12 years and for those not vaccinated at routine ages, through age 26 years for females, 21 years for males, and 26 years for special populations of males. Three HPV vaccines (i.e., bivalent, quadrivalent [4vHPV], and nonavalent [9vHPV] vaccines) are licensed for use in the U.S. From 2006 to 2015, 4vHPV was primarily used, with minimal bivalent use; 9vHPV was introduced in 2015, and since the end of 2016, only 9vHPV has been available in the U.S. [2]. 4vHPV targets HPV types 16 and 18, the causes of most HPVassociated cancers, and HPV 6 and 11, the causes of genital warts [3]. 9vHPV targets five additional cancer-causing types, HPV 31/33/45/52/58. Annual national surveys monitor HPV vaccination coverage among 13- to 17-year-olds in the U.S. Coverage among this age group has steadily increased, and in 2017, coverage of 1 dose and up-to-date coverage (two or three doses depending on age at series initiation) were 68.6% and 53.1%, respectively, among females, and 62.6% and 44.3%, respectively, among males [4]. Data from the National Health and Nutrition Examination Survey (NHANES) have been used to evaluate vaccine impact on HPV infection since vaccine introduction [5]. Within 4 years of vaccine introduction, a 56% decrease in the prevalence of 4vHPV infections was observed among 14- to 19-year-old females [6]. Most recently, within 8 years of vaccine introduction, a 71% decrease among 14- to 19-year-olds and a 61% decrease among 20- to 24-year-olds were observed [7]. The significant decrease in 4vHPV types is not only expected to lead to a decrease in HPV-associated cancers in the U.S. but also may lead to a reduction in racial/ethnic disparities in cancer incidence [8]. Non-Hispanic black (NHB) and Hispanic women have a higher cervical cancer incidence than non-Hispanic white (NHW) women [8], and NHB women have a higher cervical cancer mortality rate than NHW women [9] in part because of differences in screening participation and/or follow-up treatment [10e12]. Because HPV vaccine types 16 and 18 cause the vast majority of cervical cancers in all women, vaccination may reduce cervical cancer disparities [3]. Therefore we evaluated the impact of the U.S. vaccination program using NHANES HPV prevalence data among 14- to 34-year-old females from the four most recent vaccine era years (2013e2016) compared with the prevaccine era (2003e2006), overall and by race/ethnicity. This evaluation was limited to females because prevaccine era HPV prevalence is not available for males in NHANES. Methods
oversampled in 2003e2006, but not in 2013e2016. In 2003e2006, Mexican Americans (MAs) were oversampled. Oversampling in 2013e2016 was expanded to include all Hispanics; this group was predominantly MA. NHANES is approved by the CDC/National Center for Health Statistics Research Ethics Review Board. Study sample Female participants aged 14e34 years with adequate cervicovaginal samples were included in the overall analyses (n ¼ 4,674; Figure 1). Females who self-reported race/ethnicity as NHW, NHB, or MA were included in the race/ethnicity analyses (n ¼ 3,915). Of Hispanic females, only MA females were included to limit bias from sampling variation between 2003e2006 and 2013e2016. Other racial/ethnic groups were not analyzed separately because of small sample sizes. To achieve stable estimates, NHANES 2-year cycles were combined into 4-year eras: prevaccine era (2003e2006) and the most recent 4 years of the vaccine era (2013e2016). Study variables Demographic information including age, race/ethnicity, and poverty status (poverty level based on a ratio of household income to state poverty level). HPV vaccination history were selfreported/parent-reported during household interviews. Sexual behavior history (e.g., number of lifetime sex partners) was collected using audio computer-assisted self-interviews during examination at a MEC. Cervicovaginal samples were selfcollected at a MEC and shipped to the CDC laboratory for HPV genotyping. HPV genotyping methods have been described previously [5]. Briefly, DNA was extracted from samples and tested for 37 HPV types with a genotyping assay using nucleic acide amplification methods. HPV types were categorized as any HPV, non-4vHPV types (26/31/33/35/39/40/42/45/51/52/53/54/
Prevaccine Era 2003–2006
Vaccine Era 2013–2016
3325 females aged 14–34 intervieweda
2549 females aged 14–34 interviewedb
3210 (96.5%) received examination in MEC
2456 (96.4%) received examination in MEC
2649 (82.5%) submitted selfcollected cervicovaginal swab
2119 (86.3%) submitted selfcollected cervicovaginal swab
2587 (97.7%) samples with adequate typing results
2087 (98.5%) samples with adequate typing results
Study design NHANES is a population-based, cross-sectional survey conducted in 2-year cycles by the National Center for Health Statistics (NCHS) at the Centers for Disease Control and Prevention (CDC) to evaluate the health and nutrition of the U.S. population. Participants are randomly selected from civilian, noninstitutionalized households using a complex, multistage sampling design. After an in-home interview, they complete a series of interviews and physical examinations in a mobile examination center (MEC). Persons aged 16 years and emancipated minors are interviewed directly. Various subpopulations are oversampled in each 2-year cycle to increase the precision of estimates. Pertinent to this analysis, adolescents were
Figure 1. Participation and sample collection among females aged 14e34 yearsdNHANES, 2003e2006 and 2013e2016. MEC: medical examination center. aThe response rates during 2003e2004 and 2005e2006 among females were 79.7% and 80.9% for the interview portion, and 76.0% and 77.8% for the examination portion, respectively; bThe response rates during 2013e2014 and 2015e2016 among females were 71.4% and 62.6% for the interview portion, and 68.8% and 60.0% for the examination portion, respectively.
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
55/56/58/59/61/62/64/66/67/68/69/70/71/72/73/81/82/83/84/89/ IS39), additional types found in 9vHPV (31/33/45/52/58), and 4vHPV types (6/11/16/18). Data analysis Number of lifetime sex partners, poverty status, HPV vaccination, and HPV prevalence were described and compared across vaccine eras by age group (14e19, 20e24, 25e29, 30e34 years) overall and by race/ethnicity (NHW, NHB, and MA). Number of lifetime sex partners was categorized as 0, 1e2, or 3 partners. Poverty status was dichotomized as above or below the poverty level based on a ratio of household income to state poverty level. Two dichotomized measures of HPV vaccination status were used: receipt of 1 dose and 3 doses. All estimates were calculated by applying examination sample weights to account for unequal probabilities of selection and adjustment for nonresponse. To account for the complex survey design, variance estimates were calculated using a Taylor series linearization. Logit confidence intervals (CIs) were calculated for prevalence estimates, with a ¼ .05. Prevalence estimates with a relative standard error >30% were considered unstable. Wald chi-square tests were used to determine differences in number of lifetime sex partners, poverty status, and HPV vaccination between the prevaccine and vaccine eras. Logistic regression was used to estimate predicted marginal prevalence ratios comparing HPV prevalence in the vaccine to prevaccine era by age group. In the overall analysis, unadjusted and adjusted (a priori on race/ ethnicity, sex partners, and poverty status) prevalence ratios (PRs and aPRs) and 95% CIs were calculated. Because of convergence concerns in adjusted models, the three category number of lifetime partners variable was collapsed to two categories (0e2 and 3), and the dichotomized variable was used for adjustments. In the analyses by race/ethnicity, only unadjusted PRs were calculated because of small sample sizes. To control for the changing proportion of young women who were sexually experienced, prevalence estimates were also determined among sexually experienced 14e19 and 20- to 24-year-olds. Statistical significance was defined as p < .05 or, for a PR, as a 95% CI that does not include 1. All analyses were conducted in SAS version 9.3 (SAS Institute, Cary, NC) and SUDAAN version 11.0 (Research Triangle Institute, Research Triangle Park, North Carolina). Results Sexual behavior and poverty status Number of lifetime partners and poverty status, by age group overall and race/ethnicity, are summarized in Table 1. Significant changes in number of lifetime partners were observed in 14- to 19-year-olds overall and among NHW and NHB females. The proportion of 14- to 19-year-olds reporting zero lifetime partners increased from 46.1% in 2003e2006 to 56.2% in 2013e2016. Similar increases were observed among NHW (48.5% to 58.3%) and NHB (38.9% to 55.1%) females. No other significant differences in number of lifetime partners or poverty status were observed between the prevaccine and vaccine eras. HPV vaccination history In 2013e2016, the proportion of females with a history of 1 dose and 3 doses of HPV vaccine was 53.9% and 34.8%,
3
respectively, among 14- to 19-year-olds and 51.5% and 30.9%, respectively, among 20- to 24-year-olds and was lower in older age groups (Table 1). The median age at first vaccine dose was 12 years in all 14- to 19-year-olds and 15e16 years in all 20- to 24-year-olds. By race/ethnicity, the proportion of 14- to 19-yearold females with a history of 1 HPV vaccine dose was between 50% and 60% among all race/ethnicities (52.6% NHW, 58.1% NHB, and 59.5% MA). Among 20- to 24-year-olds, 1 dose vaccination coverage was 58.8% in NHW, 45.0% in NHB, and 33.8% in MA females. HPV prevalence by age group Among 14- to 19-year-olds, the prevalence of all HPV-type categories decreased significantly between 2003e2006 and 2013e2016 (Table 2). Prevalence of 4vHPV types was 11.5% in the prevaccine era and 1.8% in 2013e2016. After adjustment for lifetime sex partners, race/ethnicity, and poverty status, an 86% decline (aPR ¼ .14 [95% CI: .08e.24]) in 4vHPV-type prevalence was observed. Among 20- to 24-year-olds, 4vHPV-type prevalence was 18.5% in the prevaccine era and 5.3% in 2013e2016. After adjustment, a 71% decline was observed in 4vHPV-type prevalence (aPR ¼ .29 [95% CI: .15e.56]). No significant declines were observed in the other HPV-type categories among 14- to 19-year-olds and 20- to 24-year-olds after adjustment. No significant decline in 4vHPV-type prevalence was observed in older age groups. Among 30- to 34-year-olds, there were declines in any HPV (aPR ¼ .72 [95% CI: .60e.86]) and non-4vHPV-type (aPR ¼ .76 [95% CI: .63e.91]) prevalence. Prevalence of individual HPV types in 14- to 19-year-olds and 20- to 24-year-olds is reported in Supplementary Tables 1 and 2. HPV prevalence by age group and race/ethnicity The prevalence of 4vHPV types decreased significantly from the prevaccine to vaccine era among 14- to 19-year-olds in all racial/ethnic groups, from 11.3% to 1.5% in NHW (PR ¼ .14 [95% CI: .06e.29]), 17.1% to 4.4% in NHB (PR ¼ .26 [95% CI: .12e.54]), and 10.0% to 1.3% in MA (PR ¼ .13 [95% CI: .03e.53]) females (Table 3). Prevalence of HPV 31/33/45/52/58 declined among NHB (PR ¼ .41 [95% CI: .25e.67]) and MA (PR ¼ .37 [95% CI: .18e.75]), but not NHW females. Among 20- to 24-year-olds, 4vHPV-type prevalence decreased in NHW females, from 21.1% to 5.7% (PR ¼ .27 [95% CI: .11e.67]), and in NHB females, from 19.6% to 7.5% (PR ¼ .38 [95% CI: .18e.80]), but not in MA females. Among older age groups, a decline in 4vHPV-type prevalence was not observed in any racial/ethnic group, and declines in any HPV prevalence (PR ¼ .63 [95% CI: .47e.83]) and non-4vHPV-type prevalence (PR ¼ .69 [95% CI: .51e.92]) were only observed among NHW 30- to 34-year-olds. HPV prevalence among 14- to 19-year-old and 20- to 24-year-old sexually experienced females Because of the increase in proportion of 14- to 19-year-olds reporting no lifetime sex partners between the prevaccine era and 2013e2016, we also analyzed data limited to sexually experienced (report of ever having had an oral, anal, or vaginal sex partner) females in this age group and also in 20- to 24-yearolds. Among sexually experienced 14- to 19-year-olds, overall 4vHPV-type prevalence declined from 19.3% in the prevaccine era to 3.1% in the vaccine era (PR ¼ .16 [95% CI: .09e.27];
Age group, characteristics
20e24 y Lifetime sex partners 0 1e2 3 Poverty level Living below HPV vaccination 1 dose 3 doses 25e29 y Lifetime sex partners 0 1e2 3 Poverty level Living below HPV vaccination 1 dose 3 doses 30e34 y Lifetime sex partners 0 1e2 3 Poverty level Living below HPV vaccination 1 dose 3 doses
Non-Hispanic white
Non-Hispanic black
n ¼ 4,674
n ¼ 1,582
n ¼ 1,206
Mexican American n ¼ 1,127
2003e2006
2013e2016
2003e2006
2013e2016
2003e2006
2013e2016
2003e2006
2013e2016
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
n ¼ 1,363
n ¼ 783
n ¼ 397
n ¼ 205
n ¼ 465
n ¼ 186
n ¼ 409
n ¼ 196
47.0 (39.9e54.2) 35.5 (31.4e39.8) 17.5 (13.5e22.5)
50.7 (43.5e57.9) 33.0 (28.1e38.4) 16.3 (11.9e21.9)
a
a
a
46.1 (43.0e49.3) 28.2 (25.0e31.7) 25.6 (22.5e29.0)
56.2 (51.9e60.5) 25.2 (22.2e28.4) 18.6 (15.3e22.4)
48.5 (43.8e53.2) 26.2 (22.0e30.9) 25.3 (21.4e29.8)
58.3 (51.5e64.7) 23.3 (19.1e28.2) 18.4 (13.8e24.2)
38.9 (33.1e45.0) 31.2 (25.9e37.1) 29.9 (23.6e37.0)
55.1 (47.7e62.3) 20.2 (15.0e26.7) 24.7 (19.0e31.4)
23.0 (18.7e27.9)
21.6 (17.0e27.0)
15.1 (10.5e21.1)
11.2 (6.2e19.4)
43.9 (35.7e52.4)
35.0 (25.5e45.9)
41.5 (35.3e48.0)
43.1 (36.7e49.7)
-
53.9 (48.1e59.6) 34.8 (29.4e40.6)
-
52.6 (44.4e60.6) 33.8 (26.7e41.7)
-
58.1 (49.8e66.0) 39.8 (31.1e49.1)
-
59.5 (50.2e68.2) 37.4 (28.1e47.7)
n ¼ 432
n ¼ 413
n ¼ 179
n ¼ 147
n ¼ 110
n ¼ 93
b
c
c
n ¼ 107
n ¼ 74
b
8.6 (5.2e14.0) 30.7 (26.4e35.4) 60.7 (53.7e67.2)
5.4 (3.4e8.4) 27.5 (23.1e32.4) 67.1 (62.2e71.7)
6.3 (2.3e15.6) 29.3 (24.2e35.0) 64.5 (56.2e71.9)
3.1 (1.0e8.7) 27.3 (20.5e35.5) 69.6 (62.1e76.2)
4.6 (1.6e12.8) 16.3 (9.7e26.1) 79.1 (68.1e87.0)
6.1 (2.4e14.8) 20.5 (11.5e33.9) 73.4 (60.6e83.2)
12.0 (7.5e18.7) 46.5 (37.7e55.4) 41.6 (31.8e52.0)
14.9 (8.2e25.5) 38.9 (27.4e51.9) 46.2 (34.7e58.1)
25.7 (19.8e32.7)
29.3 (24.0e35.2)
18.3 (11.7e27.4)
19.8 (14.2e27.0)
43.1 (30.4e56.8)
48.0 (37.8e58.4)
51.5 (36.9e65.8)
44.0 (27.1e62.4)
-
51.5 (44.0e59.0) 30.9 (24.5e38.1)
-
58.8 (47.9e69.0) 34.8 (25.4e45.5)
-
45.0 (37.1e53.1) 26.0 (17.0e37.5)
-
33.8 (21.4e49.0) 17.3 (10.5e27.0)
n ¼ 403
n ¼ 447
n ¼ 190
n ¼ 150
n ¼ 76
n ¼ 91
n ¼ 104
n ¼ 71
5.0 (2.9e8.3) 21.8 (15.9e29.1) 73.2 (66.4e79.0)
4.4 (2.6e7.4) 26.7 (21.2e32.9) 69.0 (63.1e74.3)
3.2 (1.3e7.7)b 18.6 (12.8e26.2) 78.2 (71.1e84.0)
4.8 (2.3e9.9)b 26.5 (17.9e37.2) 68.7 (58.7e77.2)
3.8 (.8e16.8)c 14.4 (7.4e26.2)b 81.8 (70.1e89.7)
5.2 (1.7e14.7)c 12.9 (8.7e18.9) 81.9 (74.2e87.7)
5.7 (1.8e17.0)c 44.4 (29.5e60.4) 49.9 (34.5e65.2)
4.9 (1.2e17.8)c 29.7 (18.0e44.9) 65.4 (52.6e76.3)
16.3 (12.5e21.0)
20.8 (16.6e25.7)
11.9 (7.6e17.9)
14.4 (9.0e22.3)
30.7 (20.6e43.1)
34.7 (24.1e47.1)
30.5 (22.9e39.3)
38.4 (21.6e58.6)
-
33.3 (28.2e38.9) 24.8 (19.9e30.6)
-
37.5 (29.6e46.2) 30.1 (22.8e38.5)
-
27.4 (19.2e37.4) 14.3 (8.1e24.0)
-
26.8 (18.3e37.5) 13.9 (7.8e23.6)
n ¼ 389
n ¼ 444
n ¼ 165
n ¼ 149
n ¼ 90
n ¼ 95
n ¼ 97
n ¼ 69
1.6 (.6e4.6)c 24.8 (20.5e29.6) 73.6 (68.2e78.4)
2.7 (1.2e6.2)b 23.5 (18.7e29.1) 73.8 (68.4e78.6)
1.0 (.1e7.0)c 20.6 (14.9e27.6) 78.5 (71.8e83.9)
2.4 (.6e8.7)c 21.6 (13.9e32.2) 75.9 (65.4e84.1)
1.3 (.2e9.4)c 8.5 (3.2e20.5)b 90.2 (78.3e95.9)
1.3 (.2e9.8)c 6.3 (2.1e17.5)c 92.4 (80.9e97.2)
3.0 (.7e12.9)c 55.2 (42.4e67.3) 41.8 (29.7e55.0)
4.8 (1.4e15.0)c 37.7 (25.2e52.1) 57.6 (43.1e70.8)
16.0 (12.7e20.1)
21.3 (16.9e26.4)
9.6 (5.8e15.5)
13.0 (7.5e21.4)
25.1 (18.9e32.6)
36.1 (26.5e46.8)
36.4 (27.5e46.4)
39.3 (26.8e53.5)
-
14.3 (10.7e18.7) 9.5 (6.5e13.7)
16.8 (11.6e23.7) 13.1 (8.5e19.7)
-
16.8 (10.3e26.2) 7.3 (3.7e13.8)b
-
-
All percentages are weighted. CI ¼ confidence interval; HPV ¼ human papillomavirus. a p < .05 based on Wald chi-square test comparing 2013e2016 to 2003e2006. b Relative standard error >30% and 50%. c Relative standard error > 50%.
7.0 (2.6e17.4)b 2.1 (.3e13.0)c
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
14e19 y Lifetime sex partners 0 1e2 3 Poverty level Living below HPV vaccination 1 dose 3 doses
Overall
4
Table 1 Lifetime sex partners, poverty status, and vaccination history among females aged 14e34 years, by age group and race/ethnicitydNHANES 2003e2006 and 2013e2016
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
5
Table 2 HPV prevalence among females aged 14e34 years, by age groupdNHANES 2003e2006 and 2013e2016 Age group/HPV types
14e19 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 20e24 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 25e29 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 30e34 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV
PR (95% CI)
aPRa (95% CI)
Prevaccine era 2003e2006 % (95% CI)
Vaccine era 2013e2016 % (95% CI)
32.9 31.2 8.4 11.5
(29.5e36.5) (27.9e34.8) (6.6e10.6) (9.1e14.4)
21.6 21.0 3.9 1.8
(18.5e25.0) (18.0e24.4) (2.5e5.9) (1.1e2.9)
.66 .67 .46 .15
(.55e.79)b (.56e.81)b (.29e.74)b (.09e.26)b
.85 .89 .69 .14
(.72e1.01) (.74e1.07) (.39e1.20) (.08e.24)b
53.7 50.7 16.5 18.5
(45.9e61.4) (43.4e58.0) (11.3e23.4) (14.9e22.8)
54.6 54.4 13.3 5.3
(48.0e61.0) (47.9e60.8) (9.5e18.2) (2.9e9.4)
1.01 1.07 .80 .29
(.84e1.22) (.89e1.29) (.50e1.30) (.16e.53)b
1.01 1.08 .92 .29
(.87e1.18) (.92e1.26) (.56e1.50) (.15e.56)b
46.8 43.8 10.8 11.8
(42.9e50.8) (38.9e48.9) (7.3e15.7) (8.8e15.6)
43.5 41.8 11.5 8.0
(38.8e48.3) (36.8e46.9) (8.2e15.9) (5.6e11.4)
.93 .95 1.06 .68
(.81e1.07) (.81e1.12) (.65e1.75) (.43e1.07)
.92 .95 .92 .61
(.80e1.06) (.80e1.12) (.53e1.58) (.36e1.04)
47.9 44.5 9.8 9.5
(42.3e53.5) (39.1e50.1) (7.1e13.5) (6.7e13.2)
35.4 34.2 6.2 6.5
(30.4e40.7) (29.1e39.7) (3.7e10.3) (4.4e9.7)
.74 .77 .63 .69
(.62e.89)b (.63e.93)b (.35e1.15) (.41e1.15)
.72 .76 .58 .73
(.60e.86)b (.63e.91)b (.29e1.17) (.41e1.32)
All percents are weighted. 4vHPV ¼ quadrivalent HPV vaccine; aPR ¼ adjusted prevalence ratio; CI ¼ confidence interval; HPV ¼ human papillomavirus; PR ¼ prevalence ratio. a Adjusted for race/ethnicity, lifetime sex partners, and poverty status. b For prevalence ratio comparing 2013e2016 to 2003e2006, confidence interval does not include 1.
Figure 2A). Declines in 4vHPV-type prevalence were observed in all racial/ethnic groups: NHW PR ¼ .14 (95% CI: .07e.28), NHB PR ¼ .30 (95% CI: .14e.64), MA PR ¼ .16 (95% CI: .04e.71). Overall, non-4vHPV-type prevalence declined from 50.1% to 41.4% (PR ¼ .83 [95% CI: .71e.97]), but no significant declines were observed in individual race/ethnicity groups. Among sexually experienced 20- to 24-year-olds, 4vHPV-type prevalence declined overall from 17.9% to 5.7% (PR ¼ .32 [95% CI: .17e.59]; Figure 2B); by race/ ethnicity, declines were only observed among NHW (PR ¼ .28 [95% CI: .11e.71]) and NHB (PR ¼ .43 [95% CI: .20e.91]) females. No significant declines were observed in non-4vHPV-type prevalence in this age group. Discussion The results of this analysis, using the most recent national data, indicate continued impact of the national HPV vaccination program on 4vHPV-type prevalence among young women in the U.S. A decade after the introduction of HPV vaccine, the prevalence of 4vHPV types declined 86% among 14- to 19-year-olds and 71% among 20- to 24-year-olds. In the most recent previous analysis, comparing 2011e2014 data to the prevaccine era, 71% and 61% decreases were observed in the same age groups [7]. Importantly, this analysis extends prior reports of NHANES data by showing that the decline is occurring across the racial/ethnic groups analyzed. Among 14- to 19-year-olds, 4vHPV-type prevalence declined 86% in NHW, 74% in NHB, and 87% in MA females. Findings by race/ethnicity are consistent with prelicensure clinical trials, which found robust immunogenicity in all represented races, ethnicities, and countries [13]. Among 20- to 24-year-olds, 4vHPV-type prevalence declined 73% in NHW and 62% in NHB females; a similar decline (63%) was observed in MA females, but because of a lower prevalence of vaccine types in 2003e2006, there was less power to detect a significant decline in 2013e2016.
A continued decline in 4vHPV-type prevalence is consistent with increasing vaccination coverage among young women in the U.S. According to the National Immunization Survey, Teen (NIS-Teen), which uses provider-verified data to monitor the U.S. vaccination program among 13- to 17-year-olds, 1 dose vaccination coverage has increased from 48.7% in 2010 to 65.1% in 2016, and three-dose vaccination coverage from 32.0% to 43.0%, among U.S. females [1,14,15]. Similar increases in 1 dose vaccination coverage have also been observed among females aged 14e19 years in NHANES, from 45.2% in 2009e2010 to 54.7% in 2015e2016 [16]. Vaccine status is self-reported in NHANES, which likely contributes to the lower coverage estimates than NIS-Teen. According to NIS-Teen, vaccination coverage has increased in all reported races/ethnicities, and in 2016, 1 dose vaccination coverage was similar by race/ethnicity among 13- to 17-year-olds [14]. In addition to increased coverage, continued declines are consistent with a higher percentage of girls receiving vaccination at the routine age (11 or 12 years), before exposure to HPV through sexual activity [17]. In NHANES 2013e2016, the median age at vaccine initiation was 12 years among 14- to 19year-olds of all race/ethnicities but was 16 years among 20- to 24-year-olds. Even greater vaccine impact on 4vHPV-type prevalence will be observed as more females who were vaccinated according to the routine recommendations age into older age groups. Although increasing vaccination coverage is consistent with continued declines in 4vHPV-type prevalence, changes in sexual behavior may also contribute to declines. In this analysis, the proportion of 14- to 19-year-olds reporting no lifetime sex partners in 2013e2016 (56.2%) was higher than in 2003e2006 (46.1%). Of note, this is the first significant change in sexual behavior among 14- to 19-year-olds tested for HPV since HPV has been monitored in NHANES. Similar increases in the proportion reporting no lifetime sex partners were observed among NHW (48.5% to 58.3%) and NHB (38.9% to 55.1%) females, but not in MA
6
Table 3 HPV prevalence among females aged 14e34 years, by age group and race/ethnicitydNHANES 2003e2006 and 2013e2016 Age group/HPV types
Non-Hispanic black n ¼ 1,206
Mexican American n ¼ 1,127 Vaccine era 2013e2016 % (95% CI)
(.56e.80)a (.56e.80)a (.25e.67)a (.12e.54)a
33.6 30.5 9.7 10.0
(27.5e40.3) (24.4e37.3) (5.9e15.6) (6.7e14.5)
21.6 20.8 3.6 1.3
(15.7e28.9) (15.3e27.7) (2.1e6.1) (.3e5.1)c
.64 .68 .37 .13
(.45e.92)a (.48e.98)a (.18e.75)a (.03e.53)a
(.76e1.12) (.80e1.17) (.43e1.38) (.18e.80)a
45.9 44.0 10.6 7.2
(32.3e60.1) (29.8e59.3) (5.4e19.7)b (2.5e19.2)c
42.4 42.4 6.4 2.7
(30.1e55.8) (30.1e55.8) (2.4e15.9)b (.6e11.4)c
.92 .96 .60 .37
(.60e1.42) (.61e1.52) (.19e1.87) (.06e2.23)
1.23 1.29 1.68 .76
(.99e1.52) (1.03e1.62)a (.74e3.80) (.31e1.81)
49.0 47.8 11.1 8.3
(39.4e58.6) (38.2e57.5) (6.0e19.6) (3.2e19.8)b
40.5 40.5 9.4 1.3
(29.6e52.5) (29.6e52.5) (3.9e21.1)b (.1e10.8)c
.83 .85 .84 .15
(.59e1.17) (.60e1.20) (.30e2.34) (.01e1.59)
.96 .94 1.17 1.10
(.80e1.14) (.78e1.14) (.49e2.81) (.48e2.54)
46.2 46.2 12.5 3.7
(38.1e54.5) (38.1e54.5) (5.9e24.3)b (1.1e11.5)c
37.1 35.5 10.0 3.9
(23.8e52.7) (21.8e52.1) (4.0e22.6)b (1.3e10.7)c
.80 .77 .80 1.06
(.52e1.23) (.48e1.22) (.27e2.42) (.22e4.97)
Vaccine era 2013e2016 % (95% CI)
(.49e.91)a (.51e.94)a (.24e1.20) (.06e.29)a
49.5 47.8 16.5 17.1
(44.8e54.2) (43.0e52.7) (13.2e20.5) (13.5e21.6)
33.0 31.9 6.8 4.4
(28.1e38.4) (27.2e37.0) (4.3e10.6) (2.1e9.0)b
.67 .67 .41 .26
1.08 1.16 .98 .27
(.86e1.36) (.91e1.47) (.49e1.98) (.11e.67)a
76.3 72.5 29.6 19.6
(64.1e85.3) (61.7e81.2) (19.1e42.8) (12.0e30.4)
70.0 70.0 22.9 7.5
(59.2e79.0) (59.2e79.0) (14.5e34.1) (4.2e13.2)
.92 .97 .77 .38
(31.5e45.6) (29.1e43.7) (6.0e16.0) (6.1e15.6)
.85 .86 1.04 .76
(.68e1.06) (.67e1.12) (.50e2.17) (.41e1.40)
52.7 48.4 12.7 14.5
(43.3e61.9) (39.5e57.4) (6.5e23.2)b (7.6e26.0)b
64.7 62.6 21.3 11.0
(55.9e72.6) (53.8e70.6) (12.1e34.6) (5.6e20.2)b
(22.5e35.6) (22.2e35.5) (1.9e11.2)b (3.1e12.0)b
.63 .69 .53 .57
(.47e.83)a (.51e.92)a (.20e1.43) (.26e1.24)
62.0 60.0 11.4 10.8
(53.1e70.2) (51.6e67.8) (6.1e20.4) (5.4e20.5)b
59.2 56.6 13.4 12.0
(51.9e66.2) (48.9e64.1) (6.8e24.7)b (7.0e19.8)
Vaccine era 2013e2016 % (95% CI)
29.5 28.0 7.0 11.3
(25.2e34.2) (23.8e32.6) (4.9e9.8) (8.0e15.7)
19.7 19.3 3.7 1.5
(14.9e25.7) (14.7e25.1) (1.8e7.8)b (.8e3.0)b
.67 .69 .53 .14
50.9 47.6 14.1 21.1
(42.0e59.8) (38.9e56.6) (7.9e23.7) (16.1e27.3)
55.1 55.1 13.8 5.7
(46.6e63.4) (46.6e63.4) (8.7e21.3) (2.3e13.4)b
45.3 41.7 9.6 13.1
(39.5e51.2) (35.0e48.7) (5.3e16.6) (8.5e19.6)
38.3 36.1 10.0 9.9
45.7 41.3 8.8 10.8
(37.9e53.8) (34.0e49.0) (5.5e13.7) (7.1e16.0)
28.6 28.4 4.7 6.2
All percents are weighted. PR ¼ prevalence ratio; 4vHPV ¼ quadrivalent HPV vaccine; CI ¼ confidence interval; HPV ¼ human papillomavirus. a For prevalence ratio comparing 2013e2016 to 2003e2006, confidence interval does not include 1. b Relative standard error >30% and 50%. c Relative standard error >50%.
PR (95% CI)
Prevaccine era 2003 -2006 % (95% CI)
PR (95% CI)
Prevaccine era 2003e2006 % (95% CI)
PR (95% CI)
Prevaccine era 2003e2006 % (95% CI)
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
14e19 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 20e24 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 25e29 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 30e34 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV
Non-Hispanic white n ¼ 1,582
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
A
2003-2006
2013-2016
90 80
Prevalence, %
70 60
*
50
40 30 20 10
*
0 Overall
* * NHW
NHB
* MA
Overall
4vHPV
B
NHW
NHB
MA
non-4vHPV 2003-2006
2013-2016
90 80
Prevalence, %
70 60 50 40 30 20 10
*
*
*
NHW
NHB
0 Overall
4vHPV
MA
Overall
NHW
NHB
MA
non-4vHPV
Figure 2. 4vHPV-type and non-4vHPV-type prevalences among sexually experienceda females aged (A) 14e19 years and (B) 20e24 years overall and by race/ ethnicity, NHANES 2003e2006 and 2013e2016. NHW: non-Hispanic white, NHB: non-Hispanic black, MA: Mexican American; 4vHPV, quadrivalent HPV vaccine. a Sexually experienced defined as reporting ever having had an oral, anal, or vaginal sex partner; A total of 1,090 females aged 14e19 years (NHW, n ¼ 302; NHB, n ¼ 352; MA, n ¼ 300) and 710 females aged 20e24 years (NHW, n ¼ 292; NHB, n ¼ 170; MA, n ¼ 145) were sexually experienced. *For prevalence ratio comparing 2013e2016 to 2003e2006, confidence interval does not include 1.
females. The smaller but significant declines observed in any HPV and non-4vHPV types in this age group are consistent with lower exposure to HPV. Because of the decline in reported sexual activity, we adjusted for number of lifetime sex partners in our overall analyses and also conducted an analysis limited to sexually experienced 14- to 19-year-olds. Among these females, 4vHPV-type prevalence declined 84% overall and 86% in NHW, 70% in NHB, and 84% in MA females. Among 20- to 24-year-olds, no declines were observed in non-4vHPV types. The combination of observed declines in 4vHPV-type prevalence, but not other HPV type-prevalence, while controlling for sexual behavior, supports vaccine impact as the main reason for the decline in 4vHPV-type prevalence. In addition to NHANES, other national surveys have also reported on sexual activity among adolescents in the U.S. since the introduction of HPV vaccination. Similar to the findings of this analysis, data from the Youth Risk Behavioral Surveillance System showed a declining trend in young people aged 13e 24 years reporting ever having sex between 2007 and 2017 [18]. Data from the National Survey of Family Growth also showed a declining trend over the past two decades, although no significant trend was observed during the vaccine period [19]. Furthermore, other countries have reported declines in sexual activity among adolescents since HPV vaccine introduction [20]. These findings complement other studies, which report no association between vaccination and sexual initiation [21], other sexually transmitted infections or teen pregnancy [22], or other
7
sexual risk behaviors [23]. Although reasons for downward trends in adolescent sexual behavior are unknown and likely multifactorial, no data suggest that HPV vaccination increases adolescent sexual risk behavior. This analysis uses data from a population-based survey that provides nationally representative estimates of HPV prevalence over time, but it is subject to limitations. First, analyses by race/ ethnicity do not account for all races/ethnicities because of limited sample sizes in females reporting other races. In addition, the Hispanic population reported in this analysis was restricted to MAs because of sampling changes over time; therefore, findings are likely not generalizable to all Hispanic females. The sample sizes in analyses by race/ethnicity were too small to adjust for other factors that may impact HPV prevalence, such as number of lifetime partners. However, an analysis of sexually experienced 14- to 19-year-olds and 20- to 24-year-olds was conducted as a sensitivity analysis and showed similar declines in 4vHPV-type prevalence. In conclusion, as national HPV vaccine coverage has increased, declines in vaccine-type prevalence have increased in magnitude, with 86% and 71% declines observed among 14- to 19-year-olds and 20- to 24-year-olds, respectively, in 2013e2016. These extraordinary declines are expected to lead to a meaningful reduction in HPV-associated cancers. Furthermore, with evidence of declines in 4vHPV-type prevalence among females of all reported race/ethnicities, HPV vaccination should reduce incidence of cervical cancer and other HPV-associated diseases in all racial/ethnic groups [24]. Continued monitoring in NHANES is important for ongoing evaluation of vaccine impact, particularly in light of the changes to the vaccination program in recent years. Data from NHANES will enable evaluation of the impact of 9vHPV vaccine that additionally targets HPV 31/33/45/52/58 [25], recommended in 2015, the change to a recommended 2-dose vaccination schedule among adolescents aged <15 years, and duration of protection after vaccination [26]. Continued efforts are also important to increase coverage and encourage vaccination at a routine age to further the impact of the HPV vaccination program in the U.S. Funding Sources This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The National Health and Nutrition Examination Survey is funded by the National Center for Health Statistics, Centers for Disease Control and Prevention. Supplementary Data Supplementary data related to this article can be found at https://doi.org/10.1016/j.jadohealth.2019.07.003. References [1] Markowitz LE, Dunne EF, Saraiya M, et al. Human papillomavirus vaccination: Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 2014;63:1e30. [2] Markowitz LE, Gee J, Chesson H, Stokley S. Ten years of human papillomavirus vaccination in the United States. Acad Pediatr 2018;18:S3e10. [3] Saraiya M, Unger ER, Thompson TD, et al. US assessment of HPV types in cancers: Implications for current and 9-valent HPV vaccines. J Natl Cancer Inst 2015;107:djv086. [4] Walker TY, Elam-Evans LD, Yankey D, et al. National, regional, state, and selected local area vaccination coverage among adolescents aged 13-17
8
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8 years - United States, 2017. MMWR Morb Mortal Wkly Rep 2018;67: 909e17. Hariri S, Unger ER, Sternberg M, et al. Prevalence of genital human papillomavirus among females in the United States, the National Health and Nutrition Examination survey, 2003-2006. J Infect Dis 2011;204:566e73. Markowitz LE, Hariri S, Lin C, et al. Reduction in human papillomavirus (HPV) prevalence among young women following HPV vaccine introduction in the United States, National Health and Nutrition Examination Surveys, 2003-2010. J Infect Dis 2013;208:385e93. Oliver SE, Unger ER, Lewis R, et al. Prevalence of human papillomavirus among females after vaccine introduction-National Health and Nutrition Examination Survey, United States, 2003-2014. J Infect Dis 2017;216: 594e603. Van Dyne EA, Henley SJ, Saraiya M, et al. Trends in human papillomavirusassociated cancers - United States, 1999-2015. MMWR Morb Mortal Wkly Rep 2018;67:918e24. Beavis AL, Gravitt PE, Rositch AF. Hysterectomy-corrected cervical cancer mortality rates reveal a larger racial disparity in the United States. Cancer 2017;123:1044e50. McCarthy AM, Dumanovsky T, Visvanathan K, et al. Racial/ethnic and socioeconomic disparities in mortality among women diagnosed with cervical cancer in New York City, 1995-2006. Cancer Causes Control 2010;21: 1645e55. Watson M, Benard V, King J, et al. National assessment of HPV and pap tests: Changes in cervical cancer screening, National Health Interview Survey. Prev Med 2017;100:243e7. Benard VB, Lawson HW, Eheman CR, et al. Adherence to guidelines for follow-up of low-grade cytologic abnormalities among medically underserved women. Obstet Gynecol 2005;105:1323e8. Giuliano AR, Lazcano-Ponce E, Villa L, et al. Impact of baseline covariates on the immunogenicity of a quadrivalent (types 6, 11, 16, and 18) human papillomavirus virus-like-particle vaccine. J Infect Dis 2007;196: 1153e62. Walker TY, Elam-Evans LD, Singleton JA, et al. National, regional, state, and selected local area vaccination coverage among adolescents aged 1317 years - United States, 2016. MMWR Morb Mortal Wkly Rep 2017;66: 874e82.
[15] Centers for Disease Control and Prevention. National and state vaccination coverage among adolescents aged 13 through 17 years–United States, 2010. MMWR Morb Mortal Wkly Rep 2011;60:1117e23. [16] Lewis RM, Markowitz LE. Human papillomavirus vaccination coverage among females and males, National Health and Nutrition Examination Survey, United States, 2007-2016. Vaccine 2018;36:2567e73. [17] Lin X, Shrader L, Rodgers L, et al. Increasing human papillomavirus vaccination at the recommended age. Vaccine 2019;37:686e9. [18] Centers for Disease Control and Prevention. Youth risk behavior survey: Data summary and trends report, 2007-2017. Available from: https://www.cdc. gov/healthyyouth/data/yrbs/index.htm. Accessed January 10, 2019. [19] Abma JC, Martinez GM. Sexual activity and contraceptive use among teenagers in the United States, 2011-2015. Natl Health Stat Rep 2017; 104:1e23. [20] Ogilvie GS, Phan F, Pedersen HN, et al. Population-level sexual behaviours in adolescent girls before and after introduction of the human papillomavirus vaccine (2003-2013). CMAJ 2018;190:E1221e6. [21] Mullins TLK, Rosenthal SL, Zimet GD, et al. Human papillomavirus vaccinerelated risk perceptions do not predict sexual initiation among young women over 30 months following vaccination. J Adolesc Health 2018;62:164e9. [22] Bednarczyk RA, Davis R, Ault K, et al. Sexual activity-related outcomes after human papillomavirus vaccination of 11- to 12-year-olds. Pediatrics 2012; 130:798e805. [23] Madhivanan P, Pierre-Victor D, Mukherjee S, et al. Human papillomavirus vaccination and sexual disinhibition in females: A systematic review. Am J Prev Med 2016;51:373e83. [24] Yoo W, Kim S, Huh WK, et al. Recent trends in racial and regional disparities in cervical cancer incidence and mortality in United States. PLoS One 2017;12:e0172548. [25] Petrosky E, Bocchini JA Jr, Hariri S, et al. Use of 9-valent human papillomavirus (HPV) vaccine: Updated HPV vaccination recommendations of the Advisory Committee on Immunization Practices. MMWR Morb Mortal Wkly Rep 2015;64:300e4. [26] Meites E, Kempe A, Markowitz LE. Use of a 2-dose schedule for human papillomavirus vaccination-updated recommendations of the Advisory Committee on Immunization Practices. MMWR Morb Mortal Wkly Rep 2016;65:1405e8.
www.jahonline.org Original article
Declines in Vaccine-Type Human Papillomavirus Prevalence in Females Across Racial/Ethnic Groups: Data From a National Survey Nancy M. McClung, Ph.D., R.N. a, b, *, Rayleen M. Lewis, M.P.H. a, c, Julia W. Gargano, Ph.D. a, Troy Querec, Ph.D. d, Elizabeth R. Unger, Ph.D., M.D. d, and Lauri E. Markowitz, M.D. a a
National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention (CDC), Atlanta, Georgia Epidemic Intelligence Service, Centers for Disease Control and Prevention, Atlanta, Georgia c Synergy America, Inc., Duluth, Georgia d National Center for Emerging and Zoonotic Infectious Diseases, Centers for Disease Control and Prevention (CDC), Atlanta, Georgia b
Article history: Received March 15, 2019; Accepted July 8, 2019 Keywords: Human papillomavirus; HPV vaccine; Public health
A B S T R A C T
Purpose: To monitor human papillomavirus (HPV) vaccine impact in the U.S., we evaluated quadrivalent vaccine (4vHPV)-type prevalence among females aged 14e34 years in the prevaccine (2003e2006) and vaccine (2013e2016) eras overall and by race/ethnicity in the National Health and Nutrition Examination Survey. Methods: We analyzed HPV DNA prevalence in self-collected cervicovaginal specimens, demographic characteristics, sexual behavior, and self-reported/parent-reported vaccination status. We compared prevaccine to vaccine era 4vHPV-type prevalence, using unadjusted and adjusted prevalence ratios (PR and aPR) and 95% confidence intervals (CIs). PRs were calculated by race/ ethnicity (non-Hispanic white [NHW], non-Hispanic black [NHB], and Mexican American [MA]). Overall aPRs were adjusted for race/ethnicity, lifetime sex partners, and poverty. Results: Overall, 4,674 females had HPV typing results; 3,915 reported NHW, NHB, or MA race/ ethnicity. Vaccination coverage of 1 dose was 53.9% among 14- to 19-year-olds (NHW 52.6%, NHB 58.1%, and MA 59.5%) and 51.5% among 20- to 24-year-olds (NHW 58.8%, NHB 45.0%, MA 33.8%). Among 14- to 19-year-olds, 4vHPV-type prevalence decreased overall (11.5% to 1.8%; aPR ¼ .14 [CI: .08e.24]) and in NHW (PR ¼ .14 [CI: .06e.29]), NHB (PR ¼ .26 [CI: .12e.54]), and MA (PR ¼ .13 [CI: .03e.53]). In 20- to 24-year-olds, 4vHPV-type prevalence decreased overall (18.5% to 5.3%; aPR ¼ .29 [CI: .15e.56]) and in NHW (PR ¼ .27 [CI: .11e.67]) and NHB (PR ¼ .38 [CI: .18e.80]). No significant declines were observed in older age groups. Conclusions: Within 10 years of vaccine introduction, 4vHPV-type prevalence declined 86% among 14- to 19-year-olds, with declines observed in NHW, NHB, and MA females, and 71% among 20- to 24-year-olds, with declines in NHW and NHB females. These extraordinary declines should lead to substantial reductions in HPV-associated cancers. Ó 2019 Published by Elsevier Inc. on behalf of Society for Adolescent Health and Medicine.
IMPLICATIONS AND CONTRIBUTION
This study provides updated human papillomavirus vaccine-type prevalence estimates through 2016 in females and reports new findings by race/ethnicity. Since vaccine introduction, a decline in prevalence was observed in all reported race/ethnicities among 14- to 19-yearolds, with an 86% decline overall, and a 71% decline among 20- to 24-yearolds.
Conflicts of interest: The authors have no conflicts of interest to disclose. Disclaimer: 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. * Address correspondence to: Nancy M. McClung, Ph.D., R.N., Centers for Disease Control and Prevention, 1600 Clifton Rd. NE, MS A34, Atlanta, GA 30329. E-mail address: [email protected] (N.M. McClung). 1054-139X/Ó 2019 Published by Elsevier Inc. on behalf of Society for Adolescent Health and Medicine. https://doi.org/10.1016/j.jadohealth.2019.07.003
2
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
In the U.S., routine vaccination against human papillomavirus (HPV) has been recommended since 2006 for females and 2011 for males [1]. Routine vaccination is recommended at age 11 or 12 years and for those not vaccinated at routine ages, through age 26 years for females, 21 years for males, and 26 years for special populations of males. Three HPV vaccines (i.e., bivalent, quadrivalent [4vHPV], and nonavalent [9vHPV] vaccines) are licensed for use in the U.S. From 2006 to 2015, 4vHPV was primarily used, with minimal bivalent use; 9vHPV was introduced in 2015, and since the end of 2016, only 9vHPV has been available in the U.S. [2]. 4vHPV targets HPV types 16 and 18, the causes of most HPVassociated cancers, and HPV 6 and 11, the causes of genital warts [3]. 9vHPV targets five additional cancer-causing types, HPV 31/33/45/52/58. Annual national surveys monitor HPV vaccination coverage among 13- to 17-year-olds in the U.S. Coverage among this age group has steadily increased, and in 2017, coverage of 1 dose and up-to-date coverage (two or three doses depending on age at series initiation) were 68.6% and 53.1%, respectively, among females, and 62.6% and 44.3%, respectively, among males [4]. Data from the National Health and Nutrition Examination Survey (NHANES) have been used to evaluate vaccine impact on HPV infection since vaccine introduction [5]. Within 4 years of vaccine introduction, a 56% decrease in the prevalence of 4vHPV infections was observed among 14- to 19-year-old females [6]. Most recently, within 8 years of vaccine introduction, a 71% decrease among 14- to 19-year-olds and a 61% decrease among 20- to 24-year-olds were observed [7]. The significant decrease in 4vHPV types is not only expected to lead to a decrease in HPV-associated cancers in the U.S. but also may lead to a reduction in racial/ethnic disparities in cancer incidence [8]. Non-Hispanic black (NHB) and Hispanic women have a higher cervical cancer incidence than non-Hispanic white (NHW) women [8], and NHB women have a higher cervical cancer mortality rate than NHW women [9] in part because of differences in screening participation and/or follow-up treatment [10e12]. Because HPV vaccine types 16 and 18 cause the vast majority of cervical cancers in all women, vaccination may reduce cervical cancer disparities [3]. Therefore we evaluated the impact of the U.S. vaccination program using NHANES HPV prevalence data among 14- to 34-year-old females from the four most recent vaccine era years (2013e2016) compared with the prevaccine era (2003e2006), overall and by race/ethnicity. This evaluation was limited to females because prevaccine era HPV prevalence is not available for males in NHANES. Methods
oversampled in 2003e2006, but not in 2013e2016. In 2003e2006, Mexican Americans (MAs) were oversampled. Oversampling in 2013e2016 was expanded to include all Hispanics; this group was predominantly MA. NHANES is approved by the CDC/National Center for Health Statistics Research Ethics Review Board. Study sample Female participants aged 14e34 years with adequate cervicovaginal samples were included in the overall analyses (n ¼ 4,674; Figure 1). Females who self-reported race/ethnicity as NHW, NHB, or MA were included in the race/ethnicity analyses (n ¼ 3,915). Of Hispanic females, only MA females were included to limit bias from sampling variation between 2003e2006 and 2013e2016. Other racial/ethnic groups were not analyzed separately because of small sample sizes. To achieve stable estimates, NHANES 2-year cycles were combined into 4-year eras: prevaccine era (2003e2006) and the most recent 4 years of the vaccine era (2013e2016). Study variables Demographic information including age, race/ethnicity, and poverty status (poverty level based on a ratio of household income to state poverty level). HPV vaccination history were selfreported/parent-reported during household interviews. Sexual behavior history (e.g., number of lifetime sex partners) was collected using audio computer-assisted self-interviews during examination at a MEC. Cervicovaginal samples were selfcollected at a MEC and shipped to the CDC laboratory for HPV genotyping. HPV genotyping methods have been described previously [5]. Briefly, DNA was extracted from samples and tested for 37 HPV types with a genotyping assay using nucleic acide amplification methods. HPV types were categorized as any HPV, non-4vHPV types (26/31/33/35/39/40/42/45/51/52/53/54/
Prevaccine Era 2003–2006
Vaccine Era 2013–2016
3325 females aged 14–34 intervieweda
2549 females aged 14–34 interviewedb
3210 (96.5%) received examination in MEC
2456 (96.4%) received examination in MEC
2649 (82.5%) submitted selfcollected cervicovaginal swab
2119 (86.3%) submitted selfcollected cervicovaginal swab
2587 (97.7%) samples with adequate typing results
2087 (98.5%) samples with adequate typing results
Study design NHANES is a population-based, cross-sectional survey conducted in 2-year cycles by the National Center for Health Statistics (NCHS) at the Centers for Disease Control and Prevention (CDC) to evaluate the health and nutrition of the U.S. population. Participants are randomly selected from civilian, noninstitutionalized households using a complex, multistage sampling design. After an in-home interview, they complete a series of interviews and physical examinations in a mobile examination center (MEC). Persons aged 16 years and emancipated minors are interviewed directly. Various subpopulations are oversampled in each 2-year cycle to increase the precision of estimates. Pertinent to this analysis, adolescents were
Figure 1. Participation and sample collection among females aged 14e34 yearsdNHANES, 2003e2006 and 2013e2016. MEC: medical examination center. aThe response rates during 2003e2004 and 2005e2006 among females were 79.7% and 80.9% for the interview portion, and 76.0% and 77.8% for the examination portion, respectively; bThe response rates during 2013e2014 and 2015e2016 among females were 71.4% and 62.6% for the interview portion, and 68.8% and 60.0% for the examination portion, respectively.
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
55/56/58/59/61/62/64/66/67/68/69/70/71/72/73/81/82/83/84/89/ IS39), additional types found in 9vHPV (31/33/45/52/58), and 4vHPV types (6/11/16/18). Data analysis Number of lifetime sex partners, poverty status, HPV vaccination, and HPV prevalence were described and compared across vaccine eras by age group (14e19, 20e24, 25e29, 30e34 years) overall and by race/ethnicity (NHW, NHB, and MA). Number of lifetime sex partners was categorized as 0, 1e2, or 3 partners. Poverty status was dichotomized as above or below the poverty level based on a ratio of household income to state poverty level. Two dichotomized measures of HPV vaccination status were used: receipt of 1 dose and 3 doses. All estimates were calculated by applying examination sample weights to account for unequal probabilities of selection and adjustment for nonresponse. To account for the complex survey design, variance estimates were calculated using a Taylor series linearization. Logit confidence intervals (CIs) were calculated for prevalence estimates, with a ¼ .05. Prevalence estimates with a relative standard error >30% were considered unstable. Wald chi-square tests were used to determine differences in number of lifetime sex partners, poverty status, and HPV vaccination between the prevaccine and vaccine eras. Logistic regression was used to estimate predicted marginal prevalence ratios comparing HPV prevalence in the vaccine to prevaccine era by age group. In the overall analysis, unadjusted and adjusted (a priori on race/ ethnicity, sex partners, and poverty status) prevalence ratios (PRs and aPRs) and 95% CIs were calculated. Because of convergence concerns in adjusted models, the three category number of lifetime partners variable was collapsed to two categories (0e2 and 3), and the dichotomized variable was used for adjustments. In the analyses by race/ethnicity, only unadjusted PRs were calculated because of small sample sizes. To control for the changing proportion of young women who were sexually experienced, prevalence estimates were also determined among sexually experienced 14e19 and 20- to 24-year-olds. Statistical significance was defined as p < .05 or, for a PR, as a 95% CI that does not include 1. All analyses were conducted in SAS version 9.3 (SAS Institute, Cary, NC) and SUDAAN version 11.0 (Research Triangle Institute, Research Triangle Park, North Carolina). Results Sexual behavior and poverty status Number of lifetime partners and poverty status, by age group overall and race/ethnicity, are summarized in Table 1. Significant changes in number of lifetime partners were observed in 14- to 19-year-olds overall and among NHW and NHB females. The proportion of 14- to 19-year-olds reporting zero lifetime partners increased from 46.1% in 2003e2006 to 56.2% in 2013e2016. Similar increases were observed among NHW (48.5% to 58.3%) and NHB (38.9% to 55.1%) females. No other significant differences in number of lifetime partners or poverty status were observed between the prevaccine and vaccine eras. HPV vaccination history In 2013e2016, the proportion of females with a history of 1 dose and 3 doses of HPV vaccine was 53.9% and 34.8%,
3
respectively, among 14- to 19-year-olds and 51.5% and 30.9%, respectively, among 20- to 24-year-olds and was lower in older age groups (Table 1). The median age at first vaccine dose was 12 years in all 14- to 19-year-olds and 15e16 years in all 20- to 24-year-olds. By race/ethnicity, the proportion of 14- to 19-yearold females with a history of 1 HPV vaccine dose was between 50% and 60% among all race/ethnicities (52.6% NHW, 58.1% NHB, and 59.5% MA). Among 20- to 24-year-olds, 1 dose vaccination coverage was 58.8% in NHW, 45.0% in NHB, and 33.8% in MA females. HPV prevalence by age group Among 14- to 19-year-olds, the prevalence of all HPV-type categories decreased significantly between 2003e2006 and 2013e2016 (Table 2). Prevalence of 4vHPV types was 11.5% in the prevaccine era and 1.8% in 2013e2016. After adjustment for lifetime sex partners, race/ethnicity, and poverty status, an 86% decline (aPR ¼ .14 [95% CI: .08e.24]) in 4vHPV-type prevalence was observed. Among 20- to 24-year-olds, 4vHPV-type prevalence was 18.5% in the prevaccine era and 5.3% in 2013e2016. After adjustment, a 71% decline was observed in 4vHPV-type prevalence (aPR ¼ .29 [95% CI: .15e.56]). No significant declines were observed in the other HPV-type categories among 14- to 19-year-olds and 20- to 24-year-olds after adjustment. No significant decline in 4vHPV-type prevalence was observed in older age groups. Among 30- to 34-year-olds, there were declines in any HPV (aPR ¼ .72 [95% CI: .60e.86]) and non-4vHPV-type (aPR ¼ .76 [95% CI: .63e.91]) prevalence. Prevalence of individual HPV types in 14- to 19-year-olds and 20- to 24-year-olds is reported in Supplementary Tables 1 and 2. HPV prevalence by age group and race/ethnicity The prevalence of 4vHPV types decreased significantly from the prevaccine to vaccine era among 14- to 19-year-olds in all racial/ethnic groups, from 11.3% to 1.5% in NHW (PR ¼ .14 [95% CI: .06e.29]), 17.1% to 4.4% in NHB (PR ¼ .26 [95% CI: .12e.54]), and 10.0% to 1.3% in MA (PR ¼ .13 [95% CI: .03e.53]) females (Table 3). Prevalence of HPV 31/33/45/52/58 declined among NHB (PR ¼ .41 [95% CI: .25e.67]) and MA (PR ¼ .37 [95% CI: .18e.75]), but not NHW females. Among 20- to 24-year-olds, 4vHPV-type prevalence decreased in NHW females, from 21.1% to 5.7% (PR ¼ .27 [95% CI: .11e.67]), and in NHB females, from 19.6% to 7.5% (PR ¼ .38 [95% CI: .18e.80]), but not in MA females. Among older age groups, a decline in 4vHPV-type prevalence was not observed in any racial/ethnic group, and declines in any HPV prevalence (PR ¼ .63 [95% CI: .47e.83]) and non-4vHPV-type prevalence (PR ¼ .69 [95% CI: .51e.92]) were only observed among NHW 30- to 34-year-olds. HPV prevalence among 14- to 19-year-old and 20- to 24-year-old sexually experienced females Because of the increase in proportion of 14- to 19-year-olds reporting no lifetime sex partners between the prevaccine era and 2013e2016, we also analyzed data limited to sexually experienced (report of ever having had an oral, anal, or vaginal sex partner) females in this age group and also in 20- to 24-yearolds. Among sexually experienced 14- to 19-year-olds, overall 4vHPV-type prevalence declined from 19.3% in the prevaccine era to 3.1% in the vaccine era (PR ¼ .16 [95% CI: .09e.27];
Age group, characteristics
20e24 y Lifetime sex partners 0 1e2 3 Poverty level Living below HPV vaccination 1 dose 3 doses 25e29 y Lifetime sex partners 0 1e2 3 Poverty level Living below HPV vaccination 1 dose 3 doses 30e34 y Lifetime sex partners 0 1e2 3 Poverty level Living below HPV vaccination 1 dose 3 doses
Non-Hispanic white
Non-Hispanic black
n ¼ 4,674
n ¼ 1,582
n ¼ 1,206
Mexican American n ¼ 1,127
2003e2006
2013e2016
2003e2006
2013e2016
2003e2006
2013e2016
2003e2006
2013e2016
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
n ¼ 1,363
n ¼ 783
n ¼ 397
n ¼ 205
n ¼ 465
n ¼ 186
n ¼ 409
n ¼ 196
47.0 (39.9e54.2) 35.5 (31.4e39.8) 17.5 (13.5e22.5)
50.7 (43.5e57.9) 33.0 (28.1e38.4) 16.3 (11.9e21.9)
a
a
a
46.1 (43.0e49.3) 28.2 (25.0e31.7) 25.6 (22.5e29.0)
56.2 (51.9e60.5) 25.2 (22.2e28.4) 18.6 (15.3e22.4)
48.5 (43.8e53.2) 26.2 (22.0e30.9) 25.3 (21.4e29.8)
58.3 (51.5e64.7) 23.3 (19.1e28.2) 18.4 (13.8e24.2)
38.9 (33.1e45.0) 31.2 (25.9e37.1) 29.9 (23.6e37.0)
55.1 (47.7e62.3) 20.2 (15.0e26.7) 24.7 (19.0e31.4)
23.0 (18.7e27.9)
21.6 (17.0e27.0)
15.1 (10.5e21.1)
11.2 (6.2e19.4)
43.9 (35.7e52.4)
35.0 (25.5e45.9)
41.5 (35.3e48.0)
43.1 (36.7e49.7)
-
53.9 (48.1e59.6) 34.8 (29.4e40.6)
-
52.6 (44.4e60.6) 33.8 (26.7e41.7)
-
58.1 (49.8e66.0) 39.8 (31.1e49.1)
-
59.5 (50.2e68.2) 37.4 (28.1e47.7)
n ¼ 432
n ¼ 413
n ¼ 179
n ¼ 147
n ¼ 110
n ¼ 93
b
c
c
n ¼ 107
n ¼ 74
b
8.6 (5.2e14.0) 30.7 (26.4e35.4) 60.7 (53.7e67.2)
5.4 (3.4e8.4) 27.5 (23.1e32.4) 67.1 (62.2e71.7)
6.3 (2.3e15.6) 29.3 (24.2e35.0) 64.5 (56.2e71.9)
3.1 (1.0e8.7) 27.3 (20.5e35.5) 69.6 (62.1e76.2)
4.6 (1.6e12.8) 16.3 (9.7e26.1) 79.1 (68.1e87.0)
6.1 (2.4e14.8) 20.5 (11.5e33.9) 73.4 (60.6e83.2)
12.0 (7.5e18.7) 46.5 (37.7e55.4) 41.6 (31.8e52.0)
14.9 (8.2e25.5) 38.9 (27.4e51.9) 46.2 (34.7e58.1)
25.7 (19.8e32.7)
29.3 (24.0e35.2)
18.3 (11.7e27.4)
19.8 (14.2e27.0)
43.1 (30.4e56.8)
48.0 (37.8e58.4)
51.5 (36.9e65.8)
44.0 (27.1e62.4)
-
51.5 (44.0e59.0) 30.9 (24.5e38.1)
-
58.8 (47.9e69.0) 34.8 (25.4e45.5)
-
45.0 (37.1e53.1) 26.0 (17.0e37.5)
-
33.8 (21.4e49.0) 17.3 (10.5e27.0)
n ¼ 403
n ¼ 447
n ¼ 190
n ¼ 150
n ¼ 76
n ¼ 91
n ¼ 104
n ¼ 71
5.0 (2.9e8.3) 21.8 (15.9e29.1) 73.2 (66.4e79.0)
4.4 (2.6e7.4) 26.7 (21.2e32.9) 69.0 (63.1e74.3)
3.2 (1.3e7.7)b 18.6 (12.8e26.2) 78.2 (71.1e84.0)
4.8 (2.3e9.9)b 26.5 (17.9e37.2) 68.7 (58.7e77.2)
3.8 (.8e16.8)c 14.4 (7.4e26.2)b 81.8 (70.1e89.7)
5.2 (1.7e14.7)c 12.9 (8.7e18.9) 81.9 (74.2e87.7)
5.7 (1.8e17.0)c 44.4 (29.5e60.4) 49.9 (34.5e65.2)
4.9 (1.2e17.8)c 29.7 (18.0e44.9) 65.4 (52.6e76.3)
16.3 (12.5e21.0)
20.8 (16.6e25.7)
11.9 (7.6e17.9)
14.4 (9.0e22.3)
30.7 (20.6e43.1)
34.7 (24.1e47.1)
30.5 (22.9e39.3)
38.4 (21.6e58.6)
-
33.3 (28.2e38.9) 24.8 (19.9e30.6)
-
37.5 (29.6e46.2) 30.1 (22.8e38.5)
-
27.4 (19.2e37.4) 14.3 (8.1e24.0)
-
26.8 (18.3e37.5) 13.9 (7.8e23.6)
n ¼ 389
n ¼ 444
n ¼ 165
n ¼ 149
n ¼ 90
n ¼ 95
n ¼ 97
n ¼ 69
1.6 (.6e4.6)c 24.8 (20.5e29.6) 73.6 (68.2e78.4)
2.7 (1.2e6.2)b 23.5 (18.7e29.1) 73.8 (68.4e78.6)
1.0 (.1e7.0)c 20.6 (14.9e27.6) 78.5 (71.8e83.9)
2.4 (.6e8.7)c 21.6 (13.9e32.2) 75.9 (65.4e84.1)
1.3 (.2e9.4)c 8.5 (3.2e20.5)b 90.2 (78.3e95.9)
1.3 (.2e9.8)c 6.3 (2.1e17.5)c 92.4 (80.9e97.2)
3.0 (.7e12.9)c 55.2 (42.4e67.3) 41.8 (29.7e55.0)
4.8 (1.4e15.0)c 37.7 (25.2e52.1) 57.6 (43.1e70.8)
16.0 (12.7e20.1)
21.3 (16.9e26.4)
9.6 (5.8e15.5)
13.0 (7.5e21.4)
25.1 (18.9e32.6)
36.1 (26.5e46.8)
36.4 (27.5e46.4)
39.3 (26.8e53.5)
-
14.3 (10.7e18.7) 9.5 (6.5e13.7)
16.8 (11.6e23.7) 13.1 (8.5e19.7)
-
16.8 (10.3e26.2) 7.3 (3.7e13.8)b
-
-
All percentages are weighted. CI ¼ confidence interval; HPV ¼ human papillomavirus. a p < .05 based on Wald chi-square test comparing 2013e2016 to 2003e2006. b Relative standard error >30% and 50%. c Relative standard error > 50%.
7.0 (2.6e17.4)b 2.1 (.3e13.0)c
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
14e19 y Lifetime sex partners 0 1e2 3 Poverty level Living below HPV vaccination 1 dose 3 doses
Overall
4
Table 1 Lifetime sex partners, poverty status, and vaccination history among females aged 14e34 years, by age group and race/ethnicitydNHANES 2003e2006 and 2013e2016
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
5
Table 2 HPV prevalence among females aged 14e34 years, by age groupdNHANES 2003e2006 and 2013e2016 Age group/HPV types
14e19 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 20e24 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 25e29 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 30e34 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV
PR (95% CI)
aPRa (95% CI)
Prevaccine era 2003e2006 % (95% CI)
Vaccine era 2013e2016 % (95% CI)
32.9 31.2 8.4 11.5
(29.5e36.5) (27.9e34.8) (6.6e10.6) (9.1e14.4)
21.6 21.0 3.9 1.8
(18.5e25.0) (18.0e24.4) (2.5e5.9) (1.1e2.9)
.66 .67 .46 .15
(.55e.79)b (.56e.81)b (.29e.74)b (.09e.26)b
.85 .89 .69 .14
(.72e1.01) (.74e1.07) (.39e1.20) (.08e.24)b
53.7 50.7 16.5 18.5
(45.9e61.4) (43.4e58.0) (11.3e23.4) (14.9e22.8)
54.6 54.4 13.3 5.3
(48.0e61.0) (47.9e60.8) (9.5e18.2) (2.9e9.4)
1.01 1.07 .80 .29
(.84e1.22) (.89e1.29) (.50e1.30) (.16e.53)b
1.01 1.08 .92 .29
(.87e1.18) (.92e1.26) (.56e1.50) (.15e.56)b
46.8 43.8 10.8 11.8
(42.9e50.8) (38.9e48.9) (7.3e15.7) (8.8e15.6)
43.5 41.8 11.5 8.0
(38.8e48.3) (36.8e46.9) (8.2e15.9) (5.6e11.4)
.93 .95 1.06 .68
(.81e1.07) (.81e1.12) (.65e1.75) (.43e1.07)
.92 .95 .92 .61
(.80e1.06) (.80e1.12) (.53e1.58) (.36e1.04)
47.9 44.5 9.8 9.5
(42.3e53.5) (39.1e50.1) (7.1e13.5) (6.7e13.2)
35.4 34.2 6.2 6.5
(30.4e40.7) (29.1e39.7) (3.7e10.3) (4.4e9.7)
.74 .77 .63 .69
(.62e.89)b (.63e.93)b (.35e1.15) (.41e1.15)
.72 .76 .58 .73
(.60e.86)b (.63e.91)b (.29e1.17) (.41e1.32)
All percents are weighted. 4vHPV ¼ quadrivalent HPV vaccine; aPR ¼ adjusted prevalence ratio; CI ¼ confidence interval; HPV ¼ human papillomavirus; PR ¼ prevalence ratio. a Adjusted for race/ethnicity, lifetime sex partners, and poverty status. b For prevalence ratio comparing 2013e2016 to 2003e2006, confidence interval does not include 1.
Figure 2A). Declines in 4vHPV-type prevalence were observed in all racial/ethnic groups: NHW PR ¼ .14 (95% CI: .07e.28), NHB PR ¼ .30 (95% CI: .14e.64), MA PR ¼ .16 (95% CI: .04e.71). Overall, non-4vHPV-type prevalence declined from 50.1% to 41.4% (PR ¼ .83 [95% CI: .71e.97]), but no significant declines were observed in individual race/ethnicity groups. Among sexually experienced 20- to 24-year-olds, 4vHPV-type prevalence declined overall from 17.9% to 5.7% (PR ¼ .32 [95% CI: .17e.59]; Figure 2B); by race/ ethnicity, declines were only observed among NHW (PR ¼ .28 [95% CI: .11e.71]) and NHB (PR ¼ .43 [95% CI: .20e.91]) females. No significant declines were observed in non-4vHPV-type prevalence in this age group. Discussion The results of this analysis, using the most recent national data, indicate continued impact of the national HPV vaccination program on 4vHPV-type prevalence among young women in the U.S. A decade after the introduction of HPV vaccine, the prevalence of 4vHPV types declined 86% among 14- to 19-year-olds and 71% among 20- to 24-year-olds. In the most recent previous analysis, comparing 2011e2014 data to the prevaccine era, 71% and 61% decreases were observed in the same age groups [7]. Importantly, this analysis extends prior reports of NHANES data by showing that the decline is occurring across the racial/ethnic groups analyzed. Among 14- to 19-year-olds, 4vHPV-type prevalence declined 86% in NHW, 74% in NHB, and 87% in MA females. Findings by race/ethnicity are consistent with prelicensure clinical trials, which found robust immunogenicity in all represented races, ethnicities, and countries [13]. Among 20- to 24-year-olds, 4vHPV-type prevalence declined 73% in NHW and 62% in NHB females; a similar decline (63%) was observed in MA females, but because of a lower prevalence of vaccine types in 2003e2006, there was less power to detect a significant decline in 2013e2016.
A continued decline in 4vHPV-type prevalence is consistent with increasing vaccination coverage among young women in the U.S. According to the National Immunization Survey, Teen (NIS-Teen), which uses provider-verified data to monitor the U.S. vaccination program among 13- to 17-year-olds, 1 dose vaccination coverage has increased from 48.7% in 2010 to 65.1% in 2016, and three-dose vaccination coverage from 32.0% to 43.0%, among U.S. females [1,14,15]. Similar increases in 1 dose vaccination coverage have also been observed among females aged 14e19 years in NHANES, from 45.2% in 2009e2010 to 54.7% in 2015e2016 [16]. Vaccine status is self-reported in NHANES, which likely contributes to the lower coverage estimates than NIS-Teen. According to NIS-Teen, vaccination coverage has increased in all reported races/ethnicities, and in 2016, 1 dose vaccination coverage was similar by race/ethnicity among 13- to 17-year-olds [14]. In addition to increased coverage, continued declines are consistent with a higher percentage of girls receiving vaccination at the routine age (11 or 12 years), before exposure to HPV through sexual activity [17]. In NHANES 2013e2016, the median age at vaccine initiation was 12 years among 14- to 19year-olds of all race/ethnicities but was 16 years among 20- to 24-year-olds. Even greater vaccine impact on 4vHPV-type prevalence will be observed as more females who were vaccinated according to the routine recommendations age into older age groups. Although increasing vaccination coverage is consistent with continued declines in 4vHPV-type prevalence, changes in sexual behavior may also contribute to declines. In this analysis, the proportion of 14- to 19-year-olds reporting no lifetime sex partners in 2013e2016 (56.2%) was higher than in 2003e2006 (46.1%). Of note, this is the first significant change in sexual behavior among 14- to 19-year-olds tested for HPV since HPV has been monitored in NHANES. Similar increases in the proportion reporting no lifetime sex partners were observed among NHW (48.5% to 58.3%) and NHB (38.9% to 55.1%) females, but not in MA
6
Table 3 HPV prevalence among females aged 14e34 years, by age group and race/ethnicitydNHANES 2003e2006 and 2013e2016 Age group/HPV types
Non-Hispanic black n ¼ 1,206
Mexican American n ¼ 1,127 Vaccine era 2013e2016 % (95% CI)
(.56e.80)a (.56e.80)a (.25e.67)a (.12e.54)a
33.6 30.5 9.7 10.0
(27.5e40.3) (24.4e37.3) (5.9e15.6) (6.7e14.5)
21.6 20.8 3.6 1.3
(15.7e28.9) (15.3e27.7) (2.1e6.1) (.3e5.1)c
.64 .68 .37 .13
(.45e.92)a (.48e.98)a (.18e.75)a (.03e.53)a
(.76e1.12) (.80e1.17) (.43e1.38) (.18e.80)a
45.9 44.0 10.6 7.2
(32.3e60.1) (29.8e59.3) (5.4e19.7)b (2.5e19.2)c
42.4 42.4 6.4 2.7
(30.1e55.8) (30.1e55.8) (2.4e15.9)b (.6e11.4)c
.92 .96 .60 .37
(.60e1.42) (.61e1.52) (.19e1.87) (.06e2.23)
1.23 1.29 1.68 .76
(.99e1.52) (1.03e1.62)a (.74e3.80) (.31e1.81)
49.0 47.8 11.1 8.3
(39.4e58.6) (38.2e57.5) (6.0e19.6) (3.2e19.8)b
40.5 40.5 9.4 1.3
(29.6e52.5) (29.6e52.5) (3.9e21.1)b (.1e10.8)c
.83 .85 .84 .15
(.59e1.17) (.60e1.20) (.30e2.34) (.01e1.59)
.96 .94 1.17 1.10
(.80e1.14) (.78e1.14) (.49e2.81) (.48e2.54)
46.2 46.2 12.5 3.7
(38.1e54.5) (38.1e54.5) (5.9e24.3)b (1.1e11.5)c
37.1 35.5 10.0 3.9
(23.8e52.7) (21.8e52.1) (4.0e22.6)b (1.3e10.7)c
.80 .77 .80 1.06
(.52e1.23) (.48e1.22) (.27e2.42) (.22e4.97)
Vaccine era 2013e2016 % (95% CI)
(.49e.91)a (.51e.94)a (.24e1.20) (.06e.29)a
49.5 47.8 16.5 17.1
(44.8e54.2) (43.0e52.7) (13.2e20.5) (13.5e21.6)
33.0 31.9 6.8 4.4
(28.1e38.4) (27.2e37.0) (4.3e10.6) (2.1e9.0)b
.67 .67 .41 .26
1.08 1.16 .98 .27
(.86e1.36) (.91e1.47) (.49e1.98) (.11e.67)a
76.3 72.5 29.6 19.6
(64.1e85.3) (61.7e81.2) (19.1e42.8) (12.0e30.4)
70.0 70.0 22.9 7.5
(59.2e79.0) (59.2e79.0) (14.5e34.1) (4.2e13.2)
.92 .97 .77 .38
(31.5e45.6) (29.1e43.7) (6.0e16.0) (6.1e15.6)
.85 .86 1.04 .76
(.68e1.06) (.67e1.12) (.50e2.17) (.41e1.40)
52.7 48.4 12.7 14.5
(43.3e61.9) (39.5e57.4) (6.5e23.2)b (7.6e26.0)b
64.7 62.6 21.3 11.0
(55.9e72.6) (53.8e70.6) (12.1e34.6) (5.6e20.2)b
(22.5e35.6) (22.2e35.5) (1.9e11.2)b (3.1e12.0)b
.63 .69 .53 .57
(.47e.83)a (.51e.92)a (.20e1.43) (.26e1.24)
62.0 60.0 11.4 10.8
(53.1e70.2) (51.6e67.8) (6.1e20.4) (5.4e20.5)b
59.2 56.6 13.4 12.0
(51.9e66.2) (48.9e64.1) (6.8e24.7)b (7.0e19.8)
Vaccine era 2013e2016 % (95% CI)
29.5 28.0 7.0 11.3
(25.2e34.2) (23.8e32.6) (4.9e9.8) (8.0e15.7)
19.7 19.3 3.7 1.5
(14.9e25.7) (14.7e25.1) (1.8e7.8)b (.8e3.0)b
.67 .69 .53 .14
50.9 47.6 14.1 21.1
(42.0e59.8) (38.9e56.6) (7.9e23.7) (16.1e27.3)
55.1 55.1 13.8 5.7
(46.6e63.4) (46.6e63.4) (8.7e21.3) (2.3e13.4)b
45.3 41.7 9.6 13.1
(39.5e51.2) (35.0e48.7) (5.3e16.6) (8.5e19.6)
38.3 36.1 10.0 9.9
45.7 41.3 8.8 10.8
(37.9e53.8) (34.0e49.0) (5.5e13.7) (7.1e16.0)
28.6 28.4 4.7 6.2
All percents are weighted. PR ¼ prevalence ratio; 4vHPV ¼ quadrivalent HPV vaccine; CI ¼ confidence interval; HPV ¼ human papillomavirus. a For prevalence ratio comparing 2013e2016 to 2003e2006, confidence interval does not include 1. b Relative standard error >30% and 50%. c Relative standard error >50%.
PR (95% CI)
Prevaccine era 2003 -2006 % (95% CI)
PR (95% CI)
Prevaccine era 2003e2006 % (95% CI)
PR (95% CI)
Prevaccine era 2003e2006 % (95% CI)
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
14e19 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 20e24 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 25e29 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV 30e34 y Any HPV Non-4vHPV HPV 31/33/45/52/58 4vHPV
Non-Hispanic white n ¼ 1,582
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8
A
2003-2006
2013-2016
90 80
Prevalence, %
70 60
*
50
40 30 20 10
*
0 Overall
* * NHW
NHB
* MA
Overall
4vHPV
B
NHW
NHB
MA
non-4vHPV 2003-2006
2013-2016
90 80
Prevalence, %
70 60 50 40 30 20 10
*
*
*
NHW
NHB
0 Overall
4vHPV
MA
Overall
NHW
NHB
MA
non-4vHPV
Figure 2. 4vHPV-type and non-4vHPV-type prevalences among sexually experienceda females aged (A) 14e19 years and (B) 20e24 years overall and by race/ ethnicity, NHANES 2003e2006 and 2013e2016. NHW: non-Hispanic white, NHB: non-Hispanic black, MA: Mexican American; 4vHPV, quadrivalent HPV vaccine. a Sexually experienced defined as reporting ever having had an oral, anal, or vaginal sex partner; A total of 1,090 females aged 14e19 years (NHW, n ¼ 302; NHB, n ¼ 352; MA, n ¼ 300) and 710 females aged 20e24 years (NHW, n ¼ 292; NHB, n ¼ 170; MA, n ¼ 145) were sexually experienced. *For prevalence ratio comparing 2013e2016 to 2003e2006, confidence interval does not include 1.
females. The smaller but significant declines observed in any HPV and non-4vHPV types in this age group are consistent with lower exposure to HPV. Because of the decline in reported sexual activity, we adjusted for number of lifetime sex partners in our overall analyses and also conducted an analysis limited to sexually experienced 14- to 19-year-olds. Among these females, 4vHPV-type prevalence declined 84% overall and 86% in NHW, 70% in NHB, and 84% in MA females. Among 20- to 24-year-olds, no declines were observed in non-4vHPV types. The combination of observed declines in 4vHPV-type prevalence, but not other HPV type-prevalence, while controlling for sexual behavior, supports vaccine impact as the main reason for the decline in 4vHPV-type prevalence. In addition to NHANES, other national surveys have also reported on sexual activity among adolescents in the U.S. since the introduction of HPV vaccination. Similar to the findings of this analysis, data from the Youth Risk Behavioral Surveillance System showed a declining trend in young people aged 13e 24 years reporting ever having sex between 2007 and 2017 [18]. Data from the National Survey of Family Growth also showed a declining trend over the past two decades, although no significant trend was observed during the vaccine period [19]. Furthermore, other countries have reported declines in sexual activity among adolescents since HPV vaccine introduction [20]. These findings complement other studies, which report no association between vaccination and sexual initiation [21], other sexually transmitted infections or teen pregnancy [22], or other
7
sexual risk behaviors [23]. Although reasons for downward trends in adolescent sexual behavior are unknown and likely multifactorial, no data suggest that HPV vaccination increases adolescent sexual risk behavior. This analysis uses data from a population-based survey that provides nationally representative estimates of HPV prevalence over time, but it is subject to limitations. First, analyses by race/ ethnicity do not account for all races/ethnicities because of limited sample sizes in females reporting other races. In addition, the Hispanic population reported in this analysis was restricted to MAs because of sampling changes over time; therefore, findings are likely not generalizable to all Hispanic females. The sample sizes in analyses by race/ethnicity were too small to adjust for other factors that may impact HPV prevalence, such as number of lifetime partners. However, an analysis of sexually experienced 14- to 19-year-olds and 20- to 24-year-olds was conducted as a sensitivity analysis and showed similar declines in 4vHPV-type prevalence. In conclusion, as national HPV vaccine coverage has increased, declines in vaccine-type prevalence have increased in magnitude, with 86% and 71% declines observed among 14- to 19-year-olds and 20- to 24-year-olds, respectively, in 2013e2016. These extraordinary declines are expected to lead to a meaningful reduction in HPV-associated cancers. Furthermore, with evidence of declines in 4vHPV-type prevalence among females of all reported race/ethnicities, HPV vaccination should reduce incidence of cervical cancer and other HPV-associated diseases in all racial/ethnic groups [24]. Continued monitoring in NHANES is important for ongoing evaluation of vaccine impact, particularly in light of the changes to the vaccination program in recent years. Data from NHANES will enable evaluation of the impact of 9vHPV vaccine that additionally targets HPV 31/33/45/52/58 [25], recommended in 2015, the change to a recommended 2-dose vaccination schedule among adolescents aged <15 years, and duration of protection after vaccination [26]. Continued efforts are also important to increase coverage and encourage vaccination at a routine age to further the impact of the HPV vaccination program in the U.S. Funding Sources This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The National Health and Nutrition Examination Survey is funded by the National Center for Health Statistics, Centers for Disease Control and Prevention. Supplementary Data Supplementary data related to this article can be found at https://doi.org/10.1016/j.jadohealth.2019.07.003. References [1] Markowitz LE, Dunne EF, Saraiya M, et al. Human papillomavirus vaccination: Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 2014;63:1e30. [2] Markowitz LE, Gee J, Chesson H, Stokley S. Ten years of human papillomavirus vaccination in the United States. Acad Pediatr 2018;18:S3e10. [3] Saraiya M, Unger ER, Thompson TD, et al. US assessment of HPV types in cancers: Implications for current and 9-valent HPV vaccines. J Natl Cancer Inst 2015;107:djv086. [4] Walker TY, Elam-Evans LD, Yankey D, et al. National, regional, state, and selected local area vaccination coverage among adolescents aged 13-17
8
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
N.M. McClung et al. / Journal of Adolescent Health xxx (2019) 1e8 years - United States, 2017. MMWR Morb Mortal Wkly Rep 2018;67: 909e17. Hariri S, Unger ER, Sternberg M, et al. Prevalence of genital human papillomavirus among females in the United States, the National Health and Nutrition Examination survey, 2003-2006. J Infect Dis 2011;204:566e73. Markowitz LE, Hariri S, Lin C, et al. Reduction in human papillomavirus (HPV) prevalence among young women following HPV vaccine introduction in the United States, National Health and Nutrition Examination Surveys, 2003-2010. J Infect Dis 2013;208:385e93. Oliver SE, Unger ER, Lewis R, et al. Prevalence of human papillomavirus among females after vaccine introduction-National Health and Nutrition Examination Survey, United States, 2003-2014. J Infect Dis 2017;216: 594e603. Van Dyne EA, Henley SJ, Saraiya M, et al. Trends in human papillomavirusassociated cancers - United States, 1999-2015. MMWR Morb Mortal Wkly Rep 2018;67:918e24. Beavis AL, Gravitt PE, Rositch AF. Hysterectomy-corrected cervical cancer mortality rates reveal a larger racial disparity in the United States. Cancer 2017;123:1044e50. McCarthy AM, Dumanovsky T, Visvanathan K, et al. Racial/ethnic and socioeconomic disparities in mortality among women diagnosed with cervical cancer in New York City, 1995-2006. Cancer Causes Control 2010;21: 1645e55. Watson M, Benard V, King J, et al. National assessment of HPV and pap tests: Changes in cervical cancer screening, National Health Interview Survey. Prev Med 2017;100:243e7. Benard VB, Lawson HW, Eheman CR, et al. Adherence to guidelines for follow-up of low-grade cytologic abnormalities among medically underserved women. Obstet Gynecol 2005;105:1323e8. Giuliano AR, Lazcano-Ponce E, Villa L, et al. Impact of baseline covariates on the immunogenicity of a quadrivalent (types 6, 11, 16, and 18) human papillomavirus virus-like-particle vaccine. J Infect Dis 2007;196: 1153e62. Walker TY, Elam-Evans LD, Singleton JA, et al. National, regional, state, and selected local area vaccination coverage among adolescents aged 1317 years - United States, 2016. MMWR Morb Mortal Wkly Rep 2017;66: 874e82.
[15] Centers for Disease Control and Prevention. National and state vaccination coverage among adolescents aged 13 through 17 years–United States, 2010. MMWR Morb Mortal Wkly Rep 2011;60:1117e23. [16] Lewis RM, Markowitz LE. Human papillomavirus vaccination coverage among females and males, National Health and Nutrition Examination Survey, United States, 2007-2016. Vaccine 2018;36:2567e73. [17] Lin X, Shrader L, Rodgers L, et al. Increasing human papillomavirus vaccination at the recommended age. Vaccine 2019;37:686e9. [18] Centers for Disease Control and Prevention. Youth risk behavior survey: Data summary and trends report, 2007-2017. Available from: https://www.cdc. gov/healthyyouth/data/yrbs/index.htm. Accessed January 10, 2019. [19] Abma JC, Martinez GM. Sexual activity and contraceptive use among teenagers in the United States, 2011-2015. Natl Health Stat Rep 2017; 104:1e23. [20] Ogilvie GS, Phan F, Pedersen HN, et al. Population-level sexual behaviours in adolescent girls before and after introduction of the human papillomavirus vaccine (2003-2013). CMAJ 2018;190:E1221e6. [21] Mullins TLK, Rosenthal SL, Zimet GD, et al. Human papillomavirus vaccinerelated risk perceptions do not predict sexual initiation among young women over 30 months following vaccination. J Adolesc Health 2018;62:164e9. [22] Bednarczyk RA, Davis R, Ault K, et al. Sexual activity-related outcomes after human papillomavirus vaccination of 11- to 12-year-olds. Pediatrics 2012; 130:798e805. [23] Madhivanan P, Pierre-Victor D, Mukherjee S, et al. Human papillomavirus vaccination and sexual disinhibition in females: A systematic review. Am J Prev Med 2016;51:373e83. [24] Yoo W, Kim S, Huh WK, et al. Recent trends in racial and regional disparities in cervical cancer incidence and mortality in United States. PLoS One 2017;12:e0172548. [25] Petrosky E, Bocchini JA Jr, Hariri S, et al. Use of 9-valent human papillomavirus (HPV) vaccine: Updated HPV vaccination recommendations of the Advisory Committee on Immunization Practices. MMWR Morb Mortal Wkly Rep 2015;64:300e4. [26] Meites E, Kempe A, Markowitz LE. Use of a 2-dose schedule for human papillomavirus vaccination-updated recommendations of the Advisory Committee on Immunization Practices. MMWR Morb Mortal Wkly Rep 2016;65:1405e8.