Do female condoms work?

Standard permeability tests have established that the female condom is no more likely to let through pathogens or sperm cells than the male condom. This leads to mathematical modelling which suggests that consistent and correct use of female condoms by an HIV-negative women with an HIV-positive partner, if they have sex twice a week for a year, could reduce her chance of acquiring HIV by about 90% – similar to the maximum achievable for consistent male condom use (even consistent and correct use does not result in 100% protection with condoms, as the chances of breakage and slippage cannot be eliminated entirely).

Trials in the United States and Britain show a pregnancy failure rate of 2.4% when the female condom is used perfectly, and 12.2% with typical use (source: Female Condom Company). That compares to a 2% pregnancy failure rate for perfectly used Kitemark condoms, rising to up to 15% for imperfectly used condoms (source: Durex).

There have been very few randomised studies of female condom use, and all have been of the FC1 polyurethane condom.

In a US study in 2003,¹ 1442 women attending an STI clinic were randomly assigned to receive free female or male condoms and small-group education on their use. There was a 21% reduction in new STI incidence at subsequent screening in the female condom users, which was almost statistically significant (p = 0.07).

In one study in the late 1990s in Thailand,² sex workers in four cities were randomised into two groups: 255 women were instructed to use male condoms consistently, while 249 women were provided with female condoms as well and had the option of using them if clients did not want to use male condoms. Male condom use was lower in the male/female condom group when compared with the male condom group (88.2 and 97.5%, respectively, p=< 0.001).

However, this reduction in male condom use was counterbalanced by the use of female condoms in 12% of all sexual acts in the male/female condom group, contributing to a 17% reduction in the proportion of unprotected sexual acts in this group when compared to the male condom group (5.9 versus 7.1%, though this was not statistically significant (p=0.16)).

A 2005 study in Madagascar³ enrolled 1000 sex workers and supplied and promoted male condoms for six months. They then supplied female condoms in addition to male condoms for the following six months. The mean proportion of sex acts that were protected increased from 57% at baseline to 88% at 18-month follow-up. Male condom use at six months was 78%; at 12 months this had dropped to 64% but was more than compensated for by a 19% rate of female condom use. At 18 months, 68% of sex acts involved male condoms and 20% female condoms. The proportion of clients with any STI declined from 49% at six months to 40% at 18 months. However, the proportion of sex acts with primary and non-paying partners that were protected, though increasing from 20% at baseline to 30% at two months, did not increase further.

A large multi-country trial by the World Health Organization (WHO)³ compared pregnancy rates in 1071 women using male condoms (344 women), female condoms (482 women) or both (131 women) in China, Panama, Nigeria and South Africa. At six months pregnancy rates were statistically indistinguishable between users of male versus female condoms. Discontinuation rates were interestingly different, however: in China less than 1% of the women using male condoms stopped using them but over 15% of the female condom users did. Conversely, in South Africa 78% of women stopped using male condoms in accordance with the study protocol but only 36% of the female condom users.

An overall survey⁴ of 14 randomised and observational studies of adding-in the female condom to the prevention choices available to women, including the US, Thai and Madagascar studies above, found that eleven studies reported some positive effect on STI incidence, pregnancy or protected sex acts.