CONDYLOMATA ACUMINATA, or genital warts, are grossly visible florid lesions of the genital epithelium. In women, these exophytic lesions are most commonly found on the external genitalia and adjacent anal and perianal areas but may also occur on the cervix and vagina. It has been known for centuries that genital warts may be transmitted sexually,1 although it was not until the mid‐1980s that these lesions were found to result from infection with certain mucosotropic types of human papillomavirus (HPV), usually HPV‐6 or HPV‐11.2 Genital warts are the most common clinical manifestation of HPV infection of the genitalia. The incubation period has been estimated to range from 3 weeks to 8 months.3 Infection with these viruses may also be subclinical or latent and clinically undetectable.1
Condyloma is not a legally mandated reportable sexually transmitted disease (STD) in the United States; therefore, information on secular trends of this disease is limited. Data from a national survey of private practitioners in the United States indicated that the incidence of genital warts increased substantially during the 1970s and early 1980s,4 and condyloma is currently considered to be one of the most commonly diagnosed STDs in the United States.5
Several factors have been observed to be associated with cancers of the anogenital system, including sexual behavior, income, parity, oral contraceptive use, and smoking. A history of condyloma also has been found to be associated with certain anogenital cancers in women.6,7 The HPV types considered to be the putative etiologic agents in the development of these cancers are not those usually responsible for genital warts; however, condyloma and anogenital cancers may share similar risk factors that are related to HPV exposure, host response, or viral replication or persistence.
A number of studies examined factors related to genital HPV infection among women from selected populations, primarily from student health centers and STD and family planning clinics.8–16 Most studies of genital HPV infection focused on prevalent infection of the cervix as determined by various laboratory assays that detect the presence of viral DNA.8–18 Only a few studies examined factors related to the development of condylomata acuminata among women in the general population.19,20
Analysis of data from this population‐based case‐control study was performed specifically to determine the exposures that predispose women to the development of genital warts. We placed special emphasis on exposures identified as potential cofactors in the development of anogenital cancers in women.
Study subjects for this population‐based case‐control study were western Washington enrollees of the Group Health Cooperative of Puget Sound (GHC) who received their primary health care from four GHC clinics serving approximately 280,000 residents of western Washington state. Group Health Cooperative is a health maintenance organization whose population resembles that of the surrounding community with regard to age, gender, and race.21 Although the membership includes Medicaid patients, it underrepresents the very poor. This analysis includes female subjects only. An analysis of the male subjects in this study is reported by Van Den Eeden et al.22
Potential patients included all women 18 years of age or older who were diagnosed with genital warts at one of these four primary health care clinics between April 1, 1987 and September 30, 1991. Patients were identified by reviewing all treatment record forms for patients seen by primary care practitioners, women's health care specialists, or dermatologists during this period. Each potential patient was assigned a reference date corresponding to the date that they were diagnosed with condyloma.
After first obtaining consent from the patient's primary physician, potential patients were sent a letter by the GHC research department explaining the study and asking them to send back a card indicating their interest in the study. If the card was not returned within 2 weeks, patients were called by a GHC research staff person. A female study interviewer then telephoned interested patients to schedule an interview. Of the 258 potential patients identified, 149 (58.1%) were interviewed. Reasons for nonresponse included doctor or patient refusal (32.6%) and loss to follow‐up (9.3%).
Potential control subjects were selected from computergenerated random lists of GHC enrollees. One potential control subject was matched to each potential case by clinic, age (5‐year groups), and gender. The control subject was assigned the same reference date as the patient. The contact procedures for control subjects were similar to those for the patients. Of the 230 potential control subjects, 144 (63.5%) were interviewed; (29.6%) were nonresponders because of doctor or subject refusal and (7.0%) could not be located.
Subjects were interviewed in their homes or other place of convenience. A structured questionnaire was used to collect information on demographic and lifestyle characteristics (e.g., diet, alcohol, smoking, recreational drug use) and on medical, reproductive, contraceptive, and sexual histories. A number of questions about infections were asked, including if the subjects ever had genital (or venereal) warts. Women answering yes to this question were asked how many times they had genital warts, if they were ever diagnosed by a physician, and the site(s) where the warts appeared. All interview questions were restricted to exposures and characteristics before diagnosis date (patients) or reference date (control subjects). There was a median of 8 months between diagnosis or reference date and interview. A sample of blood was drawn at the time of interview or occasionally at a later date by an interviewer or by the subject's own physician. Subjects who had blood drawn in this study were included in an analysis of serologic markers of infection to HPV type 6 or 11, reported previously by Carter et al.23
Logistic regression was used to estimate relative risks (RR; via the odds ratio) while controlling for the confounding effect of other factors.24 Ninety‐five percent confidence intervals (CIs) were calculated using the normal approximation and standard errors. Using the same control group for comparison, RRs were estimated separately for women with incident and recurrent warts. Recurrent patients were also compared directly with incident patients to determine if the two patient groups were statistically significantly different with respect to patient characteristics and exposures. We defined incident patients as those women with a first reported diagnosis of condyloma during the study period; women who reported having at least one prior episode of condyloma were considered to be recurrent patients. Control subjects (n = 11) with a history of prior genital warts were excluded from all analyses.
Selected characteristics of control subjects and incident and recurrent condyloma patients are presented in Table 1. The study population was comprised largely of young adult women; over 80% of subjects were under age 40. Approximately 37% of patients had a prior episode of genital warts. Compared with control subjects, incident patients were more likely to be less than 20 years of age, black, and to report an annual income of less than $15,000; recurrent patients were more likely to be 30 years of age or older, to have at least some college, and to have annual incomes less than $30,000.
Table 2 presents RRs for condyloma associated with several demographic variables. Women with no sexual partners during the 5 years before the reference date (nine control subjects, one incident and one recurrent patient) were excluded from the results presented in this and subsequent tables. Compared with currently married women, women who were never married, formerly married, or living with a partner were at increased risk of incident condyloma. Incident patients were also more likely than recurrent patients to be unmarried at the reference date.
Risk of incident and recurrent condyloma increased with increasing number of sexual partners in the 5 years before the reference date (Table 3). Although the numbers were small, there was some suggestion that women who were 20 years or older at first intercourse were at modestly increased risk of incident disease and decreased risk of recurrent disease.
Women diagnosed with incident condyloma were more than twice as likely to also have a history of another STD than were control women (Table 4). The associations with individual STDs (genital herpes, gonorrhea, syphilis, and Trichomonas) were somewhat elevated but were not statistically significant (some data not shown). A history of other STDs was not associated with increased risk of recurrent condyloma. No control subjects reported a sexual partner during the 5 years before the reference date who had genital warts; over 40% of incident and recurrent patients reported a sexual partner during this period who had genital warts.
A history of oral herpes was associated with an increased risk of incident, but not recurrent, condyloma (Table 5). There was a suggestion that women with a history of yeast infections were at slightly greater risk of condyloma than women without such a history. Women who reported common warts on two or more locations on the body were at substantially increased risk of recurrent condyloma, and there was some suggestion that they were also at increased risk for incident disease. However, there was no increased risk of condyloma observed among women with common warts on only one location on the body.
No clear pattern emerged for the association of cigarette smoking and risk of incident or recurrent condyloma (Table 6). Our data did not support an association between either cocaine or marijuana use and risk of condyloma.
Table 7 presents the RRs for condyloma associated with factors that may be indicators of immune status. Women who reported a history of allergies were at twice the risk of incident condyloma compared to women who did not report having allergies. Compared with control subjects, incident patients more frequently reported use of steroids during the past year, a history of diabetes, and a history of cancer; however, the number of individuals with these exposures was quite small and the RRs were extremely imprecise.
At reference date, 56% of control subjects, 74% of incident patients, and 61% of recurrent patients were nulliparous and 41%, 55%, and 31% were nulligravid, respectively. Although the numbers were small, there was a suggestion that women pregnant within a year of reference date were less likely to have incident or recurrent condyloma. This pattern was also observed when the analysis was restricted to gravid women. A total of eight control subjects, one incident patient, and two recurrent patients were pregnant at the reference date.
Approximately 80% of the study population reported using oral contraceptives (OCs) for some period before the reference date. Among women reporting some prior use of OCs, there was a suggestion of a slight decrease in risk of incident condyloma with increasing number of years of OC use (P for trend = 0.13); recurrent disease did not appear to be associated with duration of use. Women who only used barrier methods (condoms, diaphragms, cervical caps) during the year before the reference date were more likely to be diagnosed with condyloma than women who reported using OCs only during this period, but our estimate was quite imprecise (RR, 2.3; 95% CI, 0.4‐14.5).
Several factors should be considered when interpreting the results of this study. We were only able to interview 58% of patients and 64% of control subjects. The sensitive nature of this disease probably contributed to this high rate of nonresponse. If control subjects exposed to certain factors were more or less likely to participate than exposed patients, our estimates for these exposures would be biased. In addition, some misclassification of disease status may have occurred. The diagnosis of condyloma was made on the basis of clinical appearance, because only a small number of condylomas were excised as part of routine treatment and were available for HPV typing. Of the 52 tissues tested for HPV DNA (see methods in Beckman et al25), approximately 83% were positive for HPV types 6 and/or 11 and 6% were positive for type 16. A similar proportion of genital wart biopsies from women were found by Hillman et al26 to be positive for HPV DNA.
Infection with HPV is not always symptomatic, and some symptomatic woman may not seek medical attention. A number of our control subjects may have had subclinical disease, biasing our estimates toward the null. Some patients we defined as incident may have had a lesion for a period of time before seeking care. No information was collected on the reasons for the doctor visits when the diagnoses of condyloma were made. If patients with multiple sexual partners, partners with condyloma, or other exposures were more likely to see a health care provider for a suspicious genital lesion, the estimates for these exposures would be artificially high. Recurrent disease in our study was likely the result of both reactivation of latent virus and new infection.27,28 Because risk factors for new infection and reactivation may differ, this mixture of disease included in the recurrent patient group could attenuate the relative risk estimates. Further, some patients' responses to sexual behavior questions may have been influenced by the knowledge that genital warts are caused by sexually transmitted agents. If patients recalled more sexual activity than control subjects, RR estimates for these factors would be biased upward. Among women participating in a prospective study of genital HPV infection in Toronto, Canada, Rohan et al29 found a very high level of repeatability (intraclass correlation coefficient of 0.98 for number of sexual partners) of interview‐derived sexual history information. However, there was some tendency for women who were informed that they were HPV positive after the initial interview to report more sexual partners on reinterview.
We observed a strong relationship between number of sexual partners and risk of incident and recurrent condyloma. The relation between self‐reported sexual behavior and genital HPV infection has been examined in several studies of women. In a case‐control study conducted by Daling et al20, women with the highest number of sexual partners were the most likely to have a history of genital warts. Studies examining the relation between the number of sexual partners and cervical infection with HPV have usually, but not always, found a positive association.3 As with our findings, most studies have not observed a strong association between age at first intercourse and risk of genital infection with HPV.3
Although the association we observed between risk of condyloma and having a sexual partner with genital warts may have been exaggerated somewhat by recall bias, it confirms other reports of the sexual transmission of this disease.1,30 Positive associations between condyloma and various other STDs have been reported by others.20 Sexually transmitted diseases are often correlated, either because of shared sexual behavioral risk factors or because some sexually transmitted infections disrupt the genital mucosa, making it less resistant to infection with other agents.3
Cigarette smoking has been found to be associated with an increase in risk of anogenital cancers.31 This may be due in part to residual confounding by sexual behavior. However, biologic mechanisms have also been proposed. For example, smoking may suppress the immune system, increasing an individual's susceptibility to HPV infection or extending viral persistence.32,33 We only observed a very modest, and statistically insignificant, increase in risk of incident condyloma among women who were current or former smokers. Risk did not seem to vary by the number of cigarettes a day a woman reported smoking, although we observed an increase in risk of recurrent disease associated with smoking for 10 or more years. Two studies reported a positive association between smoking and risk of genital warts.20,34 However, most,11,13,14,16,17,35 but not all,12 studies of HPV infection of the cervix have not observed an association with cigarette smoking.
In a number of studies, high‐risk sexual behavior has been found to be more common when individuals use alcohol and/or recreational drugs.30,36 However, few studies reported on the association between HPV infection and these exposures. The results of an examination of the association between diet, including alcohol consumption, and risk of condyloma in this study population were reported previously.37 In that analysis, risk of condyloma was found to increase with increasing consumption of alcohol. We did not observe an increase in risk of either incident or recurrent condyloma associated with recreational drug use. In a study of sexually active adolescent females attending a family planning clinic, the positive association between marijuana or other drug use and HPV cervical infection became statistically insignificant after adjusting for number of sexual partners.10
Clinically evident HPV infection and viral positivity have been reported to be elevated during pregnancy, although some studies found no association.3 Our study included very few women who were pregnant at the reference date. There was a suggestion in our data that women with a recent pregnancy were less likely to have incident or recurrent warts compared with women whose last pregnancy was one or more years before the reference date. This reduction in risk may be partially explained by differences in the sexual behaviors of women who are planning to get pregnant and those who are not.
The relation between contraceptive use and genital HPV infection is also extremely difficult to assess because of probable selection bias or residual confounding from the association between contraceptive practices and both sexual behavior and screening for cervical neoplasia.38 Nevertheless, several studies examined this relationship. Daling et al20 found that, compared with never users of OCs, women who reported using OCs for 5 or more years were almost 10 times as likely to have genital warts. A few studies reported that past or current users of OCs were more likely to have prevalent HPV infection of the cervix than women who had never used OCs.11,39 Our findings are consistent with other studies that found only a weak or no association between HPV infection and OC use.10,12,13,16,18,35
It is not known whether barrier methods protect against genital infection with HPV. Barrier methods do not cover all potentially infective or susceptible genital surfaces. In addition, they may not be used consistently.40 Villa and Franco13 observed that women with HPV infection of the cervix were 50% more likely to have used condoms at some time in the past. Neither Moscicki et al10 nor Kjaer et al17 observed an association between barrier contraceptive use and HPV infection of the cervix. Compared with women using only OCs, there was a suggestion that women who used only barriers (condoms, diaphragm, cervical cap) within the year before the reference date were at slightly increased risk of incident condyloma. Some women in our study may have used barrier methods because they had a genital lesion that appeared several months before the actual diagnosis of condyloma by a GHC health care provider or because their partner was known to have had warts.
The role of immune function in the control or expression of HPV infection has not been fully elucidated. Even among apparently immunologically intact individuals, the immune response appears inadequate, because HPV‐associated lesions often persist for months or years.41 The increase in prevalence or size of genital warts among renal allograft recipients, patients with Hodgkin's disease or acquired immunodeficiency syndrome, and patients receiving immunosuppressive chemotherapy suggests that cell‐mediated immunity may be especially important in the control of HPV infections.2,41 We only had a few rather poor potential indicators of immune function, and it is possible that the associations we observed were due to confounding by the correlation of these conditions with increased disease surveillance. Nevertheless, we found that women with recurrent warts were significantly more likely than control subjects to report having common warts on two or more locations on the body. As in our data, a population‐based study conducted in Rochester, Minnesota observed that diabetes mellitus was more frequent among patients with condyloma.19
In summary, our data suggest that risk of condyloma is primarily related to sexual behavior. We did not observe a strong association between risk of condyloma and many of the exposures considered to be potential cofactors for anogenital cancer, although our ability to examine some of these factors was limited.
1. Jenson AB, Lancaster WD. Association of human papillomavirus with benign, premalignant, and malignant anogenital lesions. In: Pfister H, ed. Papillomaviruses and Human Cancer. Boca Raton, FL: CRC Press, 1992:12-43.
2. Shah KV, Howley PM. Papillomaviruses. In: Fields BN, Knipe DM, eds. Virology. 2nd ed. New York: Raven Press, 1990:1651-1676.
3. Schneider A, Koutsky LA. Natural history and epidemiological features of genital HPV infection. In: Munoz N, Bosch FX, Shah KV, Meheus A, eds. The Epidemiology of Cervical Cancer and Human Papilloma Virus. Lyon: International Agency for Research on Cancer, 1992.
4. Becker TM, Stone KM, Alexander ER. Genital human papillomavirus infection: a growing concern. Obstet Gynecol Clin North Am 1987;14:389-396.
5. Campion MJ. Epidemiology of genital human papillomavirus infection. Curr Probl Obstet Gynecol Fertil 1989; 12:109-120.
6. Brinton LA. Epidemiology of cervical cancer-overview. In: Munoz N, Bosch FX, Shah KV, Meheus A, eds. The Epidemiology of Cervical Cancer and Human Papilloma Virus. Lyon: International Agency for Research on Cancer, 1992:3-23.
7. Daling JR, Sherman KJ. Relationship between human papillomavirus infection and tumours of anogenital sites other than cervix. In: Munoz N, Bosch FX, Shah KV, Meheus A, eds. The Epidemiology of Cervical Cancer and Human Papilloma Virus. Lyon: International Agency for Research on Cancer, 1992.
8. Kiviat KB, Koutsky LA, Paavonen JA, et al. Prevalence of genital papillomavirus infection among women attending a college student health clinic or a sexually transmitted disease clinic. J Infect Dis 1989; 159:293-302.
9. Rosenfeld WD, Vermund SH, Wentz SJ, Burk RD. High prevalence rate of human papillomavirus infection and association with abnormal Papanicolaou smears in sexually active adolescents. Am J Dis Child 1989; 143:1443-1447.
10. Moscicki AB, Palefsky J, Gonzales J, Schoolnik GK. Human papillomavirus infection in sexually active adolescent females: prevalence and risk factors. Pediatr Res 1990; 28:507-513.
11. Ley C, Bauer HM, Reingold A, Chambers JC, Tashiro CJ, Manos MM. Determinants of genital papillomavirus infection in young women. J Natl Cancer Inst 1991; 83:997-1003.
12. Rohan T, Mann F, McLaughlin J, et al. PCR-detected genital papillomavirus infection: prevalence and association with risk factors for cervical cancer. Int J Cancer 1991; 49:856-860.
13. Villa LL, Franco EL. Epidemiologic correlates of cervical neoplasia and risk of human papillomavirus infection in asymptomatic women in Brazil. J Natl Cancer Inst 1989; 81:332-340.
14. Hildesheim A, Gravitt P, Schiffman MH, et al. Determinants of genital human papillomavirus infection in low-income women in Washington, D.C. Sex Transm Dis 1993; 20:279-285.
15. Reed BD, Zazove P, Gregoire L, Gorenflo DW, Lancaster WD, Ruffin MT. Factors associated with human papillomavirus infection in women encountered in community-based offices. Arch Fam Med 1993; 2:1239-1248.
16. Wheeler CM, Parmenter CA, Hunt WC, et al. Determinants of genital human papillomavirus infection among cytologically normal women attending the University of New Mexico student health center. Sex Transm Dis 1993; 20:286-289.
17. Kjaer SK, Engholm G, Teisen C, et al. Risk factors for cervical human papillomavirus and herpes simplex virus infections in Greenland and Denmark: a population-based study. Am J Epidemiol 1990; 131:669-682.
18. Bauer HM, Hildesheim A, Schiffman MH, et al. Determinants of genital human papillomavirus infection in low-risk women in Portland, Oregon. Sex Transm Dis 1993; 20:274-278.
19. Chuang T-Y, Perry HO, Kurland LT, Ilstrup DM. Condyloma acuminatum in Rochester, Minn., 1950-1978. Epidemiology and clinical features. Arch Dermatol 1984; 120:469-475.
20. Daling JR, Sherman KJ, Weiss JS. Risk factors for condyloma acuminatum in women. Sex Transm Dis 1986; 13:16-18.
21. Pearson DC, Grothaus L, Thompson RS, Wagner EH. Smokers and drinkers in a health maintenance organization population: lifestyles and health status. Prev Med 1987; 16:783-95.
22. Van Den Eeden SK, Habel LA, Sherman KL, McKnight B, Stergachis A, Daling JR. Risk factors for incident and recurrent condylomata acuminata among men: A population-based study. Sex Transm Dis 1998: 25:278-284.
23. Carter JJ, Wipf GC, Hagensee ME, et al. Use of human papillomavirus type 6 capsids to detect antibodies in people with genital warts. J Infect Dis 1995; 172:11-18.
24. Breslow NE, Day NE, eds. Statistical Methods in Cancer Research. Vol I. The Analysis of Case-Control Studies. Lyon: International Agency for Research on Cancer, 1980. (IARC scientific publication no. 32.)
25. Beckmann AM, Sherman KJ, Myerson D, Daling JR, McDougall JK, Galloway DA. Comparative virologic studies of condylomata acuminata reveal a lack of dual infections with human papillomaviruses. J Infect Dis 1991; 163:393-396.
26. Hillman RJ, Ryait BK, Botcherby M, Taylor-Robinson D. Changes in HPV infection in patients with anogenital warts and their partners. Genitourin Med 1993; 69:450-456.
27. Ferenczy A, Mitao M, Nagai N, Silverstein SJ, Crum CP. Latent papillomavirus and recurring genital warts. N Engl J Med 1985; 313:784-788.
28. Nuovo GJ, Pedermonte BM. Human papillomavirus types and recurrent cervical warts. JAMA 1990; 263:1223-1226.
29. Rohan TE, McLaughlin JR, Harnish DG. Repeatability of interview-derived information on sexual history: a study of women. Epidemiology 1994; 5:360-363.
30. Oriel JD. Genital warts. Sex Transm Dis 1981; 8:326-329.
31. Daling JR, Shermon KJ, Hislop TG, et al. Cigarette smoking and the risk of anogenital cancer. Am J Epidemiol 1992; 135:180-189.
32. Barton SE, Jenkins D, Cuzick J, Maddox PH, Edwards R, Singer A. Effect of cigarette smoking on cervical epithelial immunity: a mechanism for neoplastic change? Lancet 1988; 2:652-654.
33. Ferson M, Edwards A, Lind A, Milton GW, Hersey P. Low natural killer-cell activity and immunoglobulin levels associated with smoking in human subjects. Int J Cancer 1979; 23:603-609.
34. Brisson J, Roy M, Fortier M, Fortier M, Bouchard C, Meisels A. Condyloma and intraepithelial neoplasia of the uterine cervix: a case-control study. Am J Epidemiol 1988; 128:337-342.
35. Reeves WC, Brinton LA, Garcia M, et al. Human papillomavirus infection and cervical cancer in Latin America. N Engl J Med 1989; 320:1437-1441.
36. Lauchili S, Heusser R, Tschopp A, Gutzwiller F, The Research Group of the Swiss HIV Prevention Study. Safer sex behaviour and alcohol consumption. Ann Epidemiol 1996; 6:357-364.
37. Bairati I, Sherman KJ, McKnight B, et al. Diet and genital warts: a case-control study. Sex Transm Dis 1994; 21:149-153.
38. Swan SH, Petitti DP. A review of problems of bias and confounding in epidemiological studies of cervical neoplasia and oral contraceptive use. Am J Epidemiol 1982; 115:10-18.
39. Lorincz AT, Schiffman MH, Jaffurs WJ, et al. Temporal associations of human papillomavirus infection with cervical cytologic abnormalities. Am J Obstet Gynecol 1990; 162:645-651.
40. Park BJ, Stergachis A, Scholes D, et al. Contraceptive methods and the risk of Chlamydia trachomatis
infection in young women. Am J Epidemiol 1995; 142:771-778.
41. Tindle RW, Frazer IH. Immunology of anogenital human papillomavirus (HPV) infection. Aust NA J Obstet Gynaecol 1990; 30:370-375.