publications

<<previous  |  index  |  next>>

VIII. Arguments For and Against Abstinence-Only Programs

Encouraging young people to delay sex and reduce the number of their sex partners forms a rational part of any comprehensive approach to HIV prevention.  However, governments have an obligation not to censor or distort information about effective methods of HIV prevention, including condoms, and to pursue HIV prevention strategies that are scientifically valid.  When moral considerations (such as discouraging sex for its own sake or promoting the institution of marriage) overwhelm sound HIV prevention, this impedes the realization of internationally recognized human rights, including the right to information, the right to the highest attainable standard of health, and ultimately the right to life.

The following sections address two aspects of whether abstinence-only-until-marriage programs constitutes a sound approach to HIV prevention that is consistent with internationally recognized human rights: first, whether abstinence messages in fact contributed to Uganda’s decline in HIV prevalence in the 1990s; and second, whether abstinence-only programs for young people have proven effective in the United States, where they have existed since 1981.

Distortion of Uganda’s HIV prevention efforts

Between 2002 and 2004, the U.S. government sponsored at least four studies which concluded that the drop in HIV prevalence in Uganda in the 1990s resulted from increased rates of abstinence and fidelity in Uganda during that period, as well as a concerted government effort to encourage these behavior changes.226  The aim of these studies was apparently to provide a scientific basis for current abstinence-until-marriage programs.  The most recent of these studies claims that Ugandan youth adopted at least twelve “protective behaviors” between 1989 and 2000, nine of which may be grouped under the category of abstinence or fidelity.227  The remaining three behavior changes relate to increased condom use, though the study notes that few national data are available on condom use before 1995.  The study does not attempt to ascertain the causes of various behavior changes (e.g., government-funded HIV prevention campaigns versus broader social factors), nor does it measure the relative impact of different behavior changes (e.g. abstinence versus condom use) on HIV spread.  It concludes that “[i]t is likely that a combination of abstinence and partner reduction resulted in the decline in prevalence, but that the increase in condom use helped maintain the low prevalence levels throughout the rest of the nineties.”228

There are multiple problems with using survey data such as these as the basis for U.S.-funded abstinence programs.  First, the U.S. government’s own research suggests that condoms played an important role in Uganda’s HIV decline, and not only for “high-risk” populations such as sex workers.229  According to the above study, the percentage of all sexually active Ugandan women and girls who had ever used a condom increased from 9 percent in 1989 to 26 percent in 1995, a period that saw a significant decline in HIV prevalence.  Among men and boys, the percentage rose from 22 percent to 35 percent.  While consistent condom use is difficult to measure, a helpful indication of the contribution of condom use to HIV prevention is the percentage of sexually active Ugandans who used a condom the last time they had sex with a non-regular partner.  National data for this indicator are available only for the period 1995 to 2000, during which the percentage rose from 25 percent to 44 percent among women and 40 percent to 62 percent among men.  As noted above, this increase in condom use coincided with a steep increase in non-regular sexual partnerships among young men, suggesting it staved off a significant number of new HIV infections.  Median HIV prevalence among ante-natal clinic attendees dropped from 11.8 percent to 5 percent during this period, though incidence (new HIV infections) likely dropped earlier.  National data on increased condom use are supported by studies in specific regions.230

Second, of the primary behavior changes documented in Uganda in the 1990s, partner reduction (or a reduction in casual sex) appears to have played a much larger role in HIV decline than abstinence.  One indication of this is that teenage pregnancy rates did not fall in Uganda during this period, and that teenage girls who became pregnant did not do so at older ages.  This suggests that any drop in HIV prevalence among girls could not have been due to girls’ postponing sex or becoming less sexually active, but instead to their having sex in more regular partnerships.231  Mathematical models have suggested that having multiple non-regular sex partners can dramatically increase HIV spread, more than having one regular partner after another (i.e., “serial monogamy”).  The fact that Uganda engaged in an intensive campaign in the 1990s to promote fidelity (known locally as “zero grazing”) further suggests that fidelity, not abstinence, was the most successful component of its HIV prevention efforts.

Third, demographic data on the causes of HIV decline in Uganda do not substitute for evaluations of abstinence-only programs.  Program evaluations require a comparison of the attitudes, intentions, and sexual behaviors of program participants over time, as well as in comparison to those who have not participated in abstinence-until-marriage programs.  Evaluations of this nature have been occurring in the United States since the 1990s and are reviewed below.  They indicate that abstinence-only programs have little to no impact on participants’ sexual risk-taking behaviors, and that participants are less likely to use condoms once they become sexually active.  If these are a guide, abstinence-only approaches would have been more likely to detract from Uganda’s HIV decline than to contribute to it.

Finally, the HIV prevention campaigns implemented by the Ugandan government in the 1990s, which enjoyed some success, differ vastly from abstinence education as defined and implemented by the United States.  Historical accounts, including those funded by the U.S. government, disclose numerous components of what has been described as the “Museveni” approach to HIV prevention.232  A hallmark of this approach was the president’s personal commitment to fighting AIDS, combined with his engagement of numerous government ministries, active encouragement of NGOs and faith-based organizations, and relaxation of state controls on mass media.  The openness of Uganda’s approach allowed a diversity of prevention messages (including the “zero grazing” message noted above) to permeate the country’s schools, churches, and airwaves.  Central to the effort was breaking down the stigma associated with HIV/AIDS and encouraging frank discussion of sex and other causes of HIV transmission.  As one veteran AIDS educator described it, “It’s not true that Museveni talked about abstinence.  What he did was give us complete freedom of the press.  There were pictures of vaginas and penises everywhere.”  A government minister added, “It was not easy [at first] because culturally we don’t talk about sexual matters openly.  The church didn’t want to talk about condoms.  Eventually, we managed to break through.  [Especially] once we explained the multiplicity of methods of acquiring AIDS, the stigma reduced.”233  It would be a revision of history to suggest that U.S.-funded abstinence-only programs, which were pioneered in the U.S. in 1981 as a means of pregnancy prevention and before HIV/AIDS was an epidemic, are a natural outgrowth of Uganda’s early anti-AIDS efforts.

Even the so-called ABC (Abstinence, Be Faithful, Condom use) approach to HIV prevention, which is routinely cited by U.S. officials and others as the “Ugandan approach,” does not accurately capture Uganda’s anti-AIDS effort before 2002.  In November 2004, the AIDS educator cited above told Human Rights Watch, “About one and a half years ago we started hearing about ABC, and we’d never heard of it before.  We were told this is what Uganda’s model was.”  Another educator, who had directed USAID-funded HIV prevention programs in Uganda since the early 1990s, said:

In about 1999 or 2000 . . . someone made a reference to ABC, and I had to ask what ABC was.  Although everyone says we were doing it in Uganda, I’d never heard of it.  I don’t even know where it came from.  A faith-based organization recently said that Janet Museveni had founded ABC, and I thought, you must be joking.  History has been substantially rewritten here.234

It is true that some Ugandan HIV/AIDS materials dating to the 1990s refer to “delayed sexual debut” as part of a comprehensive HIV prevention strategy; however, this does not amount to a national ABC approach, much less to abstinence-only-until-marriage as currently defined by the United States.235

Reverend Gideon Byamugisha, an Anglican priest from Uganda who is known as the first African cleric to reveal his HIV-positive status, said of Uganda’s alleged ABC strategy, “It makes me angry to hear that Uganda’s success is because of ABC.  It goes far beyond that.  It’s the amount of effort, information, attitudes changing, skills for self-protection, programming, VCT [voluntary counseling and testing], blood transfusions, training counselors and doctors . . . a supportive environment.  Uganda’s success is not an ‘either/or.’  Everything is important.”236

Ultimately, Uganda’s anti-AIDS efforts in the 1990s cannot be reduced to a particular government intervention such as abstinence-only or ABC.  As one commentator recently put it, “The government is but one player in the fight against HIV-1.237  There are hundreds of nongovernmental organizations, religious groups, and community activists also working to prevent the spread of HIV/AIDS in Uganda.” 238 This multiplicity of voices stands to be jeopardized by the government’s emerging focus on abstinence as an exclusive method of HIV prevention.  More importantly, the implication behind abstinence-only programs that AIDS is a “moral” disease stemming from “promiscuous” behavior is the antithesis of Uganda’s effort to de-stigmatize AIDS early in the pandemic.

Studies discrediting abstinence-only approaches in the U.S.

The exportation of abstinence-only programs from the United States to Uganda is occurring notwithstanding unrefuted evidence of the ineffectiveness and potential harms of these programs.  Government-funded evaluations in at least twelve U.S. states, as well as a federally mandated independent evaluation authorized in 1997, indicate that abstinence-until-marriage programs show no long-term success in delaying sexual initiation or reducing sexual risk-taking behaviors among program participants, and that program participants are less likely to use contraceptives once they become sexually active.239  The Institute of Medicine, a body of experts that acts under a Congressional charter as an advisor to the U.S. federal government, noted in 2001 that there was no evidence supporting abstinence-only programs, and that investing “millions of dollars of federal…funds…in abstinence-only programs with no evidence of effectiveness constitutes poor fiscal and health policy.”240  Assessments such as these provide some indication of the likely success (or failure) of abstinence-only programs in Uganda, as U.S.-based abstinence-only programs are administered according to the same guidelines, and in some cases by the same organizations, as proposed Ugandan programs.241  No independent evaluations of abstinence-only programs exist from Uganda, largely because such programs did not exist there on a significant scale before 2004.

Evaluations of abstinence-only programs typically measure whether program participants change their sexual attitudes, intentions, and behaviors over the short and long term.242  According to a 2004 review of abstinence-only program evaluations conducted by Washington, D.C.-based Advocates for Youth, only one program showed any impact on participants’ sexual behavior, and this impact disappeared by the end of the program.  While some programs had short-term impact on participants’ attitudes and intentions to abstain, and one (in Pennsylvania) had some long-term impact on intentions, these attitudes and intentions did not translate into behavior changes.243  In one county in Pennsylvania, 42 percent of female participants were sexually active by the second year of their abstinence-only program.  In another, rates of sexual debut among females increased from 6 to 30 percent as program participants progressed from seventh to ninth grade.  In Minnesota, where an abstinence program showed mixed results on changing attitudes towards abstinence in the long-term, the percentage of youths who were sexually active was higher in several counties with abstinence programs than the state average.244 

Of equal concern is that abstinence-only programs may discourage young people from using contraception once they become sexually active.  As noted above, abstinence-only programs do not provide participants with information about contraception other than (sometimes exaggerated) failure rates.  In one county in Pennsylvania, only half of those who said they started having sex in ninth grade used any form of contraception.  The Missouri evaluation found that program participants developed a less favorable attitude toward birth control from the beginning to the end of the program.  Virginity pledges, a staple of abstinence-only programs in which students pledge to remain sexually abstinent until marriage, have been shown in peer-reviewed national surveys of adolescent sexual behavior to reduce the likelihood of contraceptive use once pledgers become sexually active.245 

Proponents of abstinence-only programs often claim that teaching young people about condoms and safer sex will contradict or otherwise undermine the message of abstinence.  However, studies that compare abstinence-only education with programs that include factual information about contraception show the latter to be more effective on all counts.  A 2001 report analyzing studies of HIV prevention programs found that programs that include information about both abstinence and condoms can delay the onset of sex and increase condom use among sexually active teens.  The same study found no evidence existed that abstinence-only programs had an effect on sexual behavior.246  A 1998 study comparing a program that educated students about safer sex (including condom use) with an abstinence-only program found that both programs affected sexual behavior in the short term, but that the safer sex program was more effective at reducing unprotected sexual intercourse and frequency of intercourse in the long term.247 

In 2001, the Institute of Medicine concluded that scientific studies have shown that comprehensive sex and HIV/AIDS education programs and condom availability programs can be effective in reducing high-risk sexual behaviors.248  A 1997 report by the Joint United Nations Programme on HIV/AIDS (UNAIDS) found evidence that sexual health education for children and young people that included the promotion of condom use and safer sexual practices, did not increase participant’s sexual activity.249

In 2004, a “gold-standard” review of HIV prevention research by the Cochrane Collaborative Review Group on HIV infection and AIDS concluded that “[p]rograms promoting abstinence were found to be ineffective at increasing abstinent behavior and were possibly harmful; more rigorous research is needed to determine the effectiveness of abstinence programs on HIV risk.”250  A 2004 consensus statement in The Lancet signed by numerous experts in HIV prevention from around the world, stressed abstinence as a “first priority” for young people who are not sexually active but concluded:

For those young people who are sexually active, correct and consistent condom use should be supported.  Young people and others should be informed that correct and consistent condom use lowers the risk of HIV (by about 80-90% for reported “always use”) and of various sexually transmitted infections and pregnancy, and they should be cautioned about the consequences of inconsistent use.251

Officials of the U.S. government did not endorse the Lancet statement, though they were asked to do so.

U.S. officials systematically ignore independent evaluations of abstinence-only programs, instead making broad and unscientific claims about the benefits of abstinence.  The U.S. global AIDS strategy, for example, posits that “[d]elaying first sexual intercourse by even a year can have significant impact on the health and well-being of adolescents and on the progress of the epidemic in communities.”252  Beyond failing to cite evidence for this claim, the strategy neglects to mention that some countries with higher average ages of sexual debut than Uganda—Zimbabwe and South Africa, for example—have much higher rates of HIV incidence.  The important point is that delaying sex does not protect people from HIV unless they protect themselves once they become sexually active.  Abstinence-only programs in fact increase HIV risk by withholding information about contraception and safer sex and by suggesting that married people are safe from HIV infection.

As further “proof” of abstinence-only programs, proponents frequently cite evidence of reduced teen pregnancy rates in the United States in the 1980s and 1990s, a period that saw increased federal funding for abstinence-only programs.253  This logical fallacy assumes that just because abstinence-only programs occurred at the same time as a reduction in teen pregnancy, they must have caused this reduction.  Indeed, studies also show that contraceptive use increased during the same period, and (as noted above) that rates of premarital sex are higher in some regions with abstinence-only programs than in those without these programs.  The fact that participants in abstinence-only programs are less likely to use contraception once they become sexually active suggests that teen pregnancy rates might have dropped even further were it not for these programs.



[226] See Janice A. Hogle, ed., What Happened in Uganda?  Declining HIV Prevalence, Behavior Change, and the National Response, USAID Project Lessons Learned Case Study, September 2002; Edward C. Green, Rethinking AIDS Prevention: Learning from Success in Developing Countries (Westport, CT and London: Praeger, 2003); Uganda HIV/AIDS Partnership, Uganda Ministry of Health, Uganda AIDS Commission, and MEASURE Evaluation Project, AIDS in Africa during the Nineties: Uganda. Young People, Sex, and AIDS in Uganda Chapel Hill, NC: MEASURE Evaluation, Carolina Population Center, University of North Carolina at Chapel Hill, 2004); USAID Bureau for Global Health, Phase I Report of the ABC Study: Summary of HIV Prevalence and Sexual Behavior Findings, August 2003.

[227] The nine behaviors are premarital sex, age at first sex, sexual debut before age fifteen, age at first marriage, past year abstinence, past year secondary abstinence (i.e. abstinence among those already sexually active), non-marital partnerships, extramarital sex, and multiple partnerships.

[228] UHP/MOH/UAC/MEASURE, Young people, sex and AIDS in Uganda, p. 49.  Conclusions such as this are unfortunately highly politicized.  Studies not funded by the U.S. government have made more nuanced conclusions from the same data; see, e.g., Susheela Singh, Jacqueline E. Darroch, and Akinrinola Bankole, “A, B and C in Uganda: The Roles of Abstinence, Monogamy and Condom Use in HIV Decline” (December 2003), Alan Guttmacher Institute Occasional Report No. 9. While a comparative analysis of these studies is beyond the scope of this report, theimportant point is that the U.S. government’s own assessment does not support the establishment of abstinence-only programs as described in the U.S. Global AIDS Strategy.  

[229] As noted above, condoms are not part of abstinence education programs funded by the U.S. government, but rather are targeted at “high risk” populations.

[230] In Rakai District, for example, condom use increased from 2 percent to 66 percent of the sexually active population, a period that saw a decrease in HIV prevalence to 12 percent from 44 percent.  Sadab Kitatta Kaaya, “Rakai Condom Use Reaches 66 Percent,” The Monitor, January 27, 2005. See also, Roehr, "Abstinence programs do not reduce HIV prevalence in Uganda."

[231] Helen Epstein, “The Fidelity Fix,” The New York Times Magazine, June 13, 2004, p. 56.  The epidemiologists Rand Stoneburner and Daniel Low-Beer have also elaborated this argument in a series of articles.

[232] See, e.g., Emily Dyer, …And Banana Trees Provided the Shade: The Story of AIDS in Uganda (Kampala: Uganda AIDS Commission, September 2003); Helen Epstein, “AIDS: The Lesson of Uganda,” The New York Review of Books, July 5, 2001, pp. 18-23, Edward C.Green, Rethinking AIDS Prevention: Learning from Successes in Developing Countries, (Westport, CT and London: Praeger, 2003), pp. 141-226.

[233] Human Rights Watch interview with Miria Matembe, minister of ethics and integrity, Kampala, January 13, 2003.

[234] Human Rights Watch interview, November 10, 2004.

[235] In late 2004, the British medical the journal The Lancet published a consensus statement on HIV prevention which stated, “The ABC (Abstain, Be faithful/reduce partners, use Condoms) approach can play an important role in reducing the prevalence of HIV in a generalised epidemic, as occurred in Uganda.”  Daniel Halperin et. al., “The Time Has Come for Common Ground on Preventing Sexual Transmission of HIV,” The Lancet, Volume 364, Number 9449, November 27, 2004.  It is unfortunate that by using the term “ABC” in reference to Uganda, the statement contributed to the misperception that the U.S. AIDS Strategy, which uses the term ABC, follows the Ugandan approach.

[236] Dyer, Banana Trees, p. 18.

[237] HIV-1 is the predominant type of HIV in Uganda and most of the rest of the world.

[238] Parkhurst, "The Ugandan success story?"

[239] See Mathematica Policy Research Institute, Inc., The Evaluation of Abstinence Education Programs Funded Under Title V Section 510: Interim Report, p. 4, as well as state-level studies cited by Advocates for Youth, below.  A second federal report was completed in 2004 and submitted to the U.S. Congress and U.S. Department of Health and Human Services for review, but has yet to be released.

[240] Committee on HIV Prevention Strategies in the United States, Institute of Medicine, No Time to Lose: Getting More from HIV Prevention (Washington, D.C.: National Academy Press, 2001), pp. xi-xii and pp. 118-20.

[241] See “Uganda’s official AB policy,” above.

[242] Eleven state-funded evaluations of abstinence-only programs are reviewed in Debra Hauser, Five Years of Abstinence-Only-Until-Marriage Education: Assessing the Impact, (Washington, D.C.: Advocates for Youth, 2004).  A twelfth evaluation was released by the state of Texas in late 2004.  Patricia Goodson, B.E. (Buzz) Pruitt, Eric Buhi, Kelly L. Wilson, Catherine N. Rasberry, and Emily Gunnels, Abstinence Education Evaluation: Phase 5 Technical Report (Department of Health & Kinesiology, Texas A&M University, September 2004).  Information in this section is drawn principally from Advocates for Youth’s review.

[243] Edward Smith et al., Evaluation of the Pennsylvania Abstinence Education and Related Services Initiative: 1998-2002, Pennsylvania Department of Health, January 2003, pp. 1 and 21.

[244] Department of Health, Minnesota Education Now and Babies Later Evaluation Report 1998-2002, 2003.

[245] Peter Bearman and Hannah Brückner, "Promising the Future: Virginity Pledges as they Affect Transition to First Intercourse," American Journal of Sociology, vol. 106, no. 4 (2001), pp. 859-912; Bearman and Brückner, “After the Promise: the STD Consequences of Adolescent Virginity Pledges,” 2004, http://www.yale.edu/socdept/CIQLE/cira.ppt (retrieved November 10, 2004).

[246] Douglas Kirby, Emerging Answers:  Research Findings on Programs to Reduce Teen Pregnancy (Washington, D.C.:  National Campaign to Prevent Teen Pregnancy, 2001), pp. 5, 88-91; Jennifer Manlove, Angela Romano Papillio, and Erum Ikramullah, Not Yet: Programs to Delay First Sex Among Teens (Washington, D.C.: National Campaign to Prevent Teen Pregnancy, 2004).

[247] John B. Jemmott et al., “Abstinence and Safer Sex HIV Risk Reduction Interventions for African American Adolescents,” Journal of the American Medical Association, vol. 279, no. 19, May 20, 1998, pp. 1529-1536.

[248] Institute of Medicine, No Time to Lose.

[249] UNAIDS, Impact of HIV and Sexual Health Education on the Sexual Behavior of Young People: A Review Update (Geneva: UNAIDS, 1997), p. 15.

[250] The Cochrane Collaborative Review Group on HIV Infection and AIDS, “Evidence Assessment: Strategies for HIV/AIDS Prevention, Treatment and Care” (July 2004), University of California, San Francisco Institute for Global Health, executive summary; see also, pp. 4-8.

[251] “Comment: The time has come for common ground on preventing sexual transmission of HIV,” The Lancet, vol. 364 (November 27, 2004), p. 1913.

[252] OGAC, PEPFAR Five-Year Strategy, p. 24.

[253] See, e.g., “Abstinence Clearinghouse Points to Two Reports Supporting Abstinence Education,” The Christian Post, December 10, 2004.


<<previous  |  index  |  next>>March 2005