To the Editor: The inverse relationship between the frequencyof condom use and the incidence of human papillomavirus (HPV)infection, reported by Winer et al. (June 22 issue),1 may bedue to variation in the prevalence of HPV infection among sexpartners with different levels of condom use. Men who use condomsmore frequently may be more aware of the possibility of sexuallytransmitted infections and may be less likely to have sexualcontact with high-risk groups, such as sex workers. The prevalenceof HPV infection among men with a high level of condom use maybe lower than among men with a low level of use. Therefore,such men would have a lower risk of transmitting HPV to femalesex partners, even in the absence of a protective effect ofcondom use.
The incidence of an infectious disease is determined by theprevalence of the disease in the population, the probabilityof transmission per sexual contact, and the number of sexualcontacts per unit of time.2 Therefore, in this case, an inverseassociation between the incidence of HPV infection and the levelof condom use would be observed if there was an inverse associationbetween the prevalence of HPV infection among sex partners andthe level of condom use, even if condom use did not affect theprobability of transmission.
Wenbin Liang, M.B., B.S. Curtin University of Technology Perth WA 6102, Australia wenbin.liang{at}postgrad.curtin.edu.au
References
Winer RL, Hughes JP, Feng Q, et al. Condom use and the risk of genital human papillomavirus infection in young women. N Engl J Med 2006;354:2645-2654. [Free Full Text]
Rothman KJ, Greenland S. Modern epidemiology. 2nd ed. Philadelphia: Lippincott-Raven, 1998.
To the Editor: Winer et al. and Steiner and Cates (in the accompanyingPerspective article)1 discuss the importance of data regardingthe use of condoms in the prevention of sexually transmitteddiseases. It must be emphasized that specific methodologic recommendationsexist for data collection. For example, reporting on the numberand type of sex acts provides more accurate information thansimply reporting on the number of sex partners. A recent review2suggests that although measures of condom use have modestlyimproved over time, many recommendations have not been implemented.Furthermore, because a gold standard does not exist, data collectionmust be tailored and comprehensive. This need is particularlyimportant with high-risk groups.
In a study funded by the National Institute on Drug Abuse,3445 women who were actively using drugs and not in treatmentreported engaging in an average of 54 sex acts during a 4-monthperiod. Of those encounters, the respondents reported the useof condoms in only 31% of instances of vaginal intercourse,12% of instances of fellatio, and 4% of instances of cunnilingus.Sex workers used protection more often than did other womenbut were more likely to engage in folk practices to preventsexually transmitted diseases, such as having sex only withhealthy-looking people. Researchers must integrate recommendationsregarding the collection of data on condom use, since accuratedata are critical for the development of tailored preventionmessages.
Linda Cottler, Ph.D. Eugenia C. Garvin, B.A. Catina Callahan, M.S.W. Washington University School of Medicine St. Louis, MO 63108 garvine{at}wustl.edu
References
Steiner MJ, Cates W Jr. Condoms and sexually-transmitted infections. N Engl J Med 2006;354:2642-2643. [Free Full Text]
Noar S, Cole C, Carlyle K. Condom use measurement in 56 studies of sexual risk behavior: review and recommendations. Arch Sex Behav 2006;35:327-345. [CrossRef][Web of Science][Medline]
Cottler L, Ben Abdallah A, Callahan C, Striley C. Sex (trading) in the city: practices and beliefs about sex among female crack users who trade sex. Presented at the 68th Annual Meeting of the College on Problems of Drug Dependence, Scottsdale, AZ, June 17–22, 2006 (poster).
The authors reply: Liang and Cottler et al. address some ofthe methodologic challenges of conducting studies of the effectivenessof condom use. We agree with Cottler and colleagues that specificmeasures should be taken to ensure that data collection is accurateand study-specific. In our study, for example, we were ableto generate more precise estimates of the frequency of condomuse by collecting data on daily sexual behavior every 2 weeks.Furthermore, although the number of instances of vaginal intercoursewas not an independent predictor of HPV acquisition in our study(probably because our subjects reported having fewer instancesof vaginal intercourse with new partners than with longer-termpartners), measurements that incorporate numbers rather thanproportions of protected acts, as compared with unprotectedacts, may be appropriate for studies in other populations orof other sexually transmitted diseases.
As Steiner and Cates note, the inherent subjectivity of self-reporteddata on sexual behavior is also a concern. Computer-assistedquestionnaires may be more effective than face-to-face interviewsat eliciting more truthful reporting of sensitive behavior,1and biologic markers may be used to confirm the validity ofself-reported data on sexual behavior.2 Steiner and Cates alsostress that a failure to account for the differential risk ofexposure to sexually transmitted diseases may lead to the underestimationof the effectiveness of condom use. Indeed, the women in ourstudy tended to report more frequent condom use with "riskier"partners; controlling for the number of new partners and thenumber of previous partners reported by male partners strengthenedthe observed inverse association between condom use and HPVacquisition. The latter variable was a useful surrogate indicatorof a male partner's infection status, since no infections weredetected in women reporting previously virginal male partners.Given these observations and the fact that HPV infections arenot concentrated in high-risk populations (in a study of low-riskmen attending the same university, for example, the prevalenceof HPV infection was 33%3), Liang's suggestion that differentialrisk exposure could have led to an overestimate of the effectivenessof condom use seems unlikely.
Rachel L. Winer, Ph.D. Laura A. Koutsky, Ph.D. University of Washington Seattle, WA 98103 rlw{at}u.washington.edu
References
Locke SE, Kowaloff HB, Hoff RG, et al. Computer-based interview for screening blood donors for risk of HIV transmission. JAMA 1992;268:1301-1305. [Free Full Text]
Gallo MF, Behets FM, Steiner MJ, et al. Prostate-specific antigen to ascertain reliability of self-reported coital exposure to semen. Sex Transm Dis 2006;33:476-479. [CrossRef][Web of Science][Medline]
Weaver BA, Feng Q, Holmes KK, et al. Evaluation of genital sites and sampling techniques for detection of human papillomavirus DNA in men. J Infect Dis 2004;189:677-685. [CrossRef][Web of Science][Medline]
Previous
