Copyright © 2004 by the Johns Hopkins Bloomberg School of Public Health
LETTERS TO THE EDITOR |
SEVEN AUTHORS REPLY
1 Division of Reproductive Health, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, GA 30333
2 Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA 30322
3 Division of HIV/AIDS Prevention, National Center for HIV, STD, and TB Prevention, Centers for Disease Control and Prevention, Atlanta, GA 30333
4 Division of Sexually Transmitted Diseases, National Center for HIV, STD, and TB Prevention, Centers for Disease Control and Prevention, Atlanta, GA 30333
5 Baltimore City Health Department, Baltimore, MD 21202
6 Division of Infectious Diseases, Johns Hopkins University School of Medicine, Baltimore, MD 21205
We are grateful to Mann and Stine (1) for their comments regarding our analysis (2) of the importance of measuring partner infection status when assessing condom effectiveness against gonorrhea and chlamydia.
Our analysis evaluated whether we could reduce confounding in estimates of condom effectiveness caused by people’s tendency to use condoms more frequently with risky partners. To decrease confounding, we restricted our analyses to participants who were known to have infected partners because they presented to a sexually transmitted disease clinic after sexual contact with an infected partner. By addressing this one source of bias, we observed a markedly stronger protective effect for condoms, as well as a stronger association between the number of unprotected sexual exposures and infection (2, p. 247 (tables 3 and 4)).
Mann and Stine (1) express concern about the lack of association between the number of condom-protected exposures and risk of infection. They posit that infection risk among consistent condom users should increase with an increasing number of exposures to infected partners because condoms occasionally fail. In their proposed explanation, they point out (as we did in our article) that our measure did not distinguish precisely between protected and unprotected exposures to infectious partners during the observation period. This lack of precision, however, is typical of observational studies of condom use and curable sexually transmitted disease, and our approach represents an improvement over previous efforts to approximate the degree of exposure to infected partners.
There are two points that we would like to clarify. First, Mann and Stine’s premise may have arisen from a misinterpretation of table 4 from our article (2) regarding the relation between the number of condom-protected exposures and infection among "condom-using" subjects. Our analysis appropriately included as "condom-using" subjects both consistent and inconsistent condom users. We could not determine whether inconsistent users acquired infections during protected sex or unprotected sex, and it is probable that some infections were acquired during sex acts when condoms were not used. Such infections could obscure any relation between condom-protected exposures and risk of infection. When we limited our analyses to the 33 participants who reported only consistent condom use, we still found no association between the number of condom-protected exposures and infection (
2 = 0.08; p = 0.77).
Second, the fact that the adjusted odds ratio was greater than zero does not necessarily indicate that infections among consistent users were due to condom failure. Although condom failure is certainly possible, other factors that we noted (such as condom-use reporting error, infection prior to condom use, or laboratory error) might also account for some infections among consistent users. These factors could explain why infection risk did not increase with the number of condom-protected exposures among consistent users.
In our study, condom use clearly reduced the risk of infection for the group of patients who were known to have had infected partners. Although we could not measure the protective effect of condom use with precision, our results indicate that it is larger than observed in studies that are unable to control for partner infection status. Our approach, like that of Mann et al. (3), argues that condom effectiveness is best estimated by studying persons who have been exposed to infected partners. Determining the exact number of exposures to an infected partner during both protected and unprotected sex is clearly critical to this estimation. Although our measure of partner infection status represents a significant advance over previous efforts, we were still limited by an inability to measure other factors. Given that these factors likely introduce bias toward the null (4, 5), what remains a matter of conjecture is how much stronger our estimate of condom effectiveness might have been had we been able to address these other sources of bias.
REFERENCES
REFERENCES
- Mann JR, Stine CC. Re: "Condom effectiveness for reducing transmission of gonorrhea and chlamydia: the importance of assessing partner infection status." (Letter). Am J Epidemiol 2004;160:608.
[Free Full Text] - Warner L, Newman DR, Austin HD, et al. Condom effectiveness for reducing transmission of gonorrhea and chlamydia: the importance of assessing partner infection status. Am J Epidemiol 2004;159:242–51.
[Abstract/Free Full Text] - Mann JR, Stine CC, Vessey J. The role of disease-specific infectivity and number of disease exposures on the long-term effectiveness of the latex condom. Sex Transm Dis 2002;29:344–9.[Web of Science][Medline]
- Devine O, Aral S. The impact of inaccurate reporting of condom use and imperfect diagnosis of STD infection in studies of condom effectiveness: a simulation-based assessment. Sex Transm Dis (in press).
- Shlay JC, McClung MW, Patnaik JL, et al. Comparison of sexually transmitted disease prevalence by reported level of condom use among patients attending an urban sexually transmitted disease clinic. Sex Transm Dis 2004;31:154–60.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Warner, D. R. Newman, M. L. Kamb, M. Fishbein, J. M. Douglas Jr, J. Zenilman, L. D'Anna, G. Bolan, J. Rogers, T. Peterman, et al. Problems with Condom Use among Patients Attending Sexually Transmitted Disease Clinics: Prevalence, Predictors, and Relation to Incident Gonorrhea and Chlamydia Am. J. Epidemiol., February 1, 2008; 167(3): 341 - 349. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Warner, D. R. Newman, J. M. Douglas Jr., J. M. Zenilman, D. K. Kleinbaum, M. Macaluso, and T. A. Peterman SEVEN AUTHORS REPLY Am. J. Epidemiol., September 15, 2004; 160(6): 608 - 609. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||