OBJECTIVES: This study examined whether past condom failure (breakage, slippage, or both) can predict future failure and evaluated other predictors of condom failure. METHODS: At each of 3 international sites, approximately 130 male condom users were enrolled and given 5 condoms to use for vaginal intercourse over a 3-week period. RESULTS: Men at increased risk (history of 1 or more condoms that broke or slipped off) reported approximately twice as many condom failures as those not in this group. Condom failure increased with the number of adverse condom use behaviors reported per participant. Opening condom packages with sharp objects and unrolling condoms before donning were associated with breakage. Unrolling condoms before donning and lengthy or intense intercourse were associated with slippage. Of background characteristics evaluated, having less education was associated with condom failure. CONCLUSIONS: These data suggest that a history of condom failure predicts future failure, a finding that may be useful for targeted intervention. Moreover, these data provide further evidence that certain behaviors and lower educational attainment are associated with condom failure.

SUMMARYThe objective was to evaluate the impact of additional lubricant on condom breakage and slippage. Two hundred and sixty-eight couples used six new and six aged condoms during vaginal intercourse and were instructed to use two of each type with either water-based lubricant, oil-based lubricant or no additional lubricant.The use of either oil-based or water-based lubricant increased slippage rates of new and aged condoms, although only one pairwise comparison (oil-based lubricant vs. no additional lubricant) was statistically significant (8.5% vs. 3.8%, p -.004).The use of oil-based lubricant increased clinical breakage, although not statistically significantly, in both new and aged condoms. Waterbased lubricant did not impact the clinical breakage rate of the new condoms and decreased the breakage rate of the aged condoms (no addi tional lubricant 4.5% vs. water-based lubricant 2.1%, p -.029).From a functional perspective, this study suggests that condom users should be told not to use oil-based lubricants. The negative impact of water-based lubricant on slippage may be outweighed by the protective influence on clinical breakage, especially for aged condoms.Over three-fourth of the couples (76%) had at least some incorrect knowledge, according to current condom instructions, of the type of lubricant that should be used with condoms. If the study packet contained a lubricant, the participants were instructed to don the condoms and then lubricate them well with the lubricant found in that packet before using the condoms during vaginal intercourse.Participants were sent the three study packets and were asked to choose the order in which they used the study packets and the condoms within each study packet. Upon receipt of the completed questionnaires, couples were remunerated for each condom used. Definition of Breakage, Slippage and Total FailureThe self-administered questionnaire asked couples a series of specific questions about each condom used ( For the tests of differences in preference ratings, alpha was also set at .i .05 for the family of three paired comparisons.Each paired comparison was assessed at .017.Approximate 95 percent confidence intervals for condom failure rates were calculated using the normal approximation to the binomial distribution (with an added continuity correction factor) . RESULTSBackground Characteristics

and the United States shows that condom breakage rates during vaginal intercourse using lubricated latex condoms range from 0.6% of all condoms used in Sri Lanka to 13.3% in Ghana. Most research sites reported breakage rates below 5%. The rate at which the condom slipped off completely is as high as 9.3% in Kenya, with most of the remaining sites reporting rates below 4%. When breakage and slippage are combined, total condom failure rates range from 3.8% to 13.3%. Although such high condom failure rates may cause alarm, there is evidence that for a majority of users, condoms provide an effective method of preventing pregnancy and sexually transmitted disease infection if they are used correctly. The high overall breakage and slippage rates may be caused by incorrect behavior or by certain characteristics of a few participants.

In the absence of well-controlled studies on the clinical effectiveness of condoms against pregnancy and sexually transmitted diseases, breakage and slippage data may provide the best indication of the protection offered by condom use. According to the recent literature, condom breakage rates range from 0% to 12%, with many US studies falling in the 2% to 5% range. Few studies have collected slippage data. In addition to discussing methodological issues associated with these studies, we propose a standardized protocol for future condom breakage and slippage trials and discuss how results may be used to estimate perfect-use and typical-use pregnancy rates.

No abstract