Sexual behavior, clinical outcomes and attendance of cervical cancer screening by HPV vaccinated and unvaccinated sexually active women

ABSTRACT

Concerns were raised about HPV vaccination possibly leading to riskier sexual behavior. We assessed sexual behaviors, risk of sexually transmitted infection, and attendance to cervical cancer screening by HPV vaccinated and unvaccinated young women. In this analysis, 1475 questionnaires completed by women aged 17–29 years were included. The majority of respondents (67.9%) were vaccinated against HPV. The proportion of those vaccinated decreased with age: from 93.2% in those aged 17–19 to 72.9% in those aged 20–22, and 21.8% in 23–29-year olds. A higher proportion of unvaccinated respondents had at least one sexual intercourse under the age of 15 when compared to those vaccinated (30% vs. 23%, p < .0001). The number of sexual partners during the last 12 months was similar between vaccinated and unvaccinated participants. Vaccinated participants reported more condom use (45% versus 38%; p = .0002), and less sexually transmitted infections (10% versus 28%; p < .0001), and less anogenital condylomas (2.2% vs. 11.6%; p < .0001). A screening test has been reported by 51% and 77% of vaccinated and unvaccinated participants, respectively (p < .0001). The association between vaccination status and cervical cancer screening disappeared when adjusting for participants’ age. The study results consolidate the existing body of data regarding the absence of an impact of HPV vaccination on sexual behavior or use of contraceptives.

KEYWORDS:

• Human papillomavirus
• HPV vaccination
• Sexual behavior, cervical cancer screening

Introduction

At the time of implementation of human papillomavirus (HPV) vaccination programs, concerns were raised by some parents and health professionals about HPV vaccination possibly leading to riskier sexual behavior.¹ Although most studies showed no difference between vaccinated and unvaccinated young women sexual behaviors and attendance in cervical cancer screening programs,^1–4 some studies reported differences in barrier contraceptive use, riskier sexual behaviors and higher attendance of cervical cancer screening by unvaccinated women when compared to those vaccinated.^5,6 Other studies reported higher attendance of cervical cancer screening by vaccinated when compared to unvaccinated women.^7,8

Here, we present the results of a study that assessed sexual behaviors, risk of sexually transmitted infections (STI) and attendance to cervical cancer screening program by HPV vaccinated and unvaccinated young women in the province of Quebec (population ≈8.5 mln), Canada.

Material and methods

This study was nested in a larger study on sexual health in young adults aged 17–29 years. A complete study methodological report was published elsewhere.⁹ Information regarding participants’ socio-demographic characteristics, HPV vaccination status, cervical cancer screening attendance, history of STI including anogenital condylomas, and various behavior risk factors were assessed by a computer-assisted questionnaire.

Data collection and study sample

Participants were recruited mainly from post-secondary educational institutions (college, university, and vocational schools) located in different regions representing close to 85% of the Quebec population. A multistage sampling design was used to take into account age, region of residence, and type of institution for students.⁹ Respondents’ main characteristics are presented in Table 1.

Table 1. Main study participants socio-demographic and health characteristics by vaccination status

Characteristic	Vaccinated, N = 1002 (%)*	Unvaccinated, N = 473 (%)*	Total, N = 1475 (%)*
Age at the time of the study (years)**
17–19	564 (56)	41 (9)	605 (41)
20–22	354 (35)	131 (28)	485 (33)
23–29	84 (8)	301 (64)	385 (26)
Born in Canada	935 (93)	423 (89)	1358 (92)
Ethnicity
French Canadian	879 (88)	380 (80)	1259 (85)
Other	117 (12)	92 (20)	209 (15)
Recruited in an educational institution	901 (90)	311 (66)	1212 (82)
The highest completed degree
Primary or secondary school	606 (60)	207 (44)	813 (55)
College	314 (31)	154 (33)	468 (32)
University	62 (6)	103 (22)	165 (11)
Other or missing information	20 (2)	9 (2)	29 (2)
Sexual orientation**
Heterosexual	899 (90)	400 (85)	1299 (88)
LGBTQ	103 (10)	72 (15)	175 (12)
Ever had sex**
With males only	858 (86)	358 (76)	1216 (82)
Both males and females or females only	143 (14)	114 (24)	257 (18)
Drug consumption during sexual relations
Never or a few times	915 (91)	427 (90)	1342 (91)
More than half of time	85 (8)	43 (9)	128 (9)
Age at sexual debut (years)**
9–14	228 (23)	143 (30)	371 (25)
15–16	434 (43)	151 (32)	585 (40)
≥17	308 (31)	169 (36)	477 (32)
Use of contraception measures**
Condom	452 (45)	180 (38)	632 (43)
Hormonal methods	460 (46)	218 (46)	678 (46)
Other methods	17 (2)	15 (3)	32 (2)
Neither method	66 (7)	58 (12)	124 (8)
Number of sexual partners during last 12 months***
1–2	752 (75)	336 (71)	1088 (74)
3–4	129 (13)	71 (15)	200 (14)
≥5	121 (12)	66 (14)	187 (13)
Had a history of a STI**
Yes	104 (10)	132 (28)	236 (16)
No	893 (89)	338 (71)	1231 (83)
Had a history of anogenital warts**
Yes	22 (2.2)	55 (11.6) **	77 (5)
No	967 (97.8)	409 (88.4)	1376 (95)
Had a Pap test**
Yes	509 (51)	362 (77)	871 (59)
No	461 (46)	92 (19)	553 (37)

*Differences in numbers are due to missing answers to some questions. **p < 0.05; ***the only variable presented for last 12 months; all other variables are presented for life long period.

Statistical analysis

Statistical analysis was conducted using SAS, version 9.4 (SAS Institute, Cary, NC). Bivariate analyses were based on Chi-square test or Fisher's exact test as appropriate. Log binomial regression was conducted with inclusion of variables, which were potentially associated with the probability of vaccination or sexual behavior. Variables retained in the final models were those statistically significant (with a p-value <.05) in univariate analyses, after verification for collinearity. Crude and adjusted relative risks were calculated. Statistical significance was based on p < .05 (2 sided). Variables with missing values were excluded from the adjusted analysis.

Results

In this analysis, 1475 out of 2118 questionnaires completed by women aged 17–29 years were included. The main reasons for exclusion of 643 questionnaires were as follows: missing answer to the question regarding HPV vaccination status (n = 180) or sexual activity (n = 30), or not sexually active during the last 12 months (n = 433).

The majority of included respondents (67.9%) were vaccinated against HPV. The proportion of those vaccinated decreased with age: from 93.2% in those aged 17–19 to 72.9% in those aged 20–22, and 21.8% in 23–29-year-olds (p < .0001).

Heterosexual orientation was reported by 90% of vaccinated and 85% of unvaccinated participants (p = .007). The majority of participants reported having had sex only with males: 86% of those vaccinated and 76% of those unvaccinated (p < .0001). The remaining participants reported having had sex with women only or both males and women. A higher proportion of unvaccinated respondents had at least one consensual sexual intercourse (oral, vaginal or anal) under the age of 15 when compared to those vaccinated (30% vs. 23%, p < .0001). The number of sexual partners during the last 12 months was similar between vaccinated and unvaccinated participants (Table 1). For example, the proportion of participants reporting having had more than three sexual partners during the last 12 months was 25% and 29% among vaccinated and unvaccinated participants, respectively (p = .11). Vaccinated participants reported more condom use (45% versus 38%; p = .0002) and less sexually transmitted infections (10% versus 28%; p < .0001) when compared to those unvaccinated. A significantly lower proportion of vaccinated participants reported a diagnostic of anogenital condylomas sometime in their life (2.2% vs. 11.6%; p < .0001). A pap-test screening has been reported by 51% and 77% of vaccinated and unvaccinated participants, respectively (p < .0001) (Table 1).

In multivariate analysis, no difference was detected between HPV vaccinated and unvaccinated participants regarding the number of sexual partners or the use of contraceptives (Table 2). Unadjusted (RR 0.37; 95%CI 0.27–0.53) and adjusted relative risk (RR 0.63; 95%CI 0.44–0.90) of reporting a diagnosis of STI during the last 12 months before the survey was lower in vaccinated respondents. The unadjusted probability of attending cervical cancer screening (Pap test) during the last 12 months was lower among vaccinated respondents (RR 0.57; 95%CI 0.51–0.64). However, this difference disappeared after the adjustment for several factors, including the age of participants (RR 0.98; 95%CI 0.90–1.07) (Table 2).

Table 2. Sexual behavior, clinical outcomes and attendance of cervical cancer screening program by HPV vaccinated and unvaccinated sexually active women (multivariate analysis)

		Relative risk (95%CI)
Variables			Crude	Adjusted*
Sexual behavior	Three or more sexual partners during last 12 months
	Vaccinated		0,86 (0,72–1,03)	0,91^a (0,76–1,09)
	Unvaccinated		Ref.	Ref.
	Use of contraception
	Vaccinated		1.08 (1.04–1.13)	1.01^b (0.97–1.06)
	Unvaccinated		Ref.	Ref.
Sexually transmited infections (STI)	Diagnosis of a STI during last 12 months
	Vaccinated		0,37 (0,27–0,53)	0,63 ^c (0,44–0,90)
	Unvaccinated		Ref	Ref
Vaccination and « Pap test »	Had a Pap test during last 12 last months
	Vaccinated		0,57 (0,51–0,64)	0,98^d (0,90–1,07)
	Unvaccinated		Ref	Ref

* For the adjusted analysis, variables with missing values were excluded from the adjusted analysis: ^a Adjusted for region of recruitment, gender of sexual partners, age at the first intercourse and education level ^b Adjusted for age, gender of sexual partner, having a family physician and education level; ^c Adjusted for age, level of knowledge about STI and number of sexual partners during last 12 months; ^d Adjusted for age, ethnicity, use of contraception, having a family physician, level of knowledge about STI and number of sexual partners during the life.

Discussion

This study was conducted 5 years after the implementation of HPV immunization in the province of Quebec. The participants in this study were not eligible for vaccination in the regular school-based provincial immunization program (grade 4; age 9–11 years) because they were older at that time. However, most participants were eligible for vaccination in a catch-up offer, which was free of charge for girls aged up to 18 years. The significantly higher proportion of vaccinated participants under the age of 23 is clearly indicating that free of charge catch-up offer played an important role in vaccine uptake. The results indicate no difference between vaccinated and unvaccinated young women regarding the number of sexual partners, the use of contraceptive measures or the attendance of cervical cancer screening (when adjusted for age). This is congruent with the results of previous studies conducted generally at shorter intervals after HPV vaccination.^2,4,10–12 However, many of these studies had a relatively low number of participants,^{10,11,13–16} or were conducted in young adolescents.^2,4,12

Lower probability of STIs reported in vaccinated compared to unvaccinated participants observed in our study has been previously reported by a study conducted in the United Kingdom among 14–20 years old females (6). However, in both cases, the number of STIs was relatively small and this observation needs further confirmation by other studies. Our results are congruent with those reported by two recent systematic reviews.^3,17 The authors of one of these reviews³ concluded that they did not find sufficient evidence to support compensatory sexual risk behaviors following HPV vaccination among adolescent girls or women. The conclusions of the second review states that there is a strong body of evidence refuting the association between HPV vaccination and risky sexual behavior. Furthermore, even if risk compensation was identified as an issue related to HPV vaccination, this would not be justification for withholding vaccination, but would argue for effective pre- and post-vaccination counseling.¹⁷

Our study has some strengths, such as a relatively high number of participants (n = 1475) with around two thirds of them previously vaccinated against HPV; inclusion of females from 12 consecutive birth cohorts and from a large geographical area, which includes both urban and rural populations.

The study also has some limitations. First, this is a cross-sectional study that assessed the variables of interest at around 5 years after HPV vaccine administration, and we cannot exclude some differences at longer follow-up. However, the plausibility of sexual behaviors and preventive measures attendance change many years after vaccination and sexual debut is probably low. Second, the participation in the study was on volunteer basis and distribution of vaccination status was not randomly selected. However, recruiters clearly stated that the questionnaires were anonymous and that the study staff or anyone else would not be able to identify the participants. This should have minimized the selection bias and the probability of socially desirable answers. Third, we included in this analysis 70% of all completed questionnaires; 643 questionnaires were excluded because of missing answers to the main questions of interest (sexual activity and HPV vaccination status) or no sexual activity reported. These exclusions were in line with our main objective to assess sexual behaviors and attendance of cervical cancer screening in vaccinated and unvaccinated sexually active young women and explained some differences seen with previously published results on the entire cohort independently of sexual activity status.¹⁸ The main difference between the total study cohort and that included in this analysis (sexually active participants) consists in the proportion of vaccinated and unvaccinated participants aged 17–19 years. The proportion of vaccinated participants in the total cohort was higher (93.2%) when compared to that observed in this analysis (83.5%; p < .0001). This observation might suggest a higher vaccine coverage in young sexually non-active girls when compared to sexually active during the last 12 months. This observation needs further assessment in larger cohorts.

However, we think that the inclusion in the analysis of almost 1500 eligible questionnaires completed by sexually active participants suggests that the results are robust and allow for the comparison of the main variables of interest in vaccinated and unvaccinated participants. Fourth, the answers to some questions were missing. The proportion of missing answers to any question in the questionnaires included in this analysis was below 3%. As such, the small proportion of missing answers should have had little, if any impact on, result validity, even if participants who did not provide all answers were different from the rest of participants. Finally, the participants in this study were vaccinated during a catch-up vaccination. As previously reported, the proportion of those who were sexually active at the time of vaccination was 17%, 19% and 84% in 17–19, 20–22 and 23–29 years old, respectively.¹⁸ Routine HPV vaccination programs are targeting mainly 9–12 years old. This younger age group is not expected to be sexually active yet and not exposed to HPV. As such, the results of our study should be cautiously interpreted in the context of routine vaccination programs. However, it is reassuring that our study results are not divergent from those conducted in females vaccinated at a younger age during routine vaccination programs.^2,4,19

In summary, our study results consolidate the existing body of data regarding the absence of an impact of HPV vaccination on sexual behavior, use of contraceptive measures or attendance to cervical cancer screening. In fact, the observed lower risk of reporting a history of STI in vaccinated young women when compared to unvaccinated suggests less risky sexual behavior in this population, a few years post-vaccination.

Disclosure of potential conflicts of interest

F. Coutlee has received grants through his institution from Merck Sharp and Dome and Roche, as well as honoraria from Merck and Roche for lectures on human papillomavirus (HPV). PG, CS, VG, FD, GL and SMC declare no conflicts of interest.

Ethics

Ethics approval for this study was received from the research ethics boards of the Agence de la Sante et des Services sociaux de Montreal and the Centre hospitalier de l’Universite de Montreal.

Acknowledgments

Julie Guenoun, B. Sc. Émilie Cométe, M. Sc. Laboratoire de virologie moléculaire, Centre de recherche du Centre hospitalier de l’Université de Montréal (CRCHUM); Patricia Goggin, Médecin retraité et Marie-Hélne Mayrand, MD, PhD, Centre de recherche du Centre hospitalier de l'Université de Montréal (CRCHUM). The authors also want to thank Anne-Marie Berard, who coordinated the data collection and all the other contributors and the participants.

Additional information

Funding

This analysis was funded by the Ministére de la Sante et des Services Sociaux du Quebec.