Correspondence

Decline in HIV Infections in Thailand

N Engl J Med 1996; 335:1998-1999December 26, 1996

Article

To the Editor:

Nelson et al. (Aug. 1 issue)1 report that seropositivity for the human immunodeficiency virus (HIV) declined among Thai military conscripts from a range of 10.4 to 12.5 percent in the period 1991 to 1993 to 6.7 percent in 1995, and they attribute the decline to the increased use of condoms. The authors rely on the post hoc ergo propter hoc fallacy, attributing causation on the basis of the temporal sequence (of a single pair of events). Among Ugandan men of comparable age, the prevalence of HIV declined from 11.8 percent in 1989 to 2.7 percent in 1994, in the absence of either a major program of condom distribution or a behavioral intervention.2 Thus, the decline in seroprevalence in Thailand might best be attributed to the natural course of epidemics rather than to the distribution of condoms.3 Table 3 of Nelson et al. shows that the use of condoms with prostitutes was not significantly related to protection from HIV in 1993 or 1995 and was associated with an increased risk in the 1991 cohort.

As in most studies of risk factors for HIV infection, no attempt was made to verify the self-reports of the soldiers, even though there is frequent misreporting by respondents in such investigations 4 and other reasons for the misclassification of risk.5 One trend that may be observed in Table 4 of the article is a cohort effect of increasing risk of infection attributable to injection-drug use and homosexual contact and a decreasing risk attributable to heterosexual prostitution. This may be due to seropositive persons' becoming more forthcoming about their more serious risk factors.

Stuart Brody, Ph.D.
University of Tübingen, 72074 Tübingen, Germany

5 References

1
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
Full Text | Web of Science | Medline

2
Mulder D, Nunn A, Kamali A, Kengeya-Kayondo J. Decreasing HIV-1 seroprevalence in young adults in a rural Ugandan cohort. BMJ 1995;311:833-836
CrossRef | Web of Science | Medline

3
Brody S. Incidence of HIV infection decreases because of nature of epidemics. BMJ 1996;312:125-125
CrossRef | Web of Science | Medline

4
Brody S. Patients misrepresenting their risk factors for AIDS. Int J STD AIDS 1995;6:392-398
Web of Science | Medline

5
Brody S. Risk factors for HIV-1 seroconversion may not be what they seem. JAMA 1996;275:1543-1543
CrossRef | Web of Science | Medline

To the Editor:

Although condom use was probably responsible for some of the decline in HIV seroconversion reported by Nelson et al., it is not the only factor. The reported data include a decline in the proportion of men who had any sexual contact (from 92.3 percent to 87.2 percent) and a marked decline in the proportion who had contact with sex workers (from 81.5 percent to 63.8 percent). These men did not engage in high-risk behavior and did not become infected with HIV. It is barely mentioned that this decrease in high-risk behavior undoubtedly contributed to the decline in the reported rates of HIV seroconversion, possibly more than did the use of condoms.

Perhaps the condom campaign in Thailand raised awareness of the risk of acquiring HIV through contact with sex workers. This increased awareness may have raised doubts about the safety of such contacts in enough people that they chose not to have sexual contact with sex workers and thus did not become infected. It would be interesting to compare the results of the condom campaign with those of a campaign promoting abstinence and monogamy with regard to the decline in HIV seroconversion.

Nancy P. Lawless, M.D.
Washington, DC 20307

Author/Editor Response

The authors reply:

To the Editor: Brody believes that the recent decline in HIV-infection rates we found among 21-year-old men in Thailand was not related to changes in their high-risk sexual behavior. He cites a report from Uganda1 that found a reduction in the prevalence of HIV among men between the ages of 13 and 24 in several cross-sectional surveys between 1989 and 1994; the overall prevalence of HIV in the population and among young women did not change. Since that report contained no information on the frequency of high-risk behavior in this population, it is impossible to evaluate Brody's speculation that the apparent reduction in the prevalence of HIV among young men was not due to changes in their high-risk sexual behavior but rather to decreased infectivity of the virus. Others have suggested that the epidemic of HIV infection and AIDS might disappear spontaneously, without intervention.2 However, these unrealistic predictions, which ignore the epidemiology of HIV transmission, have not been confirmed either in the United States3 or in developing countries.4 Although the data we reported were not from a controlled clinical trial, we believe the public health intervention played an important part in the recent decline in rates of HIV infection among young men in Thailand.

Although it is neither ethical nor practical to verify a person's report of condom use or other sexual behavior, we have independent data suggesting that the use of condoms during commercial sexual relations has increased and is now the rule in northern Thailand. We confirmed the high rates of condom use claimed by the men in our study by interviewing female sex workers in northern Thailand. In a study of condom breakage during commercial sexual relations, the high frequency of reported condom use (over 90 percent) was confirmed by collecting used condoms.5

We agree with Lawless that the decline in HIV-infection rates was probably due not only to increased condom use but also to reductions in the frequency of high-risk sexual relations by the young men in our study. The public health program to prevent HIV infection in Thailand not only promoted condom use but also aimed at reducing the frequency of high-risk sexual relations, including commercial sex. However, we suspect that if the promotion of condom use had not been included in the intervention, the rates of HIV infection would not have declined nearly so quickly.

Kenrad E. Nelson, M.D.
David D. Celentano, Sc.D.
Johns Hopkins University, Baltimore, MD 21205

5 References

1
Mulder D, Nunn A, Kamali A, Kengeya-Kayondo J. Decreasing HIV-1 seroprevalence in young adults in a rural Ugandan cohort. BMJ 1995;311:833-836
CrossRef | Web of Science | Medline

2
Bregman DJ, Langmuir AD. Farr's law applied to AIDS projections. JAMA 1990;263:1522-1525
CrossRef | Web of Science | Medline

3
Karon JM, Rosenberg PS, McQuillan G, Khare M, Gwinn M, Peterson LR. Prevalence of HIV infection in the United States, 1984 to 1992. JAMA 1996;276:126-131
CrossRef | Web of Science | Medline

4
Kimball AM, Berkley S, Ngugi E, Gayle H. International aspects of the AIDS/HIV epidemic. Annu Rev Public Health 1995;16:253-282
CrossRef | Web of Science | Medline

5
Rugpao S, Wongchak T, Beyrer C, Khamboonruang C, Celentano DD, Nelson KE. Frequency of condom use and breakage and slippage in commercial sex in Northern Thailand. Presented at the 11th International Conference on AIDS, Vancouver, Canada, July 7–12, 1996. abstract.

Citing Articles (1)