The Latex Condom: Recent Advances, Future Directions

Chapter 1: Pregnancy and STD Prevention

See Also:

Condoms are the only contraceptive method proven to reduce the risk of all STDs, including HIV. Thus they can be used as a dual-purpose method, both for contraceptive and prophylactic reasons. Also, they can be used as part of a dual-method regimen, in which another method is used primarily for contraception, and the condom is used specifically for disease prevention.

Studies have found that consistent and correct use of condoms is by far the most important factor in preventing both pregnancy and disease, compared to failure of the device through breakage or slippage. Pregnancies reported with condom use are due primarily to inconsistent and incorrect use, not to defective condoms. Failure of the condom itself is rare. Research has shown that exposure to semen is nearly always due to non-use, with a very minor portion due to breakage during use and virtually none due to holes resulting from the manufacturing process. (FDA)

This chapter summarizes the effectiveness of latex condoms in preventing pregnancy and STDs, and the ability of latex condoms to block passage of genital fluids and their constituents between sex partners. It discusses various factors that influence contraceptive effectiveness and can lead to disparate estimates of pregnancy rates. The chapter shows that even among couples with one partner infected with HIV, consistent condom use results in a near-zero risk of HIV transmission.

Factors that Influence Pregnancy Rates

Results of studies of the contraceptive effectiveness of condoms vary considerably, as they do for other coital-dependent contraceptive methods. Twelve-month life table pregnancy rates for condom users have ranged from 2 to 14 per 100 women. (Hatcher) This compares with about 85 per 100 couples who would conceive a pregnancy in a year if not using contraception. These rates are slightly better than for other barrier methods but not as good as for hormonal methods or IUDs.

Variations in reported pregnancy rates, for condoms as well as other coital-dependent methods, derive from several factors: correct and consistent use, the capacity to conceive, and the frequency and timing of intercourse. (Steiner) In addition, the study methodology can affect the rates.

Correct and Consistent Use. Contraceptive use results in different pregnancy rates among different types of study participants (single or married, adults or teenagers, educated or less educated, etc.). Much of this variation stems from differences in the consistency and correctness of using the device, and leads to consideration of two terms: perfect use and typical use pregnancy rates. Providers and users need to become familiar with both perfect use and typical use pregnancy rates -- estimates of optimal and average pregnancy risk -- to ensure informed choice of family planning methods.

The lowest pregnancy rate follows from perfect use. This refers to the experience of those who report using the method exactly as it should be used (correctly) and at every intercourse (consistently). Estimated pregnancy rates during perfect use of condoms is 3 percent at 12 months. (Trussell) The pregnancies that do occur in reported perfect use are due either to condom failure or to inaccurate reports from study participants.

A recently completed randomized controlled trial of latex versus male plastic condoms reported pregnancy rates for couples who used condoms at every intercourse, called "consistent use" rates. The six-month consistent use rates were 1.2 per 100 women using latex condoms and 2.6 per 100 women using plastic condoms, which would equate with 2 to 4 pregnancies per 100 women at 12 months. (A Nelson)

Most frequently cited is the pregnancy rate in typical use: how many unintended pregnancies would occur in the first year among 100 couples using a method. This rate can include perfect and imperfect use. The pregnancy rate during typical use can be much higher than during perfect use, especially for a coital-dependent method like the condom. Multiple studies can help identify the range of typical pregnancy rates and point out sub-groups that are particularly well or poorly suited to use a method.

Many assume that pregnancy rates cited for condom users correspond to the proportion of condoms that fail, equating, for example, a 10 percent typical pregnancy rate with 10 percent condom failure. In fact, condom failure is rare. Condom failure refers to the device breaking or slipping off completely during intercourse. The higher typical-use rates are due primarily to inconsistent and incorrect use, not to condom failure. Research has shown that only a minority of users report condom failure. Also, condom failure does not invariably lead to pregnancy.

Two models demonstrate the crucial importance of consistent condom use to contraceptive effectiveness. In both cases, inconsistent condom use has a far greater impact on increased pregnancy rates than condom breakage, regardless of the cause of condom breakage.

One model suggests that failing to use a condom during even a single fertile cycle has more impact on annual probability of pregnancy than does higher condom breakage. With condom breakage rates of 2 percent, if there is one cycle of non-use out of 13 cycles in the year, the expected annual probability of pregnancy jumps more than four-fold, from 3.8 to about 18 per 100 women. In contrast, if the condom breakage doubles from 2 percent to 4 percent, but condoms are used all the time, the annual probability of pregnancy approximately doubles from 3.8 to 7.5 per 100 couples (see Figure 1-1).* (Dominik)

A second model, developed by the U.S. Food and Drug Administration (FDA), estimates the average semen exposure per coital act due to condom non-use, condom breakage and holes in condoms used. Semen exposure due to breakage was assumed to occur in about 2 percent of devices used. Semen exposure due to holes in the latex membrane was estimated from earlier in vitro work. (Carey et al; Lytle) Given the relatively low frequency of breakage, and the size and frequency of holes in the devices, exposure to semen was found to be nearly always due to non-use, with a very minor portion due to breakage and virtually none due to holes. (FDA)

Capacity to Conceive. The inherent fecundability of a couple (the capacity to become pregnant), which is influenced by various physiologic factors, is also crucial to the observed pregnancy rates. For example, within the reproductive years, older persons are less fertile on average than younger. Women with irregular, or extremely short or long, menstrual cycles may have hormonal abnormalities and hence be less fecund. People with a history of STD may be infertile, such as women who have had pelvic inflammatory disease. The dimensions of fecundability may be difficult to measure, however, and there is no single index that encapsulates the concept. Nevertheless, it is likely to contribute to observed variability in pregnancy rates across different user populations.

Frequency and Timing of Coitus. Conception can occur only during a few specific days during the menstrual cycle. The more often a couple has intercourse, the more likely they will do so during these days and be exposed to pregnancy. Ideally, the frequency and timing of intercourse must be accounted for in the calculation of pregnancy rates. Therefore, coital logs are an important design feature of any condom effectiveness study. The condom study with the lowest reported pregnancy rate was conducted among older users with (presumably) a lower coital frequency, and with a lower likelihood of conception than younger users. (Potts)

Given the strict inclusion criteria in most clinical trials, cohorts are likely to differ from the general population in the above factors that affect pregnancy rates. An inclusion criterion for parity, or one for a threshold coital frequency, may be associated with greater fecundity and a higher pregnancy rate, for instance. Conversely, frequent follow-up and counseling of study participants would tend to improve compliance, which would depress pregnancy rates. Thus, the results of clinical studies must be generalized with care.

Study Methodology. Apart from behaviors and characteristics of the study cohorts, there are also methodological sources for the variation in pregnancy rates. (This is true for other methods as well as condoms.) The timing of data collection is important. A retrospective study with interviews about past contraceptive behavior and pregnancy history will almost certainly yield different results than a prospective study that requires regular clinic visits and pregnancy testing. Most condom effectiveness data are from retrospective studies. This shortcoming will be rectified by prospective studies recently completed or currently under way. (A Nelson)

Different computations can yield widely differing pregnancy rates. A Pearl index is the ratio of the number of pregnancies observed in a study divided by the total number of years that all study participants have been studied. Pearl rates decline over the length of a study because participants likely to conceive a pregnancy tend to do so early, leaving more consistent and correct (or less fecund) users in the study.

The life table method is the preferred method for calculating effectiveness because each month of the study is treated as a separate calculation. The monthly proportions are multiplied together to arrive at a cumulative proportion conceiving, usually within six or 12 months of beginning the method. Only participants still in the study in a given month are considered at risk of pregnancy during that interval, but participants who become pregnant, leave the study for other reasons, or are lost to follow-up are all accounted for, since the results from all months are included. Unlike the Pearl index, life table pregnancy rates vary from zero to 100, rise with the length of the study, and can take into account other reasons for leaving the study besides pregnancy.

Effectiveness of Latex Condoms against Disease

Unlike pregnancy prevention, in which the unintended outcome can only occur during a portion of the menstrual cycle and during the reproductive years, disease transmission can occur during each and every act of intercourse for all persons, regardless of age. Despite this heightened risk, inconsistent or non-use of condoms is common even among persons at very high risk of STD exposure.

Numerous studies have been conducted on the risk of STDs in condom users. These studies usually present a summary relative risk measure, i.e., a comparison of the disease risk in condom users versus the risk in non-users. For each relative risk estimate, a value of less than 1.0 indicates protection, or a lower risk in the condom group. Virtually all clinical and epidemiological studies have found substantial reductions in the risk of disease among condom users. (Cates; Daly)

It appears that latex condoms, with or without the use of spermicides, are effective in preventing disease. The presence or absence of spermicides does not appear to increase protection against STDs. FHI data show that cervical and vaginal STD rates among female sex workers using plain lubricated or spermicidally lubricated condoms were approximately equal. (Roddy)

As with pregnancy prevention, using condoms consistently results in better protection from disease. Population-level data from Thailand show that a condom promotion program that reportedly led to near-universal condom use during commercial sex is associated with steep declines in the nationwide numbers of bacterial STD cases, (Hanenberg) and reduced HIV prevalence in male Thai military conscripts. (KE Nelson)

The most convincing data on risk factors for HIV infection come from prospective studies of serodiscordant couples, those where one partner is infected and the other is not. In these situations, regular sexual exposure to an infected partner is known, and more accurate record-keeping on condom use is usually possible than in retrospective surveys. (Feldblum) Three recent studies of heterosexual serodiscordant couples have compared rates of HIV seroconversion (where the uninfected partner becomes infected) in couples using condoms with varying consistency (see Table 1-1).

Two extremely important conclusions are supported by all three studies. First, with consistent condom use, the HIV infection rate among the uninfected partners was less than 1 percent per year. Second, in situations where one partner is definitely infected, inconsistent condom use can be as risky as not using condoms at all.

In a multi-center Italian study that followed 305 sexually active seronegative female sexual partners of HIV-infected men for a median of 24 months, 3.9 infections occurred per 100 person-years (py). (Saracco) The HIV incidence was reduced by 84 percent in women who always used condoms compared with women who used them inconsistently or never (rate ratio 0.2; 95% confidence interval [CI] 0.1-0.5).* Of the 171 women who always used a condom, 1.7 percent became infected. The percent infected was higher among those using condoms some of the time (14.5 percent of 55 women), compared with those who never used condoms (10.1 percent of 79 women).

A multi-country European collaborative study enrolled 378 seronegative regular partners of HIV-infected men or women; 256 of the couples continued to have vaginal or anal intercourse. (deVincenzi) About one half of the couples used condoms at every intercourse, and no seroconversions occurred among these couples (95% CI 0-1.5 per 100 py). Nearly 10 percent of the 121 couples who used condoms inconsistently or not at all seroconverted, an HIV incidence rate of 4.8/100 py (95% CI 2.5-8.4). These infections occurred even though 50 percent of the inconsistent users reported using condoms at least half the time.

In a Haitian serodiscordant couples study, 63 percent of couples ceased sexual activity soon after enrollment. The HIV seroconver-sion rate among the 177 couples who remained sexually active was 5.4 per 100 py. (Deschamps) The infection rate was 1.0 per 100 couples who always used condoms (2.4 percent of 42 couples), compared with 6.8 per 100 couples who used condoms inconsistently or not at all (rate ratio 0.1). As in the Italian study, there was almost no difference in the risk of infection between inconsistent users and non-users. Of the 45 inconsistent users, 13.3 percent became infected; of the 90 non-users, 14.4 percent became infected.

These three studies show that consistent condom use is extremely effective against HIV transmission. Even with regular exposure to infection, self-reported consistent condom users have a near-zero risk of HIV. However, among these serodiscordant couples, inconsistent condom use carries considerable risks of HIV infection, since exposure to infection is guaranteed at every unprotected intercourse. In this sexual context, inconsistent condom use offers little protection against HIV, compared with non-use.

In most circumstances, however, condom use, even if inconsistent, has considerable public health benefit. Unlike serodiscordant couples, the majority of persons at risk of HIV infection do not have a guaranteed risk of infection with each sexual encounter. Hence, most persons will reduce their risk of infection by using condoms even inconsistently. This is because some of the non-use of condoms takes place with uninfected partners, and some condom use takes place with infected partners.

Relative cell sizes 

	Neisseria Gonorrhoeae (Gonorrhea)
Treponema Pallidum (Syphilis)		HIV
	Sperm

 

Latex Membrane Permeability

Latex condoms are virtually certain of blocking passage of genital fluids and their constituents between sex partners. The organisms that condoms are designed to block range in size from spermatozoa, about 0.003 millimeters (mm), or 3000 nanometers (nm) in largest diameter, through the STD pathogens: N. gonorrhoeae (800 nm), C. trachomatis (200 nm), HIV (125 nm) and hepatitis B virus (40 nm).

To be effective for disease prevention, defects in latex membranes must not allow small pathogens to pass through otherwise intact devices. Scanning electron microscopy of latex condoms reveals pits and imperfections but no pores that penetrate the entire membrane. (Kish) Laboratory studies generally have found that viruses do not pass through intact latex condoms, even when the devices are stretched and stressed. (Conant; Katznel-son; Minuk; Van de Perre; Reitmeijer; Judson)

Two FDA teams have devised laboratory tests to simulate body fluid transfer to test the permeability of latex condoms. Carey and colleagues used fluorescing plastic micro-spheres 110 nm in diameter, approximately equivalent in size to HIV. In addition, the degree of acidity, surface tension, viscosity, temperature, pressure, geometry and time were all considered and controlled in the study of 89 condoms. Although 29 condoms showed evidence of leakage, the estimated amount of fluid crossing the membrane was 0.1 microliter or less, corresponding to 0.01 percent of typical ejaculate, an amount that would be expected to be virus-free in an HIV-infected man. (Carey) Thus, even with pinhole leakage in some devices, the authors estimated that condom use would decrease exposure to HIV 10,000-fold. They concluded from this simulation that the latex condom offers extremely reliable protection against HIV but does not totally eliminate the risk. (Carey)

The second FDA study tested 470 latex condoms using a virus of about 30 nm in diameter (i.e., about four times smaller than HIV), controlling viscosity, surface tension and pressure. With 12 condoms (2.6 percent), minute amounts of viral particles went through the latex membrane, a median of 7 x 10-4 ml. (Lytle) But the study estimated that actual use conditions would allow 100-fold less viral penetration, because of the assumptions made in the laboratory. Like the Carey study, this laboratory simulation found that latex condoms are extremely effective but not absolutely perfect barriers to viral passage.

Conclusion

Laboratory research shows that intact latex condoms form a barrier that is nearly always impermeable to spermatozoa and pathogens. It follows that consistent and correct use of latex condoms substantially reduces the twin risks of unintended pregnancy and STDs, although those risks are not zero.

It is important to remember the difference between perfect use and typical use in studies reporting pregnancy rates. Another important message is that spermicidally lubricated condoms do not seem to improve protection against cervical and vaginal STDs over non-spermicidally lubricated condoms.

Two important public health messages emerge from the research reviewed here. First, the latex condom is an extremely reliable device in terms of its ability to block the passage of sperm or pathogens. Second, consistent and correct use of condoms is far more important in preventing both pregnancy and disease than the failure of the device through breakage or slippage. It is important to remember that a 10 percent pregnancy rate, for example, does not mean the condoms failed 10 percent of the time. These pregnancy rates occur primarily because of inconsistent and incorrect use, not condom failure, which is rare.

*The model assumes that the probability of pregnancy during a single unprotected act during the fertile period is 0.15, that there is one act per fertile period, that there are 13 cycles per year, that a condom that breaks provides no protection, and that a condom that does not break is 100 percent effective. (Dominik)

*"Rate ratio" is a measure of relative risk. "Confidence interval" is a measure of the precision of a measurement, in this case, the rate ratio.

References

by Paul J. Feldblum

References

1. Carey RF. Letter in response to "Condom safety and HIV." Sex Transm Dis 1994;21:60.
2. Carey RF, Herman WA, Retta SM, et al. Effectiveness of latex condoms as a barrier to human immunodeficiency virus-sized particles under conditions of simulated use. Sex Transm Dis 1992;19:230-34.
3. Cates W Jr, Stone KM. Family planning, sexually transmitted diseases and contraceptive choice: a literature update part I. Fam Plann Perspect 1992;24:75-84.
4. Conant MA, Spicer DW, Smith CD. Herpes simplex virus transmission: condom studies. Sex Transm Dis 1984; 11:94-95.
5. Daly CC, Helling-Giese GE, Mati JK, et al. Contraceptive methods and the transmission of HIV: implications for family planning. Genitourin Med 1994;70:110-17.
6. Deschamps M-M, Pape JW, Hafner A, et al. Heterosexual transmission of HIV in Haiti. Ann Intern Med 1996;125:324-30.
7. de Vincenzi I for the European Study Group on Heterosexual Transmission of HIV. A longitudinal study of human immunodeficiency virus transmission by heterosexual partners. N Engl J Med 1994;331:341-46.
8. Dominik R. Male condom evaluation: statistical considerations for equivalence studies and extrapolating breakage and slippage to pregnancy rates. Presented at the NIH/FDA Workshop on Contraceptive Efficacy and STD Prevention: Issues in the Design of Clinical Trials. Bethesda, MD, April 6-8, 1994.
9. Feldblum PJ, Morrison CS, Roddy RE, et al. The effectiveness of barrier methods of contraception in preventing the spread of HIV. AIDS 1995; 9(suppl A):s85-s93.
10. Food and Drug Administration (FDA). Office of Science and Technology -- Annual Report, Fiscal Year 1996. Rockville, MD: U.S. Food and Drug Administration, 1997.
11. Hanenberg RS, Rojanapithayakorn W, Kunasol P, et al. Impact of Thailand's HIV-control programme as indicated by the decline of sexually transmitted diseases. Lancet 1994;344:243-45.
12. Hatcher RA, Trussell J, Stewart F, et al, eds. Contraceptive Technology, 16th Revised Edition. New York: Irvington Publishers Inc., 1994.
13. Judson FN, Ehret JM, Bodin GF, et al. In vitro evaluations of condoms with and without nonoxynol-9 as physical and chemical barriers against chlamydia trachomatis, herpes simplex virus type 2, and human immunodeficiency virus. Sex Transm Dis 1989;16:51-56.
14. Katznelson S, Drew WL, Mintz L. Efficacy of the condom as a barrier to the transmission of cytomegalovirus. J Infect Dis 1984;150:155-57.
15. Kish LS, McMahon JT, Bergfeld WF, et al. An ancient method and a modern scourge: the condom as a barrier against herpes (letter). J Am Acad Dermatol 1983; 9:769-70.
16. Lytle CD, Routson LB, Seaborn GB, et al. An in vitro evaluation of condoms as barriers to a small virus. Sex Transm Dis 1997;24:161-64.
17. Minuk GY, Bohme CE, Bowen TJ. Condoms and hepatitis B virus infection (letter). Ann Intern Med 1986;104:584.
18. Nelson A, Bernstein GS, Frezieres R, et al. Study of the efficacy, acceptability and safety of a non-latex (polyurethane) condom. Final Report for NIH project N01-HD-1-3109. Unpublished paper. National Institutes of Health, 1997.
19. Nelson KE, Celentano DD, Eiumtrakol S, et al. Changes in sexual behavior and a decline in HIV infection among young men in Thailand. N Engl J Med 1996; 335:297-303.
20. Potts M, McDevitt J. A use-effectiveness trial of spermicidally lubricated condoms. Contraception 1975;11:701-10.
21. Rietmeijer CAM, Krebs JW, Feorino PM, et al. Condoms as physical and chemical barriers against human immunodeficiency virus. JAMA 1988;259:1851-53.
22. Roddy RE, Cordero M, Ryan KA, et al. A randomized controlled trial comparing nonoxynol-9 lubricated condoms with silicone lubricated condoms for prophylaxis. Genitourin Med (in press).
23. Saracco A, Musicco M, Nicolosi A, et al. Man-to-woman sexual transmission of HIV: longitudinal study of 343 steady partners of infected men. J Acq Immun Def Syndr 1993;6:497-502.
24. Steiner M, Dominik R, Trussell J, et al. Measuring contraceptive effectiveness: a conceptual framework. Obstet Gynecol 1996;88:24s-30s.
25. Trussell J. Contraceptive efficacy. In: Hatcher RA, Trussell J, Stewart F, et al. Contraception Technology, 17th Revised Edition. New York NY: Irvington Publishers Inc. (in press).
26. Van de Perre P, Jacobs D, Sprecher-Goldberger S. The latex condom, an efficient barrier against sexual transmission of AIDS-related viruses. AIDS 1987;1:49-52.

Return to table of contents