The female condom: is "just as good" good enough?

Abstract

EVIDENCE SUPPORTING THE effectiveness of the female condom has accumulated steadily in the past decade. Is it finally sufficient to convince the majority of public health policy makers to invest in the method? The answer appears to be “not yet.” Public sector and donor support for female condom distribution is still weak in most countries facing the full enormity of the HIV/AIDS epidemic. What will it take to convince more decision makers of the value of including the female condom among the relatively few options currently available for the primary prevention of HIV and other STIs? The results reported by French and colleagues in this issue 1 may move reluctant decision makers in the right direction. The study edges towards a head-to-head comparison of the effectiveness of male and female condoms in preventing new infections. Study participants were randomly assigned to one of two study arms: one receiving training, support, and supplies for male condom use; the other similarly equipped for female condom use. Unavoidably, the latter group had access to male condoms outside the clinic, as they are widely available in the urban setting in which the study was conducted. As male condoms are the standard of STI prevention for sexually active persons, it is entirely appropriate that the investigators did not discourage participants in the female condom arm from using male condoms during the study. Since women in the female condom group used male condoms to an unknown degree, the results cannot be construed as a direct comparison of the efficacy of the two devices. Rather, it is a comparison of the effectiveness of two prevention programs, one with and one without female condoms. Aside from the inadvertent and unmeasured “contamination” of the female condom study arm, the French et al. 1 study carries important strengths that make its positive findings about the female condom convincing. Incorporating random assignment of study participants into ongoing, high-volume clinical work, the investigators successfully achieved two study arms with similar features. Citywide record linkage throughout the public health infrastructure enhanced the investigators’ capacity to ascertain new infections, using fairly accurate test methods for three of the four studied STIs. Data analysis completed in two different ways—either including or excluding those lost to follow-up—produced similar results: women in the female condom arm had a slightly reduced risk of new STI compared with women in the male condom arm. The authors conclude that women counseled on and provided female condoms are at least as protected as women provided male condoms alone. Perhaps more notable than the trial methods and results is the authors’ objection to placement of the female condom in a position inferior to the male condom. This is implied, they assert, by label instructions that encourage use of the female condom only in situations in which a male condom cannot be used. They contend that the female condom should no longer be offered as a mere “second best” option to the male condom, given the parity in protection offered by the two methods. Equivalence in protection is supported by unpublished data from a weaker cohort study conducted by Macaluso and colleagues in Alabama, 2 in which all trial participants had access to both female and male condoms. As in the French et al. cohort, 1 STI incidence was lower among women reporting female condom use. In fact, the most common way to achieve consistent use of protection was through mixed use of the two devices for different sex acts. The study had an observational design, however, and could not exclude self-selection bias stemming from inherent differences between successful female condom users and other women. 2 Decades of family planning program experience have shown that contraceptive prevalence increases in a population as a broader variety of family planning methods is offered. 3-6 Thus we should expect increased use of prophylactics with expanded choice, as couples have access to more methods that may meet their needs and preferences. Qualitative studies have shown that women are able to use the female condom in situations in which they cannot negotiate male condom use. 7-10 Most investigations that have compared condom use among women who have access to both the male and female condom versus those with access to the male condom alone, however, have found only slightly higher levels of protected coitus when both methods are available. 11-16 A recent exception was a study conducted among family planning clients in California that showed a significantly higher level of sex acts protected in the three months after the female condom was added to the method mix (from 44% to 59%), 17 although this was a simple before-after comparison. Why have so few studies even attempted to document increased protection levels associated with the addition of the female condom to a male condom distribution system? Experience with male condom studies has shown that tracking changes in protection levels resulting from promotion interventions is challenging. The best support comes from studies conducted with targeted high-risk groups. 18-26 These have shown repeatedly that intensive, individualized educational and skill-building interventions can drive up male condom use consistency. Yet in the past 10 years, most condom promotion in developing countries hardest hit by the HIV/AIDS epidemic has occurred through large-scale distribution systems, so vast that individualized counseling is not feasible. Social marketing programs have been shown to dramatically increase the number of condoms distributed in developing countries;27 generally left unexamined is the extent to which a net increase in condom use occurs in the general population, as opposed to a shift in the supply source. Assessments that have attempted to measure changes in preventive behaviors in response to large-scale male condom distribution programs have shown varying levels of success. 28-36 While documenting increases in condom use levels is difficult, demonstrating an inverse relationship between condoms distributed and infections transmitted is even more challenging. A recently conducted review of HIV prevention studies involving male condom promotion revealed that only eight interventions were designed to measure HIV/STI outcomes, of which five were successful. 37 An expert panel convened in June 2000 by the National Institutes of Health concluded that male condoms are effective against HIV and gonorrhea, but that gaps remain in the literature on their effectiveness in preventing other STIs. 38 The same methodological challenges that constrain studies on male condom effectiveness — the common inaccuracy of self-reported behaviors, the complex association between behavioral and biologic outcomes, and the variability in effect measures depending on the prevalence and infectiousness of the disease being examined 39,40 — are likely to hinder female condom studies as well. French and colleagues are therefore to be congratulated for contributing to the body of evidence supporting the female condom’s effectiveness in preventing STI transmission. Aside from French et al. 1 and Macaluso et al., 2 only three published studies have measured female condom use and STIs. 11,16,41 Two were randomized comparisons of high-risk women receiving male condoms, with half of the participants also having access to female condoms. Female sex workers in Thailand had lower STI incidence with access to the female condom, 11 whereas reproductive age female plantation workers in western and central Kenya enjoyed no measured benefit from female condoms. 16 The third study, based on a small, self-selected cohort, showed lower trichomoniasis incidence with consistent female condom use. 41 The finding that the female condom is as effective as the male condom may not be sufficiently compelling to policy makers asked to pay the female condom’s higher price, approximately 21 times that of the male condom. When an individual or couple decides which condom to purchase, the population-based benefits gained from including the female condom in the method mix are not relevant: individuals are free to use their money to purchase the method of their choice. However, when program managers are purchasing the female condom with public monies or donor funds that might otherwise be spent on male condoms, then interest in the added benefits to be gained from increased expenditure is entirely valid. From a budget perspective, the worst-case scenario for female condom distribution would be for total expenditures to increase without an attendant increase in levels of protection. This would occur if the more expensive female condom serves as a pure substitute for a male condom. While this extreme scenario is unlikely, what remains unknown is the magnitude of the expected increase in levels of protection when female condoms are added to existing male condom distribution. Cost-effectiveness analysis can be applied to assist decision makers in determining whether the additional investment of resources is likely to generate a sufficient increase in desired outcomes to warrant the investment, or whether the resources would be better deployed elsewhere. One published mathematical modeling study examined the cost-effectiveness of the female condom in preventing HIV and STI in commercial sex workers in rural South Africa. 42 Assuming that the female condom is used as a substitute for a male condom in 25% of the cases in which a female condom is used, the investigator concluded that the introduction of female condoms would be cost saving due to reduced transmission of HIV, syphilis, and gonorrhea. The estimated net savings would persist even if the substitution rate were as high as 75%. Although these results are encouraging, models based on assumed values of key variables may not be sufficiently convincing to decision-makers reluctant to invest in the method. Still needed are real-world behavioral and biologic outcome measures from studies comparing populations with and without access to the female condom as a complement to the male condom. Male condoms versus female condoms—it is not a competition. Accumulating evidence supports the notion that correct and consistent use of either the male or female condom substantially reduces the risk of STI transmission. Proponents of female condom distribution who focus only on the method’s effectiveness, while downplaying or ignoring cost considerations, may be discrediting themselves. Policy makers struggling to allocate limited resources among a host of HIV/AIDS control interventions require information on both effectiveness and costs of proposed interventions. Public health advocates should seek to assist policy makers in uncovering the mix of interventions that achieves maximum public health impact from the options available, within resource constraints.