Dinh, Thu-Ha MD*†; Sternberg, Maya PhD*; Dunne, Eileen F. MD*; Markowitz, Lauri E. MD*
GENITAL WARTS, also called condylomata acuminata, are a common sexually transmitted disease (STD) in the United States. It is estimated that 1% of sexually active adults in the United States have genital warts.1 Genital warts can cause discomfort, anxiety, embarrassment, anger, and shame and can interfere with relationships.2 The annual direct health care costs of genital warts in the United States are estimated to be $200 million, with the cost of a successful course of treatment to be $436.3
Human papillomavirus (HPV) types 6 and 11 cause more than 90% of genital warts.4,5 Recently, a quadrivalent HPV vaccine was shown in clinical trials to prevent HPV types 6 and 11 associated genital warts.6,7 This vaccine was licensed in June, 2006 for use in females 9 to 26 years of age and recommended for routine immunization of 11-to 12-year-old girls.8 Widespread use of the vaccine would likely decrease the occurrence of genital warts.
There are no representative data on genital warts in the US population. Using data collected in the National Health and Nutrition Examination Survey (NHANES) from 1999–2004, we determined the percentage of sexually active persons ages 18 to 59 years who reported having a history of genital warts diagnosed by a health care provider. In addition, we identify factors associated with a reported history of genital warts in this population.
Materials and Methods
Study Populations and Study Design
NHANES is a series of cross-sectional national surveys conducted by the National Center for Health Statistics (NCHS), Centers for Disease Control and Prevention (CDC). Details of the survey methods have been published previously.9 Briefly, during each survey, a random sample of the US civilian noninstitutionalized population was selected using a complex, multistage probability cluster design. Some populations, such as adolescents, non-Hispanic blacks, and Mexican Americans were oversampled.
During NHANES 1999–2004, the questionnaire module with questions on history of genital warts and sexual behaviors was administered to persons aged 18 to 59 years in the mobile examination center where questions about sexual behavior were asked using audio computer-assisted self interview (ACASI). A total of 11,454 persons aged 18 to 59 years were interviewed and 10,887 (95%) were examined. Of those, 2038 (18%) were considered nonresponders as they either had a missing value (1103) or answered “no” to the question “have you ever had sex,” (747) or had a missing value, refused, or answered “don’t know” to the question “Has a doctor or healthcare professional ever told you that you had genital warts.”(188) A total of 8849 (77%) responders were included in our analysis.
NHANES was approved by the institutional review board of CDC. Written informed consent was obtained from survey participants.
Statistical analyses were carried out using SAS (version 9.1, SAS Institute, Cary, NC) and SAS callable SUDAAN (RTI, Research Triangle Park, NC) to account for the NHANES survey design. We estimated percentage of persons who reported having been diagnosed with genital warts by a health care provider and sociodemographic and behavioral characteristics. All estimates were weighted to represent the 18- to 59-year-old civilian, noninstitutionalized population in the United States and to account for oversampling and nonresponse. The 6-year examination weights were created by taking two-third of the 4-year examination weights, made available for 1999–2002 and adding it to one-third of the 2-year examination weights made available as part of the 2003–2004 cycle. The weighting methodology has been described previously.9 We investigated the impact of nonresponse by identifying significant demographic predictors of the response propensity through a logistic regression model. The predicted probabilities from this model along with a poststratification step were used to further adjust the weights. The estimates using further adjusted weights changed slightly but were within the confidence intervals for the percentage estimates based on the standard weights. After this investigation, we decided to use the standard weights provided by NCHS for this article.
Taylor series linearization was used to calculate variance estimates.10 Confidence intervals were calculated using a log transformation with the standard error of the log percentage based on the δ method and applying SUDAAN estimated standard errors.11 The association between having been diagnosed with genital warts and the demographic or behavioral characteristics was based on the Wald χ2 statistic at α = 0.05.
For our multivariate analysis, logistic regression methods were used to study the independent association between a history of diagnosed genital warts and sociodemographic or behavioral characteristics. Goodness of fit for the model was assessed using Hosmer-Lemeshow Goodness of Fit Satterthwaite-adjusted F-test. All variables that had a P value of ≤0.10 in bivariate analysis were included in the multivariate model and eliminated using a backward elimination approach, such that all the remaining variables had a Satterthwaite-adjusted F-test P value ≤0.05. Confounding was assessed at each step by confirming, after the removal of a given variable, the subsequent model’s coefficients did not change by more than 30%. When all variables in the model were statistically significant at α = 0.05 and did not meet entry criteria as a confounder, all pairwise interactions in the model were explored. When an interaction was entered into the model the main effects variables were always retained, regardless of statistical significance. A pairwise interaction was retained only if the overall P value for the interaction was <0.05; however, none qualified for inclusion in the final model. Findings from 2 logistic regression models conducted separately for females and males did not have statistically significant differences; therefore, we presented results from the model using both women and men.
From 1999–2004, 5.6% [95% confidence interval (CI), 4.9%–6.4%] of sexually active 18-to 59-year olds in the United States reported that they had ever been diagnosed with genital warts by a health care provider (Table 1). The percentage was higher in women (7.2%, 95% CI, 6.2%–8.4%) than in men (4%, 95% CI, 3.2%–5.0%) (Table 1). The overall percentage of persons reporting having been diagnosed with genital warts increased with age, peaking at 7.7% (95% CI 6.1%–9.6%) in the 35- to 44-year-old age group, and then declining to 4.4% (95% CI, 3.6%–5.4%) in the oldest age group (Fig. 1). The percentage of women with a history of genital warts peaked in the 25- to 34-year-old age group (10.4%, 95% CI, 8.2%–13.1%); the percentage of men peaked in the 35- to 44-year-old age group (6.0%, 95% CI, 4.2%–8.6%) (Fig. 1).
The percentage of persons reporting a diagnosis of genital warts varied significantly by other sociodemographic characteristics. Differences by race or ethnicity were observed; the overall percentage was highest (6.6%, 95% CI, 5.7%–7.7%) among non-Hispanic whites, followed by non-Hispanic blacks (4.1%, 95% CI, 3.4%–5.0%), and then Mexican Americans (2.5%, 95% CI, 1.8%–3.4%) (Table 1). Similar differences by race or ethnicity were found among both men and women. The percentage of circumcised men reporting a diagnosis of genital warts was significantly higher than uncircumcised men, 4.5% (95% CI, 3.6%–5.6%) versus 2.4% (95% CI, 1.5%–4.0%) (Table 1).
The percentage of persons reporting a diagnosis of genital warts was greater with increased number of lifetime sex partners:1.7% (95% CI, 1.1%–2.7%) with 1 to 2 partners, 3.1% (95% CI, 2.3%–4.1%) with 3 to 5 partners, 7.2% (95% CI, 6.0%–8.8%) with 6 to 10 partners, and 10.7% (95% CI, 8.9%–12.8%) with more than 10 sex partners (Table 2). This pattern was similar when stratified by sex. The percentage of persons reporting a diagnosis of genital warts was significantly higher among those who had ever used cocaine or street drugs compared with those who never did, 10.5% (95% CI, 8.7%–12.7%) versus 4.4% (95% CI, 3.7%–5.1%) (Table 2). There was a higher percentage of wart diagnoses among women who reported having ever had sex with another woman (12.3%; 95% CI, 7.9%–19.4%) compared with those who did not (6.9%, 95% CI, 5.9%–8.0%) (Table 2).
In multivariate analysis, factors associated with history of ever having been diagnosed with genital warts were female sex, age, race or ethnicity, increasing number of lifetime sex partners, and history of ever having used cocaine or street drugs (Table 3). We found no other confounding or effect modification; therefore, no other factors were included in our final logistic regression model. Women were almost 3 times as likely to report having been diagnosed with genital warts as males [adjusted odds ratios (AOR) = 2.9; 95% CI, 2.1–3.8] (Table 3). Having been diagnosed with genital warts was most strongly associated with the number of lifetime sex partners. Persons with more than 10 lifetime sex partners were 8 times as likely to report ever having a diagnosis of genital warts compared with those who reported having one or two lifetime sex partners (AOR = 7.6; 95% CI, 4.1–13.9) (Table 3).
This is the first national, representative study of genital warts in the United States. Overall, 5.6% of the sexually active persons aged 18 to 59 years reported that they had been diagnosed with genital warts. Experts have estimated that 1% of persons (1000 per 100,000) aged 18 to 45 years in the United States have prevalent genital warts.1 Recent studies of privately insured patients revealed estimates for genital warts of 0.17% (170 per 100,000) and 0.21% (205 per 100,000) in 2000 and 2001, respectively, in the United States.12,13 NHANES did not include a question about current warts, so we can not directly compare data from NHANES to those estimates.
We found that sex, age, race or ethnicity, lifetime number of sex partners, and history of ever having used cocaine or street drugs were significantly associated with ever having been diagnosed with genital warts among 18- to 59-year-old men and women. Our data show that the percentage of persons who reported having a diagnosis of genital warts was significantly higher in non-Hispanic whites than non-Hispanic blacks. In contrast, a recent study of HPV-11 antibody found higher seroprevalence of HPV-11 in non-Hispanic blacks (particularly among women).14 This suggests that the differences in the prevalence of wart diagnoses we found, may be due to differences in health care-seeking behavior. Few studies have described racial or ethnic differences with regard to genital warts. Because our data are based on self-report of genital warts diagnosis and not on actual detection of genital warts, the study estimates may reflect recognition and factors influencing detection of genital warts, in addition to acquisition. For example, variation by sociodemographic groups, including race or ethnicity, could be due to differences in health care access.
Overall, we found that the percentage of persons reporting having been diagnosed with genital warts increased through age 35 to 44 years and then decreased. The same pattern was maintained in the multivariate analysis. The decrease in the oldest age group may be explained by a combination of unmeasured factors. As discussed above, older people may have forgotten a diagnosis of genital warts that occurred when they were younger. In addition, age-cohort differences in incidence of warts and/or temporal changes in health care-seeking behaviors among this group could partially explain these findings.
Adjusting for other factors, we found that 3 times more women reported ever having a diagnosis of genital warts than men. The finding contrasts with data from previous reports, which found the prevalence of genital warts to be similar among men and women.3,13,15 Differences by sex in our study could be due to actual differences in the risk of warts; differences in genital wart diagnosis, differences in health care access, or recognition of warts. Our data should be interpreted in the context of several important limitations. First, our findings could be underestimates because it is possible that some people may have forgotten being diagnosed with genital warts, or had genital warts and never went to a provider. Second, because questions regarding sexual behavior were limited on the survey, all sexual and behavioral risk factors related to genital warts may not have been captured. Finally, although ACASI can enhance the complete reporting of sexual and drug use behaviors, our findings may be subject to bias due to under reporting of behavioral risk factors.
These are the first national data on the burden of genital warts in the United States. Overall, 5.6% of 18- to 59-year-old men and women in the United States reported a history of ever having a genital warts diagnosed by a health care provider. The substantial burden of genital warts in the United States could be reduced by a prophylactic HPV vaccine to types 6 and 11.
1. Koutsky L. Epidemiology of genital human papillomavirus infection. Am J Med 1997; 102:3–8.
2. Ireland JA, Reid M, Powell R, et al. The role of illness perceptions: Psychological distress and treatment-seeking delay in patients with genital warts. Int J STD AIDS 2005; 16:667–670.
3. Insinga RP, Dasbach EJ, Elbasha EH. Assessing the annual economic burden of preventing and treating anogenital human papillomavirus-related disease in the US: Analytic framework and review of the literature. Pharmacoeconomics 2005; 23:1107–1122.
4. Brown DR, Schroeder JM, Bryan JT, et al. Detection of multiple human papillomavirus types in Condylomata acuminata lesions from otherwise healthy and immunosuppressed patients. J Clin Microbiol 1999; 37:3316–3322.
5. Lacey CJN, Lowndes CM, Shah KV. Burden and management of non-cancerous HPV-related conditions: HPV-6/11 disease. Vaccine 2006; 24:S35–S41.
6. Villa LL, Costa RL, Petta CA, et al. Prophylactic quadrivalent human papillomavirus (types 6, 11, 16, and 18) L1 virus-like particle vaccine in young women: A randomized double-blind placebo-controlled multi centre phase II efficacy trial. Lancet Oncol 2005; 6:271–278.
7. Villa LL, Costa RL, Petta CA, et al. High sustained efficacy of a prophylactic quadrivalent human papillomavirus types 6/11/16/18 L1 virus-like particle vaccine through 5 years of follow-up. Br J Cancer 2006; 95:1459–1466.
8. Markowitz LE, Dunne EF, Saraiya M, et al; Centers for Disease Control and Prevention. Recommendations and Reports. Quadrivalent Human Papillomavirus Vaccine: Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 2007; 56:1–24.
9. Botman S, Moore T, Parsons VL. Design and estimation for the National Health Interview Survey, 1995–2004. National Center for Health Statistics. Vital Health Stat 2000; 2:1–31.
10. Korn E, Graubard B. Analysis of Health Surveys. New York: Wiley, 1999.
11. Casella G, Berger R. Statistical Inference. Pacific Grove, CA: Wadsworth & Brooks/Col, 1990.
12. Insinga RP, Dasbach EJ, Myers ER. The health and economic burden of genital warts in a set of private health plans in the United States. Clin Infect Dis 2003; 36:1397–1403.
13. Koshiol JE, Laurent SA, Pimenta JM. Rate and predictors of new genital warts claims and genital warts-related healthcare utilization among privately insured patients in the United States. Sex Transm Dis 2004; 31:748–752.
14. Hariri S, Dunne EF, Sternberg MR, et al. Seroepidemiology of HPV-11 in the United States: Results from the Third National Health and Nutrition Examination Survey, 1991–1994. Sex Transm Dis 2008; 35:298–303.
15. Insinga RP, Glass AG, Rush BB. The health care costs of cervical human papillomavirus-related disease. Am J Obstet Gynecol 2004; 191:114–120.