Genital human papillomavirus (HPV) is the most common sexually transmitted infection (STI) in the United States with an estimated 6.2 million new HPV infections, occurring every year, and an estimated 80% of sexually active women acquiring genital HPV by the age of 50 years.1,2 Although the majority of infections are benign and cause no symptoms, persistent infection with high-risk HPV (HR-HPV) types can cause cervical cancers in women and other types of anogenital cancers among both men and women.3–5 HPV 16 and 18 have most commonly been associated with cervical cancer with approximately 70% of cervical cancers worldwide being attributed to these 2 types of HPV.5 Other high-risk types also considered oncogenic include Types 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, 68, 73, and 82.3,4 In addition to cervical cancer, these oncogenic HPV types have also been associated with anogenital cancers, such as cancer of the vulva, vagina, penis, and anus.6,7
Nearly all studies of risk factors of HPV infection have shown that the most consistent predictors of infection relate to measures of sexual activity. Specifically, number of sex partners, combining sex with alcohol, substance abuse, and partner and partnership characteristics have been independently associated with HPV infection.8–16 However, no previous studies have examined the role of concurrency as a risk factor for HPV infection. Concurrency is defined as having overlapping (or simultaneous) sexual partners (i.e., when someone forms a new partnership while they are in an existing partnership).17,18 Compared to serial monogamy, even in situations where there are many sequential partners, concurrent partnerships allow for more rapid spread of STIs given that an individual infected by 1 partner already has another partner to infect. This has been shown to play a fundamental role in the spread of HIV and bacterial STIs including Chlamydia trachomatis, Neisseria gonorrhoeae, and syphilis, even after adjusting for other risk factors such as number of partners.19–23 In addition, mathematical models of concurrency have suggested that concurrent partnerships may play a significant role in the HIV epidemic, influencing the rapidity of spread of HIV as well as the total number of individuals who become infected.17,24,25 The role of concurrency in the transmission of HPV is less clear. Furthermore, unlike HIV infection, HPV infections may clear and is not always persistent. In fact, 70% of new HPV infections clear within 1 year and approximately 90% clear within 2 years, with a median duration of infection of 8 months.8,26,27
The prevalence of concurrent partnerships is relatively high and varies by setting and duration of concurrency, with estimates ranging from 12% to 54%.20,22,28–30 Furthermore, there is evidence to suggest that the prevalence of concurrent sexual partnerships varies significantly by race and ethnicity, with estimates ranging from 6% in Asians to 21% among non-Hispanic blacks.28 Likewise, the prevalence of HR-HPV is high and varies by race and ethnicity, with estimates ranging from 17% among Asian women and 25% among African American women.31 Given the relatively high prevalence and incidence of HPV, including high-risk types of HPV and the relatively common practice of concurrency, there is a need to examine the association between concurrency as a risk factor for HR-HPV infection, independent of other known risk factors and across different racial and ethnic groups. This is particularly relevant given the recent suggestion that the availability of a vaccine against HPV Types 16 and 18, which may be an effective method of prevention for these types, may in fact lead to an increase in the prevalence of other high-risk types of HPV for which there is no effective vaccine.32 A better understanding of the role of concurrency as a risk factor for HR-HPV would not only allow for targeted prevention strategies, but can also help to clarify the role of concurrency in transmission dynamics of HPV. In this study, we examine the association between HR-HPV infection and reported concurrent partnerships among a diverse group of women attending different clinic types, including a family planning clinic, a primary care clinic, and 2 public STI clinics. We hypothesize that the prevalence of concurrency will vary by race/ethnicity and is higher among women who test positive for high-risk HPV, even after controlling for other risk behaviors.
MATERIALS AND METHODS
Study Population and Design
The HPV Sentinel Surveillance (HSS) study was a multisite 3-year surveillance study led by the Centers for Disease Control and Prevention (CDC) and designed to measure HR-HPV prevalence among women from the cervical screening population throughout the United States. Details of the study including enrollment and recruitment have been described previously.31 Women were recruited from 26 clinics in 6 US cities including Boston, MA; Baltimore, MD; Denver, CO; New Orleans, LA; Seattle, WA; and Los Angeles, CA. Participants were enrolled from January 1, 2003 to December 31, 2005, and the methods for enrollment, data collection, and laboratory testing were standardized through a common surveillance protocol. Demographic and behavioral data were collected through a standardized questionnaire and cervical specimens were collected for HR-HPV testing and typing. Women enrolled in 3 of the study sites – Denver, CO, Los Angeles, CA, and Seattle, WA—completed the standard study questionnaire, but from 2004, they also responded to additional questions about their sexual partners and concurrency with their most recent sex partner. Furthermore, given our interest in examining the epidemiology of HR-HPV among different race/ethnic groups, women in Los Angeles were recruited from sites which would allow for an adequate number of participants per race/ethnic group. Specifically, participants were recruited from STI and family planning clinics, as well as a primary care clinic which served a predominantly Asian population. This analysis is of the 812 women who were enrolled in Los Angeles, CA, and completed the questionnaire after the concurrency questions were added in 2004.
Women were eligible to participate if they were 18 to 65 years of age, needed routine cervical screening, and did not have a Papanicolau (Pap) test in the past 12 months. Those eligible were offered participation after informed consent and the study was approved by the human subjects review committees at CDC, Los Angeles County Department of Public Health, and University of California, Los Angeles.
All women had cervical samples collected for cervical screening (Pap testing) at the time of cervical examination. After providers collected a cervical specimen for Pap testing, an additional cervical sample of ecto- and endocervical cells was also collected for HPV testing using the Digene Cervical Sampler (Digene cervical brush and specimen transport medium; Digene, Gaithersburg, MD). The samples were tested for the presence of high-risk HPV DNA by both Hybrid Capture 2 (HC2) assay and L1 consensus polymerase chain reaction. The HC2 high-risk probe, which was used by the local laboratory, renders a positive result if it detects the presence of any 1 of the 13 HR-HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and/or 68). Given that type specific information is not provided by the HC2 assay, an aliquot of all HC2 samples were sent to CDC for HPV typing using L1 consensus polymerase chain reaction. The Roche line blot assay (Roche Molecular Systems, Pleasanton, CA) was used to detect the specific type of HPV which includes the 13 types detected by HC2 as well as other HR-HPV types including 26, 53, 66, 73, and 82. Concomitant infection with other STIs was also assessed. Results for those tested for C. trachomatis, N. gonorrhoeae, and/or Trichomonas vaginalis during the same clinic visit that HPV status was assessed were abstracted from the patient's medical record.
Demographic and behavioral data were collected using an interviewer administered 42-item questionnaire. In addition to sociodemographic characteristics, including age, race/ethnicity, marital status, and country of origin, participants were also asked about their sexual history. Age at first sex, number of sex partners (lifetime and past 6 months), and demographic characteristics of current sex partner was assessed. Sexual partnership concurrency was assessed over the duration of the last sexual partnership. Women were asked to report whether they had had sexual activity with anyone other than their current partner (individual concurrency) after the current partnership had been established. Participants were also asked if they thought their current partner had sexual activity with others after the partnership had been established (partner concurrency). Specifically, women were asked the following 2 questions: (1) “Since the first time you had sex with him (the last sex partner), have you had sex with anyone else?” (2) “Since the first time you had sex with him (the last sex partner), do you think he has had sex with anyone else?” Women could respond as either “Yes,” “No,” or “Don't Know.” While individual concurrency was intended to describe the actions of the respondent alone, partner concurrency describes the actions of the partner alone (as perceived by the respondent).
Starting in October 2004 women enrolling in study sites in Los Angeles answered additional questions pertaining to partner concurrency status. Data for these analyses were based on 812 of the 1778 women enrolled in Los Angeles who completed this questionnaire. The women included in this analysis did not differ significantly with respect to age, race/ethnicity, marital status, and age at first sex from women enrolled before the addition of the partner concurrency items to the questionnaire (n = 966; P >0.05, data not shown).
Descriptive statistics including means and frequency distributions were conducted for the total sample and by HR-HPV status for each race/ethnic group. Differences were evaluated using t test for continuous variables and chi-square methods for categorical variables. Given that HR-HPV—our outcome of interest—is considered a common outcome event (prevalence of 10% or more) logistic regression analysis results in biased estimates of risk ratios, with an increasing differential between the risk ratio and the odds ratio. Consequently, we used Poisson regression models with a robust error variance to examine associations between HR-HPV status and other factors including concurrency.33–35 Two different multivariate models were examined based on the concurrency measure (individual concurrency and partner concurrency, as described above). In addition, separate models were analyzed for each race/ethnic group to examine race/ethnicity as an effect modifier of the association between concurrency and HR-HPV. Variables tested for inclusion in the multivariate models for each race/ethnicity differed and were on the basis of findings from the stratified univariate analyses and factors previously associated with HPV infection. In addition, deviance tests and score statistics for Type 3 GEE analysis were used to achieve the best model. All analyses were conducted using SAS version 9.1 (SAS Institute Inc, Cary, NC).
Characteristics of Participants
The mean age of participants was 34.2 years (range: 18–65), with 31.8% identifying as African American and 28.4% as Hispanic. Data collection at a community clinic providing care for a predominantly Asian (Korean) population provided a substantial number of Asians (32.8%) to the overall dataset. The median age at first sex was 17 years (range: 10–41) with a median of 4 (range: 1–500) reported number of lifetime partners. Overall, 22.6% of women reported having concurrent sexual partners, with 36.5% reporting that their partners had concurrent partnerships (Fig. 1). Participant recruitment across clinic type included 49.2% from sexually transmitted disease clinics (predominantly African American and Hispanic clients), 32.1% from primary care clinics (predominantly Asian), and 18.7% from family planning clinics (mostly Hispanic).
Prevalence of HPV Infection and Types
The overall prevalence of HR-HPV in our sample was 21.7%. The prevalence of HR-HPV was highest among women in the youngest age groups with a prevalence of 41.5% among those aged 18 to 19 years and 32.4% in those aged 20 to 29. The most common type of HR-HPV identified was HPV 16 (4.7%), followed by HPV 51 (3.0%) and HPV 59 (2.7%). The majority of women were infected with only 1 type of HR-HPV, although 34.9% were infected with more than one type and as many as 5 types of HR-HPV.
Factors Associated With HPV Infection
Compared to participants without HR-HPV infection, women with infection were more likely to be younger (mean age, 29 years vs. 36 years, P <0.001). The prevalence of HR-HPV was higher among those younger at first sex and those who reported more sex partners (lifetime and past 6 months). Likewise, HR-HPV was identified among a greater percent of participants who were diagnosed with a concomitant STI (C. trachomatis, N. gonorrhoeae, and/or T. vaginalis) (32.1% vs. 21.3%; P = 0.015). Women who reported having a concurrent partnership (i.e., having sexual activity with another partner after their current partnership had been established) were also more likely to have HR-HPV as compared to women who reported no concurrency (29.4% and 19.3%, respectively; P = 0.004). We also found that the relationship between demographic factors, sexual risk behaviors, and HR-HPV varied by race/ethnicity. For instance, among Asian and Hispanic women the prevalence of HR-HPV was higher among those who were US born and varied by number of sex partners and having a new sex partner in the past 6 months (Table 1).
In order to examine the association between concurrency and HR-HPV by race/ethnic group, stratified analyses were conducted. In addition, 2 different multivariate models were considered in order to determine the association between HR-HPV status and the different measures of concurrency (individual concurrency and partner concurrency). Among Hispanic women, even after controlling for other risk factors, those who reported concurrency were nearly twice as likely to have HR-HPV as compared to those who did not report concurrency (Table 2). In addition, the partner's concurrency status was also associated with HR-HPV. Although a similar trend was observed among Asian women, based on multivariate analysis, individual concurrency was not predictive of HR-HPV (AOR = 1.47; 95% CI: 0.42–5.14). Among African American women, after adjustment for age, age at first sex, and lifetime number of sex partners, those who reported concurrency were less likely to have HR-HPV as compared to those who did not report concurrency (AOR = 0.60; 95% CI: 0.37–0.98). On the basis of the univariate analysis other factors considered but not included in the final multivariate model included presence of other STIs, being foreign born or having a foreign born partner, age discordance with partner, race/ethnic discordance with partner, having a new sex partner, and clinic type.
In this cross-sectional study of a diverse group of women attending different clinic types we found a relatively high prevalence of HR-HPV (21.7%) and partnership concurrency (22.6%–36.5%). Furthermore, we found that women diagnosed with HR-HPV shared characteristics and behavioral profiles of women diagnosed with other sexually transmitted infections. In general, women with HR-HPV were younger and reported more recent and lifetime sexual partners. Yet HR-HPV was more prevalent than the other STIs measured in this study, including C. trachomatis (8.7%), N. gonorrhoeae (2.6%), and T. vaginalis (6.5%) and was also found among women who do not report practice of risk behaviors. For instance, 12% of women who reported only 1 sex partner in their lifetime had HR-HPV infection. Although the role of sexual networks in the transmission of HR-HPV has been unclear, these findings highlight the potential role of sexual partnership concurrency in the transmission dynamics of HR-HPV, given that a high prevalence of concurrent partnerships ensures continuing infection, even in low-risk groups.36 These results reiterate that it is not just individual behavior, but the behavior of one's sexual partner that affects the acquisition of STIs, including HPV.
Our findings also suggest that the relationship between concurrency and HR-HPV maybe different depending on race/ethnicity. Among Hispanic women, the finding that a woman in a concurrent sexual partnership was more likely to be diagnosed with HR-HPV suggests that it is not only her exposure to multiple partners, but the timing of this exposure (concurrently vs. serial monogamy) that is associated with HPV. Furthermore, while perceiving her partner as having a concurrent partnership was not associated with HR-HPV, Hispanic women who were unsure of their partner's concurrency status were also more likely to be diagnosed with HR-HPV infection. Given evidence that those who are unaware of their partner's concurrency status are more likely to have partners in concurrent relationships,22 our data suggest that it is the partners' risk behavior as well as her own that increases the likelihood of acquiring HR-HPV infection. This replicates earlier findings on partner concurrency for chlamydial infection and reinforces the importance of measuring partner behavior as well as a woman's own behavior in studying STI epidemics.19
In contrast, African American women who reported having concurrent partners were less likely to be diagnosed with HR-HPV as compared to those who did not report concurrency. This association, which is counter to what we hypothesized and what we found among Hispanic women, suggests that there may be differences by race/ethnicity in the individual or partnership characteristics of those who report concurrency. For instance, on an individual level both African American and Hispanic women who reported concurrency had a similar age of sexual debut and lifetime number of partners. However, African American women who reported concurrency were older (mean age: 32.3 years vs. 27.8 years) and less likely to be diagnosed with Chlamydia and gonorrhea as compared to Hispanic women who reported concurrency (Chlamydia prevalence: 11.5% vs. 20.0%, respectively; gonorrhea prevalence: 0% vs. 4.4%, respectively). On a partnership level, African American women who reported concurrent partnerships were less likely to report partnerships discordant by race/ethnicity as compared to Hispanic women (4.9% vs. 36.0%). Partnerships discordant on race have been associated with increased rates of STIs.30,37 Furthermore, other studies have shown that concurrency and discordance coexist and have suggested that the combination of these 2 factors may be multiplicative on the transmission of STIs within networks.38
There are several limitations to this study. The behavioral information collected as part of this study was based on self-reports using an interviewer administered questionnaire. Reporting bias could have been minimized through the use of computer-based interviews, which improves reporting of sensitive behaviors.39,40 In addition, the partner concurrency status was on the basis of the woman's assessment and was not validated by any information collected by the partner, with evidence suggesting limited accuracy of reports of partner behavior.22,41 Our study collected data from a few community sites, which served specific clients at each site and included very few white women, which diminished our ability to consider differences by race/ethnicity and was limited to comparisons between minority women. Additionally, these clinics did not capture a representative sample of women in Los Angeles, but rather settings where screening for HR-HPV is likely to occur—community clinics, family planning clinics, and STI clinics. Finally, the lack of concurrency questions at other HPV surveillance sites limits the generalizability of the findings. It would be of interest to determine if the same pattern prevails in other geographic settings where there may be different patterns of sexual mixing as well as HR-HPV prevalence by race/ethnicity group.
In conclusion, this study revealed that concurrent sex partnerships were associated with HR-HPV. However these associations varied by race/ethnicity. While no associations were found among Asian women, concurrent African American women were less likely to have HR-HPV and concurrent Hispanic women were more likely to have HR-HPV. This suggests the impact of concurrent sex partners not only depends on the network that an individual is connected with but also on the individual level factors such as condom use. The finding that the partner's concurrency can place an individual at higher risk for HR-HPV reiterates that this STI, like many others, is transmitted between sexual partners and it is not the behavior of one, but of both, that influences transmission.
1.Myers ER, McCrory DC, Nanda K, et al. Mathematical model for the natural history of human papillomavirus infection and cervical carcinogenesis. Am J Epidemiol 2000; 151:1158–1171.
2.Weinstock H, Berman S, Cates W Jr. Sexually transmitted diseases among American youth: Incidence and prevalence estimates, 2000. Perspect Sex Reprod Health 2004; 36:6–10.
3.Munoz N, Bosch FX, de Sanjose S, et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med 2003; 348:518–527.
4.Walboomers JM, Jacobs MV, Manos MM, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999; 189:12–19.
5.Bosch FX, de Sanjose S. Chapter 1: Human papillomavirus and cervical cancer–burden and assessment of causality. J Natl Cancer Inst Monogr 2003; 3–13.
6.Cogliano V, Baan R, Straif K, et al. Carcinogenicity of human papillomaviruses. Lancet Oncol 2005; 6:204.
7.Parkin DM. The global health burden of infection-associated cancers in the year 2002. Int J Cancer 2006; 118:3030–3044.
8.Ho GY, Bierman R, Beardsley L, et al. Natural history of cervicovaginal papillomavirus infection in young women. N Engl J Med 1998; 338:423–428.
9.Moscicki AB, Hills N, Shiboski S, et al. Risks for incident human papillomavirus infection and low-grade squamous intraepithelial lesion development in young females. JAMA 2001; 285:2995–3002.
10.Koutsky L. Epidemiology of genital human papillomavirus infection. Am J Med 1997; 102:3–8.
11.Karlsson R, Jonsson M, Edlund K, et al. Lifetime number of partners as the only independent risk factor for human papillomavirus infection: A population-based study. Sex Transm Dis 1995; 22:119–127.
12.Peyton CL, Gravitt PE, Hunt WC, et al. Determinants of genital human papillomavirus detection in a US population. J Infect Dis 2001; 183:1554–1564.
13.Manhart LE, Holmes KK, Koutsky LA, et al. Human papillomavirus infection among sexually active young women in the United States: Implications for developing a vaccination strategy. Sex Transm Dis 2006; 33:502–508.
14.Trottier H, Franco EL. The epidemiology of genital human papillomavirus infection. Vaccine 2006; 24(suppl 1):S1–S15.
15.Burk RD, Ho GY, Beardsley L, et al. Sexual behavior and partner characteristics are the predominant risk factors for genital human papillomavirus infection in young women. J Infect Dis 1996; 174:679–689.
16.Burk RD, Kelly P, Feldman J, et al. Declining prevalence of cervicovaginal human papillomavirus infection with age is independent of other risk factors. Sex Transm Dis 1996; 23:333–341.
17.Morris M, Kretzschmar M. Concurrent partnerships and the spread of HIV. AIDS 1997; 11:641–648.
18.Garnett GP, Johnson AM. Coining a new term in epidemiology: Concurrency and HIV. AIDS 1997; 11:681–683.
19.Potterat JJ, Zimmerman-Rogers H, Muth SQ, et al. Chlamydia transmission: Concurrency, reproduction number, and the epidemic trajectory. Am J Epidemiol 1999; 150:1331–1339.
20.Rosenberg MD, Gurvey JE, Adler N, et al. Concurrent sex partners and risk for sexually transmitted diseases among adolescents. Sex Transm Dis 1999; 26:208–212.
21.Koumans EH, Farley TA, Gibson JJ, et al. Characteristics of persons with syphilis in areas of persisting syphilis in the United States: Sustained transmission associated with concurrent partnerships. Sex Transm Dis 2001; 28:497–503.
22.Drumright LN, Gorbach PM, Holmes KK. Do people really know their sex partners? Concurrency, knowledge of partner behavior, and sexually transmitted infections within partnerships. Sex Transm Dis 2004; 31:437–442.
23.Adimora AA, Schoenbach VJ, Martinson FE, et al. Heterosexually transmitted HIV infection among African Americans in North Carolina. J Acquir Immune Defic Syndr 2006; 41:616–623.
24.Kretzschmar M, Morris M. Measures of concurrency in networks and the spread of infectious disease. Math Biosci 1996; 133:165–195.
25.Watts CH, May RM. The influence of concurrent partnerships on the dynamics of HIV/AIDS. Math Biosci 1992; 108:89–104.
26.Molano M, Van den BA, Plummer M, et al. Determinants of clearance of human papillomavirus infections in Colombian women with normal cytology: A population-based, 5-year follow-up study. Am J Epidemiol 2003; 158:486–494.
27.Franco EL, Villa LL, Sobrinho JP, et al. Epidemiology of acquisition and clearance of cervical human papillomavirus infection in women from a high-risk area for cervical cancer. J Infect Dis 1999; 180:1415–1423.
28.Adimora AA, Schoenbach VJ, Bonas DM, et al. Concurrent sexual partnerships among women in the United States. Epidemiology 2002; 13:320–327.
29.Ford K, Sohn W, Lepkowski J. Am adolescents: Sexual mixing patterns, bridge partners, and concurrency. Sex Transm Dis 2002; 29:13–19.
30.Manhart LE, Aral SO, Holmes KK, et al. Sex partner concurrency: Measurement, prevalence, and correlates among urban 18–39-year-olds. Sex Transm Dis 2002; 29:133–143.
31.Datta SD, Koutsky LA, Ratelle S, et al. Human papillomavirus infection and cervical cytology in women screened for cervical cancer in the United States, 2003–2005. Ann Intern Med 2008; 148:493–500.
32.Garland SM, Hernandez-Avila M, Wheeler CM, et al. Quadrivalent vaccine against human papillomavirus to prevent anogenital diseases. N Engl J Med 2007; 356:1928–1943.
33.Greenland S. Model-based estimation of relative risks and other epidemiologic measures in studies of common outcomes and in case-control studies. Am J Epidemiol 2004; 160:301–305.
34.McNutt LA, Wu C, Xue X, et al. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol 2003; 157:940–943.
35.Zou G. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004; 159:702–706.
36.Anderson R. Transmission dynamics of sexually transmitted infections. In: Holmes KK, St Louis ME, Mardh PA, eds. Sexually Transmitted Disease. New York, NY: McGraw-Hill, 1999:25–38.
37.Aral SO, Hughes JP, Stoner B, et al. Sexual mixing patterns in the spread of gonococcal and chlamydial infections. Am J Public Health 1999; 89:825–833.
38.Gorbach PM, Drumright LN, Holmes KK. Discord, discordance, and concurrency: Comparing individual and partnership-level analyses of new partnerships of young adults at risk of sexually transmitted infections. Sex Transm Dis 2005; 32:7–12.
39.Metzger DS, Koblin B, Turner C, et al; HIVNET Vaccine Preparedness Study Protocol Team. Randomized controlled trial of audio computer-assisted self-interviewing: Utility and acceptability in longitudinal studies. Am J Epidemiol 2000; 152:99–106.
40.Newman JC, Des J, Turner CF, et al. The differential effects of face-to-face and computer interview modes. Am J Public Health 2002; 92:294–297.
41.Stoner BP, Whittington WL, Aral SO, et al. Avoiding risky sex partners: Perception of partners' risks v partners' self reported risks. Sex Transm Infect 2003; 79:197–201.