The Latex Condom: Recent Advances, Future Directions

Chapter 3: User Behaviors and Characteristics Related to Condom Failure

See Also:

Most condom failure occurs as a result of the behaviors of the user, not due to a faulty device. Some user characteristics are also related to condom failure. Hence, understanding which behaviors and characteristics are most strongly associated with condom failure is critical to improving the effectiveness of condoms through public health messages and counseling.

Condom failure (breakage or complete slippage) is often concentrated among certain groups of people, rather than being distributed evenly among all users. What distinguishes this minority of condom users who are at increased risk of condom failure? What are the important background characteristics that are useful for identifying these condom users? What are their specific behaviors that lead to condom failure?

Among men who have used condoms in the past, those who have experienced condom failure are about twice as likely to have a condom failure in the future, compared to men without a past failure. Lack of experience with condoms and several other characteristics may also be associated with condom failure.

Research is less clear about which behaviors lead to condom failure. Behaviors that may be related to failure include opening condom packages with sharp objects, unrolling condoms before putting them on, using oil-based lubricants, having lengthy or intense intercourse, and practicing anal intercourse or vaginal drying.

Better knowledge of characteristics and behaviors that lead to failure may be useful for screening or targeting condom users for counseling, education or other interventions. More understanding of these issues could also improve overall information, education and communication efforts about condoms, as well as help in the development of new types of condoms.

This chapter discusses research regarding condom failure in terms of:

• unequal distribution of condom failure among users
• user characteristics associated with condom failure
• user behaviors associated with condom failure, and
• issues for future research relating to user characteristics and behaviors.

Unequal Distribution of Failure

For most condom users, condom failure is rare. Several published studies that have evaluated the distribution of condom failures among participants suggest that a minority of condom users experience a disproportionate amount of failure. One published study and two reports, however, found more evenly distributed condom failure (see Table 3-1).

The most scientific analysis of uneven distribution patterns involved a prospective study of 177 couples who used 11 condoms each (1,947 total). The study compared the actual distribution of condom failure with a hypothetical distribution assuming condom failure is random and not related to the couple (calculated using a binomial probabil- ity distribution). Sixty-two percent of the couples experienced no failure, almost twice as many as the 37 percent predicted to have no failure by the hypothetical model. In actual use, 29 percent had one to three failures, compared to 62 percent in the hypothetical; 9 percent experienced four or more failures compared to 1 percent in the model. The distributions in actual use and in the hypothetical model were significantly different, showing that condom failure among this cohort was associated with particular couples (p<0.001). (Steiner 1993)

 

This study found that a small minority of users, 9 percent, were responsible for half of all failures. Other studies have shown similar patterns. The retrospective arm of a study among 44 female sex workers from Nevada, USA, gathered data on more than 41,000 condoms used prior to the interview. One woman reported 41 percent of all condom breaks, and three women reported nearly half of all slippage. (Albert 1995) During a study among 540 male and female family planning clients who used more than 3,700 condoms, 3 percent of the participants were responsible for 34 percent of the breaks. (Sparrow) Similarly, among 87 male STD clients who used more than 50 condoms in the previous year, four men experienced 30 percent of the total breakage. (Richters 1993) Distribution of condom failure was uneven, but less pronounced, in two evaluations comparing polyurethane with latex condoms (Nelson 1997, 1996)

In general, these findings suggest that specific characteristics and user behaviors of a minority of users lead to condom failure. Because condom failure is not equally distributed among all users, most condom users are likely to experience condom failure less often than the average failure rates reported in published studies imply. Moreover, published condom failure rates may vary across studies in part because of the differences in characteristics and behaviors among different groups of study participants.

User Characteristics Associated with Condom Failure

A review of studies on condom use and characteristics of users indicates that a history of condom failure and less experience using condoms are risk factors for future failure. Other characteristics that may be associated with condom failure are young age, less education, less income, and large or circumcised penis (see Table 3-2).

A recent prospective study was designed specifically to test the hypothesis that past failure is predictive of future failure. It involved 386 men and 1,810 condoms. The men who reported condom failure during the year prior to the study reported approximately twice as many condom failures during the study as those who had also used condoms but did not experience failure during the year prior to the study (p<0.03, depending on site). (Spruyt)

Other studies have found similar results, as indicated in the first column of Table 3-2. Two prospective trials (805 couples and 3,686 condoms in larger of the two) involving polyurethane and latex condoms found that history of condom breakage was significantly associated with polyurethane condom failure during the study (odds ratio > 2.1, p<0.01). (Nelson 1997, 1996) In general, one should note that factors related to failure of polyurethane and latex condoms may not be the same, because of differences in product attributes (see Chapter 7).

In a subset of 143 couples who used 1,573 condoms during a prospective study, reported condom failure was more than twice as high among couples who had experienced condom breakage in the year prior to the study, compared to those who had also used condoms, but had not experienced breakage (p=0.0001). (Steiner 1993) A multi-site prospective study of 540 family planning clients (3,754 condoms) found that a history of breakage during the three months prior to the study was significantly related to breakage and slippage during the study (p<.001). (Sparrow) A prospective study among 41 sex workers involving 353 acts of vaginal intercourse found that condom slippage during withdrawal was related to slippage prior to the study. (Albert 1995) Although two studies did not find statistically significant associations with past failure, these results may have been a consequence of few cases of failure or relatively small study samples. (Trussell 1992a, 1992b)

Findings from several studies among diverse populations and sites suggest that less experience using condoms is associated with condom failure. In the two large polyurethane studies described above, two measures of condom inexperience (couple and lifetime) were related to breakage and/or slippage. (Nelson 1997, 1996) The prospective study among 540 family planning clients described above found that experienced couples (both partners had used condoms for more than five years) reported significantly less breakage during the study than couples with less condom experience (p<0.003). (Sparrow) Similarly, in the prospective study among 177 couples outlined above, couples who did not use a condom in the year prior to the study were twice as likely to experience condom failure during the study, compared to those who had used at least one condom during the year prior to the study (p<0.001). (Steiner 1993)

Results of retrospective studies conducted in different countries, among diverse groups of men and women (e.g. of various sexual orientations, ages, occupations), also suggest that less experience with condom use is related to condom failure. (Richters 1995; Linberg; Grady; Chan-Chee; Hatcher) Although other factors may contribute, some researchers attribute exceptionally low failure rates among sex workers to the considerable experience the sex workers have using condoms. (Albert 1995; Chan-Chee; Richters 1988) In contrast to such results, two studies did not find an association between level of experience and condom failure. (Spruyt; Trussell 1992b)

Research has also evaluated age, education and income for potential relationships with condom failure. Results from the two evaluations of polyurethane condoms suggest that being 25 years old or younger may be related to condom breakage (Nelson 1997, 1996) A study among commercial sex workers found that breakage was less likely among clients who were more than 30 years old. (Rugpao) Another study including men from health clinics found higher slippage rates among young men. (Richters 1995) At least three studies did not find this association, however. (Spruyt; Grady; Steiner 1993)

Some research also suggests that less education (measured in years of formal schooling) is related to condom failure. (Steiner 1993; Spruyt) Although participants with less education reported more condom failure in both of the polyurethane condom trials and in a U.S. national survey of men, there were not statistically significant differences in these rates (Nelson 1997, 1996; Grady). At least four studies in the U.S. (two prospective, two retrospective) have evaluated the relationship between income and condom failure. All four suggest less income is a risk factor for breakage and/or slippage. (Nelson 1997, 1996; Linberg; Grady)

Among U.S. studies reviewed here, four have evaluated the relationship between ethnicity/race and condom failure. Although these finding are relevant to the U.S., they may have little application to other regions and cultures. Two of these studies suggest that male ethnic/racial minorities are more likely to experience condom failure (Nelson 1997; Grady). Two others reported higher failure rates among male minorities; however, the differences were not statistically significant. (Nelson 1996; Steiner 1993)

Research has begun to evaluate important anatomical and physiological causes that may contribute to condom failure, such as penis size and circumcision status. Closely related to penis size, research has also examined condom failure in terms of different condom widths.

Participants in the two large polyurethane condom trials were asked to measure the circumference and length of their penises while erect. Based on results on polyurethane condoms only, both studies found that larger penis circumference is a risk factor for breakage, and one found that larger circumference is protective against slippage. (Nelson 1997, 1996) One of the studies also suggests that longer penis length is associated with breakage. (Nelson 1996)

The World Health Organization bases its specifications for condom width on consumer preference and penis size, citing three studies. Taken together, the studies show significant variations in penis size within all population groups, but also indicate that men of African descent on average have a slightly wider and longer penis size, Caucasian men have a medium size, and Asian men a slightly narrower and shorter size. (WHO)

Based on the consideration that anatomical differences exist among regions, a series of FHI studies were conducted in three Asian countries to compare small and standard width condoms (49 mm and 52 mm), and in three African countries to compare larger and standard width condoms (55 mm and 52 mm). Among the African sites, breakage rates were slightly higher and slippage was slightly lower for the smaller of the two condoms being compared. (Joanis) However, results from the Asian sites were inconsistent. (Neupane; Andrada) Moreover, almost none of the differences in breakage and slippage rates from either the Asian or African sites were statistically significant. Thus, results from these studies pertaining to penis size and condom failure were inconclusive.

A retrospective study in Britain, where most condoms are 52 mm in width, suggests penis size could affect condom failure. Participants classified themselves as either white, black or Asian. Eighteen percent of blacks, 7 percent of whites and no Asians reported frequent breakage. In contrast, 21 percent of Asians, 8 percent of blacks and 2 percent of whites reported frequent complete slippage. (Tovey)

Data regarding circumcised men and condom failure are limited and inconclusive. Findings of one retrospective study suggest men with a circumcised penis are more likely to experience condom slippage. (Richters 1995) However, neither of the large polyurethane condom trials found a relationship between circumcision status and condom failure. (Nelson 1997, 1996)

User Behaviors and Condom Failure

Prospective studies and anecdotal evidence suggest that several behaviors may be associated with condom failure, including: opening condom packages with sharp objects, unrolling the condom before donning, lengthy or vigorous sex, using excessive lubrication (especially oil-based lubricants), reducing natural vaginal lubrication, anal or oral inter-course, having intercourse in specific positions, and re-use of condoms (see Table 3-3).

A study in Mexico, the Philippines and the Dominican Republic (386 men, 1,810 condoms) was designed in part to evaluate behaviors that may lead to condom failure. Condom breakage was associated with opening condom packages with sharp objects (teeth, scissors, knives or pencils), and both breakage and slippage were associated with unrolling condoms before donning. (Spruyt) Findings from the two large polyurethane condom trials were mixed: difficulty donning condoms was associated with breakage in one study but not in the other. (Nelson 1997, 1996) Another study conducted among male STD and general practice clients provides evidence that pulling condoms on with fingers on the inside may reduce the risk of failure. (Richters 1995)

In the multi-country study described above, slippage was also related to lengthy or intense intercourse. (Spruyt) Similarly, a prospective study among sex workers found breakage was related to lengthy or intense intercourse. However, the studies of polyurethane condoms and the study among male clients do not suggest duration of intercourse is related to failure. (Nelson 1997, 1996; Richters 1995)

A common instruction for proper condom use is to pull out when the penis is still erect and to hold the condom when pulling out. However, in the multi-country study, more than half of the withdrawals took place after loss of erection, and 44 percent of the condoms were not held during withdrawal. Despite such high rates, these behaviors did not appear to be associated with condom slippage in the bivariate analysis. (Spruyt) Sizable proportions of participants in other studies of the general population have also reported these two behaviors. (Oakley)

Another common instruction for condom use is not to use oil-based lubricants. This is based on laboratory research that suggests that latex can deteriorate rapidly when oil-based lubricants are added, making them more likely to break, (Voeller) and on anecdotal evidence of breakage and slippage when using such oil-based lubricants as petroleum jelly, hand lotion, or mineral or vegetable oils.

In two studies that have evaluated the impact of adding oil-based lubricants, both have found an association with slippage. A prospective study, using new and aged condoms pre-lubricated with silicone, compared breakage and slippage rates with no additional lubricant, adding oil-based lubricants, and adding water-based lubricants. Adding an oil-based lubricant resulted in higher slippage rates, which were statistically significant (p=0.004). Adding oil-based lubricants also resulted in the highest breakage rates for both new and aged condoms, but the differences were not statistically significant when compared to using no lubricant or a water-based lubricant. (Steiner 1994) Findings of a study among 70 family planning clients also suggests that use of additional oil-based lubricants is related to slippage. (Trussell 1992a)

In the prospective study comparing various lubricants, water-based lubricants decreased breakage rates among old condoms and showed some increase in slippage rates (although not statistically significant). The study concluded that the negative impact of water-based lubricants on condom slippage may be outweighed by the protective influence on breakage, especially with older condoms. (Steiner 1994)

Another study found that using water-based lubricants was significantly associated with lower condom breakage rates but not with increased condom slippage. (Gabbay) The study compared 434 women seeking post-coital contraception after recent condom failure with a control group of 393 women seeking other health services. Among the study group (all of whom reported recent failure), fewer than 2 percent reported that they had used additional water-based lubricants with their condom. Among the 133 controls who reported no history of condom failure, 14 percent reported using additional water-based lubricants; among the 270 controls reporting some condom failure, 4 percent reported using water-based lubricant. Use of water-based lubricants was associated with lower failure rates, comparing the study group both to controls with no history of failure (odds ratio=8.9) and to controls who had experienced failure (odds ratio=3.6). Moreover, the relationship between use of water-based lubricants and increased slippage rates was not significant (odds ratio=1.0).

In one study in which additional lubricants used were almost exclusively water-based and among several studies in which the lubricant type is unspecified, most findings suggest no association between lubricant use and condom failure. (Rugpao; Nelson 1997, 1996; Richters 1995; Sparrow; Trussell 1992b; Chan-Chee) Regarding saliva use as a lubricant, a study among 540 family planning clients found it to be associated with breakage (p=0.015). (Sparrow)

There is limited research concerning the impact of altering vaginal lubrication and size on condom breakage. Although the prevalence of this behavior is unknown, several vaginal preparation methods have been documented in Zimbabwe, Zaire and other sub-Saharan African countries, as well as Saudi Arabia, Haiti and Costa Rica. (Van de Wijgert; Civic; Pitts; Brown RC; Brown JE; Runganga) Women have reported douching with various solutions; wiping with newspaper, tissue or cloth; and inserting leaves, powders, cotton, wool or tampons into the vagina before sex. The purpose is to make the vagina drier and tighter, in order to reduce vaginal secretions and enhance sexual pleasure for both partners. Results of prospective research among sex workers, in addition to findings of focus group interviews among health clinic attendees and providers, suggest that these types of practices are associated with condom breakage. (Van de Wijgert; Runganga)

Data from several studies suggest that condom use during anal intercourse is associated with condom failure. (Linberg; Richters 1995; Sparrow; Grady; Chan-Chee) One of these also found an association between oral sex and failure. (Sparrow) Findings are limited and inconsistent regarding the influence of sexual position and condom failure. (Nelson 1997, 1996)

Conclusion and Future Research

More research is needed concerning user characteristics and behaviors that may be associated with condom failure. Sufficient data should be collected to enable analysis of potential correlates of condom breakage and slippage (e.g. detailed data on each condom used, regardless of whether failure occurred). Future research will also benefit from increasing the probability of cases of condom failure in a study population by oversampling (to a greater extent than in previous research) of condom users who have experienced condom failure in the past.

Key research questions that need attention are:

• With respect to characteristics of condom users, are age, education level, penis size or other characteristics predictive of failure?

• Are behaviors that have not been studied, such as squeezing air out of the condom tip and individual condom storage practices, related to condom failure?

• What is the relative contribution of specific behaviors to breakage and slippage, and what is the relative prevalence of these behaviors? If a given behavior is likely to result in condom failure, but almost never occurs, it should have a minor impact on failure rates. If a behavior is only weakly associated with failure, but commonly occurs, efforts to curb the behavior may have a significant impact on overall condom failure.

• Are factors related to condom failure different for non-latex versus latex condoms?

Building on current knowledge of characteristics and behaviors associated with condom failure, targeted intervention studies to test the applicability of these findings on counseling strategies would be useful. Future human use trials could be designed to determine if condom failure can be reduced through identifying condom users at increased risk of condom failure and providing them with counseling to discourage behaviors associated with condom failure.

A history of condom failure and less experience using condoms are risk factors for future failure. This knowledge could be built into personal counseling, information brochures and mass media messages about condom use. Thus, persons with these characteristics could be alerted to their greater potential risk of condom failure. While research findings on other characteristics are not as clear, characteristics that may be associated with condom failure include young age, less education, less income, and large or circumcised penis.

Research is less clear about which behaviors lead to condom failure. Several behaviors that may be associated with condom failure --opening a condom package with sharp objects, unrolling condoms before putting them on, using oil-based lubricants and re-using condoms -- are addressed by instructions for correct condom use. Other behaviors that may be associated with condom failure include lengthy or vigorous sex, anal sex or vaginal drying.

It is important to remember that condom failure is often concentrated among a minority of condom users. Certain characteristics and behaviors have been shown to be related to condom failure. For most condom users, condom failure is rare.

by Alan B. Spruyt

References

1. Albert AE, Warner DL, Hatcher RA, et al. Condom use among female commercial sex workers in Nevada's legal brothels. Am J Public Health 1995;85:1514-20.
2. Albert AE, Hatcher RA, Graves W. Condom use and breakage among women in a municipal hospital family planning clinic. Contraception 1991;43:167-76.
3. Andrada A, Ravelo N, Spruyt A, et al. Acceptability and Functionality of Standard and Smaller Latex Condoms during Human Use: Philippines. Durham, NC: Family Health International, 1993.
4. Brown RC, Brown JE, Ayoa OB. The use and physical effects of intravaginal substances in Zairian women. Sex Transm Dis 1993;20:96-99.
5. Brown JE, Ayoa OB, Brown RC. Dry and tight: sexual practices and potential AIDS risk in Zaire. Soc Sci Med 1993;37:989-94.
6. Chan-Chee C, De Vincenzi I, Sole-Pla M, et. al. Use and misuse of condoms (letter). Genitourin Med 1991;67:173.
7. Civic D, Wilson D. Dry sex in Zimbabwe and implications for condom use. Soc Sci Med 1996;42:91-98.
8. Folmar S, Alam S, Sharif A. Condom Use in Bangladesh: Final Report. Durham, NC: Family Health International, 1992.
9. Gabbay M, Gibbs A. Does additional lubrication reduce condom failure? Contraception 1996;53:155-58.
10. Grady WR, Tanfer K. Condom breakage and slippage among men in the United States. Fam Plann Perspect 1994;26:107-12.
11. Hatcher RA, Hughes MS. The truth about condoms. SIECUS Report 1988;17:1-9.
12. Joanis C, Brookshire T, Piedrahita C, et al. Evaluation of Two Condom Designs: A Comparison of Standard and Larger Condoms in Ghana, Kenya, and Mali. Durham, NC: Family Health International, 1990.
13. Linberg L, Sonenstein F, Ku L et al. Young men's experience with condom breakage. Fam Plann Perspect 1997;29:128-31.
14. Nelson A, Bernstein GS, Frezieres R, et al. Study of the efficacy, acceptability and safety of a non-latex (polyurethane) male condom - final report, NIH contract NO1-HD-1-3109. Unpublished paper. National Institutes of Health, 1997.
15. Nelson A, Frezieres R, Walsh T, et al. Controlled, randomized evaluation of a commercially available polyurethane and latex condom (Avanti versus Ramses Sensitol) - final report, NIH contract NO1-HD-1-3109. Unpublished paper. National Institutes of Health, 1996.
16. Neupane S, Abeywickrema D, Martinez K, et al. Acceptability and Actual Use Breakage and Slippage Rates of Standard and Smaller Latex Condoms: Nepal and Sri Lanka. Durham, NC: Family Health International, 1992.
17. Oakley D, Bogue E. Quality of condom use as reported by female clients of a family planning clinic. Am J Public Health 1995;85:1526-30.
18. Piedrahita C, Steiner M. Why condoms break. IPPF People 1992;19:44.
19. Pitts M, Magunje N, McMaster J. Students' knowledge of the use of herbs and other agents as preparation for sexual intercourse. Health Care Women Int 1994; 15:91-99.
20. Richters J, Donovan B, Gerofi J. How often do condoms break or slip off in use? Int J STD AIDS 1993;4:90-94.
21. Richters J, Gerofi J, Donovan B. Why do condoms break or slip off in use?: an exploratory study. Int J STD AIDS 1995;6:11-18.
22. Richters J, Donovan B, Gerofi J, et al. Low condom breakage rate in commercial sex (letter). Lancet 1988;24:1488.
23. Rugpao S, Beyrer C, Toanabutra S, et al. Multiple condom use and decreased condom breakage and slippage in Thailand. J of Acquired Immune Def Syndr and Hum Retrovirol 1997;14:169-73.
24. Runganga A, Pitts M, McMaster J. The use of herbal and other agents to enhance sexual experience. Soc Sci Med 1992;35:1037-42.
25. Sparrow MJ, Lavill K. Breakage and slippage of condoms in family planning clients. Contraception 1994;50:117-29.
26. Spruyt A, Steiner MJ, Joanis C, et al. Identifying condom users at risk of breakage and slippage: three international sites. Am J Public Health 1998 (in press).
27. Steiner M, Piedrahita C, Glover L, et al. Can condom users likely to experience condom failure be identified? Fam Plann Perspect 1993;25:220-26.
28. Steiner M, Piedrahita C, Glover L, et al. The impact of lubricants on latex condoms during vaginal intercourse. Int J STD AIDS 1994;5:29-36.
29. Tovey SJ, Bonell CP. Condoms: a wider range needed. (letter) BMJ 1993;307:987.
30. Trussell J, Warner DL, Hatcher RA. Condom slippage and breakage rates. Fam Plann Perspect 1992a;24:20-23.
31. Trussell J, Warner D, Hatcher R. Condom performance during vaginal intercourse: comparison of Trojan-Enz and Tactylon condoms. Contraception 1992b;45:11-19.
32. Van de Wijgert, JH. The Effect of Douching, Wiping and Inserting Herbs inside the Vagina on the Vaginal and Cervical Mucosa, on the Vaginal Flora, and on the Transmission of Human Immunodeficiency Virus and Other Sexually Transmitted Diseases in Women in Zimbabwe. Unpublished Dissertation. University of California at Berkeley, 1997.
33. Voeller B, Coulson AH, Bernstein GS, et al. Mineral oil lubricants cause rapid deterioration of latex condoms. Contraception 1989;39:95102.
34. WHO Global Programme on AIDS. Specification and Guidelines for Condom Procurement. Appendix VII, Regional or Ethnic Differences in Erect Penis Size. Geneva: WHO, 1995.

Return to table of contents