Background on the Weller Study

Date: January 1, 1997
Source: Centers for Disease Control and Prevention (CDC)

The Food and Drug Administration (FDA) and the Centers for Disease Control and Prevention (CDC) continue to receive inquiries about an article by Susan Weller called "A Meta-Analysis of Condom Effectiveness in Reducing Sexually Transmitted HIV" (published in Social Science Medicine, June 1993; vol. 36, no. 12, pages 1635-44). This review of previously published reports of condom use among "discordant couples' (couples in which one partner is HIV-infected and the other isn't) has been incorrectly interpreted to indicate that condoms are not effective protection against sexual transmission of HIV, the virus that causes AIDS. The following information can be used to answer questions.

FDA and CDC believe this analysis was flawed. Weller reviewed a number of published studies of condom use among heterosexual couples described as discordant. However, because participants' HIV status was not always assessed before they took part in the studies, it is possible, even likely, that some were infected before starting to use condoms. These participants would then be incorrectly counted as "condom failures."

Taken alone, each study was too small to draw reliable conclusions about condom effectiveness. Using a technique called meta-analysis (combining multiple studies and analyzing the cumulative results), Weller attempted to derive a summary estimate of condom effectiveness for HIV protection. Her summary figure is a "pooled risk ratio" of 0.31, suggesting that condom use is indeed associated with a significantly reduced prevalence of HIV infection among heterosexual partners of HIV-infected individuals. This is a reassuring finding -- and it's important to note that it probably represents the worst case scenario, since studies and the meta-analysis itself lumped together consistent users with inconsistent users and nonusers.

From that pooled risk ratio, Weller goes on to calculate that condom effectiveness against HIV infection is only 69%. There are several problems with this calculation. First, pooled risk is in no way equivalent to or applicable to an estimate of condom effectiveness, for several reasons. Second, the studies Weller combined varied in their ability to verify the couples' HIV status before the start of condom use -- that is, some people were possibly infected before beginning to use condoms. Third, the studies often were unable to rule out other HIV risk behaviors, such as intravenous drug use, that could result in infection. Fourth, and perhaps most problematic, most of the studies Weller reviewed divided condom users into arbitrary categories such as "always/sometimes," "never," and "regular," offering no possibility of actually verifying true frequency of use or possible exposure to HIV. Even people placed into the "user" category could have had unprotected intercourse. By combining "always" users with "sometimes" users to come up with a total effectiveness rate, Weller makes a serious error.

Not making a distinction among consistent and inconsistent users makes assessing condoms' true effectiveness from this type of research impossible. It's similar to using statistics about the number of auto-related injuries to people who aren't wearing seatbelts to "prove" seatbelts don't work. Weller's review failed to measure condom effectiveness. Instead, the results reflected "user effectiveness" -- the failure to use condoms consistently and correctly -- and point to the need for continued education and support for condom use. This is something most people know from common sense: Condoms don't work if they aren't taken out of their packages and used.

HIV infection is a complex, multifaceted problem that cannot be solved by a single preventive measure. Weller's review does suggest that condom use -- even inconsistent condom use -- is associated with reduced risk of HIV infection. Other studies of discordant couples -- more recent and larger than the ones Weller reviewed, and conducted over several years -- have demonstrated that consistent condom use is highly effective at preventing HIV infection (DeVincenzi, I et al. A Longitudinal Study of Human Immunodeficiency Virus Transmission by Heterosexual Partners. New Engl J of Medicine 1994; vol. 331, no. 6: pages 341-346 and Saracco, A et al. Man to Woman Sexual Transmission of HIV: Longitudinal Study of 343 Steady Partners of Infected Men. J of Acquired Immune Deficiency Syndromes 1993; vol. 6: pages 497-502.)

Used consistently and correctly, latex condoms offer an extremely high degree of protection against HIV, other sexually transmitted diseases (STDs), and unintended pregnancy. Weller's review and other studies point to the pressing need for continued public education about the importance of using condoms consistently and correctly. For more information on condom use, call the CDC National AIDS Hotline at 1-800-342-AIDS or the CDC National STD Hotline at 1-800-227-8922.