Correspondence
HIV Infection in Rural Florida
N Engl J Med 1993; 328:1351-1352May 6, 1993
- Article
To the Editor:
In their analysis of risk factors for infection with the human immunodeficiency virus (HIV) in pregnant women in rural Florida, Ellerbrock and colleagues (Dec. 10 issue)1 do not mention anal intercourse. I assume, from the information given in the Methods section, that questions about this sexual practice were asked, and I further assume that univariate analysis failed to show a significant association. The absence of an association is implied by their statement that “other known risk factors for HIV infection . . . were infrequently reported and not significantly associated with HIV infection.” Data, however, are preferable to default assumptions, especially in view of a recent report that associates anal intercourse with the sexual transmission to women (one third of whom were recruited in Florida) of a virus with similar epidemiologic characteristics2.
If none of the 11 HIV-infected women with no risk factors other than having had two to five sexual partners gave histories of anal intercourse, explicitly stating so would strengthen the case for penovaginal transmission unaided by cofactors such as genital infection.
2 ReferencesJohn J. Potterat
El Paso County Department of Health and Environment, Colorado Springs, CO 809101
Ellerbrock TV ,Lieb S ,Harrington PE , et al. Heterosexually transmitted human immunodeficiency virus infection among pregnant women in a rural Florida community. N Engl J Med 1992;327:1704-1709
Full Text | Web of Science | Medline2
Rosenblum L ,Darrow W ,Witte J , et al. Sexual practices in the transmission of hepatitis B virus and prevalence of hepatitis delta virus infection in female prostitutes in the United States. JAMA 1992;267:2477-2481
CrossRef | Web of Science | Medline
To the Editor:
The conclusion by the Centers for Disease Control and Prevention (CDC) and the Florida State Department of Health and Rehabilitative Services that pregnant women in rural Florida acquired HIV by heterosexual contact is not justified by the results of their seroepidemiologic survey1. Over 20 percent of the seropositive women in this study had no known high-risk partners, no history of drug use, and no positive tests for sexually transmitted diseases. These women were assumed to have been infected with HIV by heterosexual intercourse, even though no detailed investigation of their sexual contacts was performed. An important question -- What are the possible means of transmission of HIV and AIDS in persons without a recognized risk factor for the disease? -- was not answered. Previous studies by the CDC and the Department of Health and Rehabilitative Services claimed that AIDS in this region could be explained by “sex and dirty needles,” with no relation to poverty or environmental factors2. A prospective study showing seroconversion to HIV after sexual contact with an infected person might not exclude the possibility of shared environmental exposure. When I worked as a consulting physician for the Palm Beach County Health Department in Belle Glade from 1984 through 1987, it was demonstrated by putting pins in a map that virtually all persons with AIDS (and persons with active tuberculosis) lived in the same economically depressed neighborhoods3. The extraordinarily high prevalence of disease in the same slums today represents a public health emergency. Squalid living conditions contribute to high rates of infection and progression of disease in persons with defective immune systems. Despite the conclusions of the CDC and the Department of Health and Rehabilitative Services about the transmission of HIV and AIDS in rural Palm Beach County, it is apparent that the epidemic of AIDS will continue unabated unless education is combined with public health measures and environmental improvements.
3 ReferencesMark E. Whiteside, M.D., M.P.H.
Old Town Medical Center, Key West, FL 330401
Ellerbrock TV ,Lieb S ,Harrington PE , et al. Heterosexually transmitted human immunodeficiency virus infection among pregnant women in a rural Florida community. N Engl J Med 1992;327:1704-1709
Full Text | Web of Science | Medline2
Castro KG ,Lieb S ,Jaffe HW , et al. Transmission of HIV in Belle Glade, Florida: lessons for other communities in the United States. Science 1988;239:193-197
CrossRef | Web of Science | Medline3
Whiteside ME, Withum D, Tavris D, et al. Outbreak of no identifiable risk AIDS in Belle Glade, Florida. Presented at the First International Conference on AIDS, Atlanta, April 14-17, 1985. abstract.
To the Editor:
Ellerbrock et al. note that “one fifth . . . of the HIV-infected women in this study had sexual histories that were unremarkable as compared with those of other groups of women in the United States” and that “52 percent [of college women in another study have] had two to five sexual partners, and 21 percent reported six or more.” However, any number of multiple partners requires attention to the woeful arithmetic of sexual networks.
If each actual partner is conservatively assumed to have had the same number of partners as the respondent, then the total number of partners -- actual partners plus “phantom” partners -- can be calculated by the equation T = 2N - 1, where N is the number of actual partners. With the use of this formula for 2, 3, 4, 5, and 6 actual partners, the total number of potentially infective sources (actual plus phantom partners) soars upward to 3, 7, 15, 31, and 63, respectively.
Our just-completed regional survey (which involved face-to-face interviews and was supported by the National Science Foundation) of 180 adults, ranging from 18 to 59 years of age, measured respondents' perceptions of “phantom sex partners.” They failed to understand the exponential relation between sexual promiscuity and the risk of AIDS, even when we provided unlimited time for calculation and explicit instructions -- i.e., “to actual partners add the partners of those actual partners as well as the partners of their partners.” For 3, 6, and 9 actual partners, the median estimates of the number of total partners were 12, 42, and 90, respectively, whereas the correct values (according to the formula) are 7, 63, and 511. The gross underestimation of the magnitude of the sexual networks in our sample was unaffected by age, sex, or the number of sexual partners.
Laura A. Brannon, Timothy C. Brock, Timothy C. Brock, Ph.D.
Ohio State University, Columbus, OH 43210Author/Editor Response
The authors reply:
To the Editor: As Mr. Potterat's letter and several recent studies suggest, anal intercourse may increase the risk of male-to-female sexual transmission of HIV1,2. In our study, we initially included questions about anal intercourse, but decided to remove them because some prenatal patients had been either confused or embarrassed by the questions during field testing. However, in an ongoing study of female crack-cocaine users among patients attending the same prenatal clinic, about one fourth of these women reported having had anal intercourse before ever using crack cocaine, suggesting that this behavior may contribute to the heterosexual transmission of HIV in the community.
Dr. Whiteside states that the conclusion of our study is not justified. However, we consider the epidemiologic evidence overwhelming that most of the women in our study acquired HIV through heterosexual contact. For instance, only 4 of 1009 women (0.4 percent) reported ever injecting drugs, and those 4 were HIV-seronegative. In addition, four of the five independent predictors of HIV infection among study participants -- crack-cocaine use, history of multiple sexual partners, sexual intercourse with a high-risk partner, and positive serologic test for syphilis -- suggest heterosexual transmission of HIV infection.
Dr. Whiteside also seems to suggest that environmental factors contribute to the transmission of HIV in the community. However, a population-based, seroepidemiologic study designed to investigate possible environmental risk factors, especially the spread of disease by insects, was conducted in this community in 1986, and it found no evidence of HIV transmission through environmental exposure3. In addition, numerous studies examining household contacts of HIV-infected persons have found no evidence of nonpercutaneous, nonsexual transmission of HIV4. Current studies continue to indicate that the transmission of HIV is the result of specific behavior and that avoiding such behavior can prevent transmission.
4 ReferencesTedd V. Ellerbrock, M.D.
Centers for Disease Control and Prevention, Atlanta, GA 30333Sandra A. Schoenfisch, R.N., Ph.D.
Florida State Department of Health and Rehabilitative Services, Tallahassee, FL 32399Margaret J. Oxtoby, M.D.
Centers for Disease Control and Prevention, Atlanta, GA 303331
European Study Group on Heterosexual Transmission of HIV
CrossRef | Web of Science2
Lazzarin A, Saracco A, Musicco M, Nicolosi A, Italian Study Group on HIV Heterosexual Transmission
CrossRef | Web of Science | Medline3
Castro KG ,Lieb S ,Jaffe HW , et al. Transmission of HIV in Belle Glade, Florida: lessons for other communities in the United States. Science 1988;239:193-197
CrossRef | Web of Science | Medline4
Gershon RRM ,Vlahov D ,Nelson KE . The risk of transmission of HIV-1 through non-percutaneous, non-sexual modes -- a review. AIDS 1990;4:645-650
CrossRef | Web of Science | Medline
- Citing Articles (3)
Citing Articles
1
Stuart Brody. (2010) The Relative Health Benefits of Different Sexual Activities. The Journal of Sexual Medicine 7:4pt1, 1336-1361
CrossRef2
Laura A. Brannon, Amy E. McCabe. (2003) Schema-Derived Persuasion and Perception of AIDS Risk. Health Marketing Quarterly 20:2, 31-48
CrossRef3
John J. Potterat, Stuart Brody. (2000) More of the Same is Not Validation. Sexually Transmitted Diseases 27:1, 60-61
CrossRef
