I'm sure you won't mind linking me to a legitimate study that affirms the 20% figure you mention. Frankly, that seems utterly implausible, even when factoring in human error.
I've seen this figure before, too. IIRC it's much lower, around single digits, with "perfect use". I think the 20% is with "typical use", meaning all the times people don't bother to use one or they use it incorrectly. For all I know, people in African countries use a condom once a year, thinking it's some sort of magical charm instead of understanding how it's really supposed to work. I guess that would make for a pretty high typical failure rate, too.
How effective are condoms in preventing pregnancy?
For adults, the failure rate is about 14% per year of use. That means every year about 1 in 7 condom users experience an unplanned pregnancy. For persons under the age of 18, condoms were found to have a failure rate of 18% over one year. For unmarried minorities, the condom failure rate is 36% per year, and for unmarried Hispanics, the failure rate is as high as 45% annually. Spermicidal condoms have not been proven more effective than the non-spermicidal type.
University of Texas, Medical Branch, Galveston, Texas
Press Release: June 7, 1993, Condoms Fail 31% of the time to prevent HIV
Contact: Susan Weller at 409-772-2618. published: June, 1993 in Social Science and Medicine, #36-120, analysis of data from 11 different studies
1995 American Journal of Epidemiology article, reported a Brazilian study of 204 heterosexual couples, each comprising of an HIV-positive man and an uninfected woman. Among couples who used condoms 100% of the time, 23% of the women were infected with HIV.
Fact Sheet on Condom Failure
1. "24 sets of condoms tested and all failed" and almost 71% failed "In respect of one or more of the physical requirements of the specification, notably freedom from pinholes." SABS report April 89.
2. "Spillage from condoms occurs as much as 65% to 75% of the time." Bjorklund and Gordon. Univ of Manitoba. Nov. 1990.
3. "The rubber comprising latex condom has intrinsic voids about 5 microns in size." The HIV virus is 0.1 micron. Roland, Rubber World. June 1993. Roland and Sobieski, Rubber Chemistry and Technology. Vol. 62, 1989.
4. Condoms reduce the risk of HIV infection by about 70% if they are used "consistently and correctly" IPPF (International Planned Parenthood Federation) Medical Bulletin Feb. 1997.
5. "It is not established whether the condom is as effective at preventing heterosexual transmission of HIV as it is for preventing conception." "The level of protection approximates 87%, with a range depending upon the incidence (of HIV) among condom nonusers. Thus the condom's efficacy at reducing heterosexual transmission may be comparable to or slightly lower than its effectiveness at preventing pregnancy." Family Planning Perspectives, 1999.
6. The failure rate for condoms in preventing pregnancy is 10%. K. Niswander. Manual of Obstetrics 1980.
7. The ISO standard for condoms allows 2 per 350 to be defective (about six defects per thousand.) (Tough luck if you happen to be one of those six)
8. "Increased condom use will increase the number of [HIV/AIDS] transmissions that result from condom failure" and "a vigorous condom promotion policy could increase rather than decrease unprotected sexual exposure if it has the unintended effect of encouraging a greater overall level of sexual activity." "Condoms and seat belts: the parallels and the lessons" The Lancet, 29 Jan 2000
9. In one test, 33% of latex condoms leaked HIV sized particles. Sexually Transmitted Diseases. vol.19. 1992
10. Ontario Ministry of Health campaign to promote condoms by means of televised AIDS messages made respondents more inclined to use condoms but less inclined to avoid casual sexual partners. Wilde, Target Risk, PDE Publications, 1994.
11. IPPF indicates that the risk of contracting AIDS during so-called "protected sex" approaches 100 percent as the number of episodes of sexual intercourse increases. Cates Medical Bulletin, IPPF 1997.
12. The only sure ways to avoid sexual transmission of diseases (including AIDS, chlamydia, genital herpes, genital warts, gonorrhoea, hepatitis B, and syphilis) are not to have sex at all or to limit sex to one uninfected partner who is also monogamous. Food and Drug Administrationc (USA) Consumer Magazine Sep 1990.