HERPES SIMPLEX VIRUS TYPE 2 (HSV-2) is one of the most common sexually transmitted diseases (STDs). Approximately 25% of the U.S. population is seropositive by 30 years of age. 1 Genital HSV-2 infection is predominantly transmitted sexually from partners with asymptomatic or unrecognized disease. 2 Therefore, HSV-2 seroprevalence data provide a better epidemiologic measure of the population burden of infection than clinical data. Herpes simplex virus type 2 has received renewed public health attention because of a reported HSV-2 seropositivity increase in the United States from 1976 to 1994 (30%) 1 and reports of similar increases in other countries. 3,4 Because HSV-2–associated ulceration facilitates HIV transmission, 5,6 HSV-2 seroprevalence data may be useful in determining specific subgroups of a population at high HIV risk. 7 However, in most less-developed countries, data on HSV-2 type-specific prevalence and trends are sparse or nonexistent.
Previous epidemiologic studies of HSV-2 infection have been hampered by the nonspecificity of serologic assays, which do not clearly differentiate HSV-1 and HSV-2 infections because of common antigen reactivity. 8 Recently, HSV-2 type-specific serologic assays have been developed. 9 Herpes simplex virus type 1 infection, characterized by oral–labial lesions, is generally transmitted nonsexually. 10 Detection of type-specific HSV-2 antibodies almost exclusively indicates a genital infection because HSV-2 oral infection rarely is seen in the absence of genital infection, and HSV-2 seropositivity in adults from neonatal or nosocomial HSV-2 infections is rare. 3
Data comparing type-specific HSV-2 seroprevalence and risk factors between comparable population groups from different countries are largely unavailable. This article examines the prevalence and risk factors for the detection of type-specific HSV-2 serum antibodies among middle-age women in São Paulo, Brazil, and Manila, the Philippines.
This analysis is based on cross-sectional data from 552 women who participated as hospital control subjects in two case-control studies of cervical cancer in São Paulo, Brazil, 11 and Manila, the Philippines. 12 Both studies are part of the International Agency for Research on Cancer’s (IARC’s) multicenter cervical cancer case–control study. The two studies used an identical protocol and questionnaire for recruitment and data collection, and central laboratories for the detection of sexually transmitted agents.
The fieldwork was conducted from June 1990 to June 1991 in Brazil and from April 1991 to April 1993 in the Philippines. Women between 18 and 80 years of age were invited to participate and selected to match the expected age distribution of cervical cancer case subjects. In Brazil, women were enrolled at outpatient clinics and inpatient wards of five public hospitals in the city of São Paulo. In the Philippines, women were identified at outpatient clinics (surgical and medical wards) of the Philippines General Hospital in Manila. Women were ineligible to participate if they had received any previous cervical cancer treatment or a diagnosis of any disease sharing risk factors with cervical cancer including cardiovascular or cerebrovascular disease, chronic bronchitis, emphysema, or neoplasia of the breast, reproductive or respiratory organs, anus, oral cavity, esophagus, bladder, or liver.
Interviews were conducted by specially trained interviewers using a standardized questionnaire that included sociodemographic characteristics, reproductive and contraceptive history, and sexual behavior. Testing to detect human papillomavirus (HPV) DNA was performed on cervical exfoliated cells using polymerase chain reaction (PCR) assays. Because this case–control study was designed to investigate HSV-2 as a cofactor of HPV in the etiology of invasive cervical cancer, control participants having both HSV-2 serology and HPV DNA results were included in this analysis. Protocols were cleared by the local and IARC ethical review committees.
In Brazil, 238 women were eligible for investigation, and 225 (94.5%) agreed to be interviewed. Altogether, 181 women (76.1%) with adequate sera for HSV-2 serology testing and HPV test results were included in this analysis. The main reason for exclusion was unavailable HPV results because of β-globin negativity. Participating and nonparticipating women had similar sociodemographic and sexual behavior characteristics (data not shown). Six women were excluded because of a symptomatic STD diagnosis at admission. The main diagnostic categories of the Brazilian women were diseases of the circulatory system (21.5%), infectious and parasitic diseases (12.7%), diseases of the digestive tract (12.2%), neoplasms (9.4%), diseases of the nervous system (8.8%), and endocrine diseases (8.3%).
In the Philippines, 392 women were eligible for investigation, and 387 (98.7%) agreed to participate. A total of 371 women (94.6%) with HPV test results and adequate sera for HSV-2 serology testing were included in this analysis. The main diagnostic categories of the Filipino participants were urinary tract infections (19.4%), benign disorders of the genital tract (14.3%), menstrual disorders (11.3%), diseases of the circulatory system (8.1%), and mild mental disorders (7.3%). Also included in this study were 55 healthy women accompanying outpatients (14.8%), whose distribution of various risk factors was similar to that of the other Filipino participants. Current “husbands” of the participating women, selected as previously described for Colombia 13 and in Spain, 14 were interviewed in both countries to determine the sexual behavior of a woman’s male partner. In brief, husbands were defined as men who had been engaged in regular sexual intercourse with the women in the study for at least 6 months, irrespective of a formal marriage or a common house. Husbands of 37.6% of the participating women from Brazil (n = 68) and 28.6% of the participating women from the Philippines (n = 106) were interviewed by specially trained male interviewers.
Participating women were asked to provide 10 ml of blood for tests to identify sexually transmitted agents. Blood samples were processed by centrifugation at the site of collection. The separated serum was frozen at −20 °C for shipment to IARC for storage.
Herpes Simplex Virus Type 2 Serology Testing
Sera were screened initially for HSV-2 immunoglobulin G (IgG) antibodies using the Gull HSV-2 gG enzyme-linked immunoassay (ELISA), which has an estimated sensitivity of 96% to 98% and an estimated specificity of 96% to 97%. 9 The Gull HSV-2 ELISA (Gull Laboratories, Salt Lake City, UT) has since been replaced with the Meridian HSV-2 ELISA (Meridian Laboratories, Cincinnati, OH) because of a change in company management.
For quality control, a validation study was performed in which 39 samples were initially tested using ELISA assay with Western blot confirmation of equivocal and borderline negative results. On the basis of these results, the Western blot was used to retest 20 HSV-2–positive and 19 HSV-2–negative sera. The Gull ELISA with Western blot confirmation of all HSV-2 negative and equivocal sera had a sensitivity of 100% and a specificity of 90.1%, as compared with the Western blot results, which were considered the reference gold standard. 15 To improve the specificity of this testing approach, all Gull HSV-2–positive sera were retested with the Western blot assay for definitive serostatus. Thus, the final HSV-2 serologic results were based on an initial screen of the sera with the Gull HSV-2 gG ELISA, and all positive, equivocal, and borderline negative ELISA results were retested with the Western blot assay.
The statistical significance of differences in dichotomous variables was tested using a chi-square test. Odds ratios and 95% CIs were calculated as approximations of relative risks by unconditional, multiple logistic regression (STATA 5.0, Stata Press, College Station, TX), adjusted for age categorized into four groups: younger than 40, 40 to 49, 50 to 59, and 60 years or older. The statistical significance of trends for odds ratios was assessed by considering the categorical variable as a continuous variable in the logistic model. Standard methods were used to calculate HSV-2–attributable fractions (AFs):
AF = p (r − 1)/(1 + p(r − 1)>,
where p is the proportion of the population with a given exposure, approximated by the prevalence among women surveyed, and r is the relative risk, approximated by the odds ratio.
The prevalence of HSV-2 IgG serum antibodies was significantly higher in Brazil (42%; 95% CI, 34.7–49.2) than in the Philippines (9.2%; 95% CI, 6.2–12.1). The mean age of the 181 participating women in Brazil (52.4 years) was significantly higher than that of the 371 women in the Philippines (46.6 years;P < 0.001). Although the diagnostic categories represented by these two groups of women differed, no significant difference in HSV-2 seropositivity was seen in any specific diagnostic category (data not shown).
Table 1 shows that no clear association between HSV-2 seropositivity and age was found in either country. In the age-adjusted analyses, HSV-2 seropositivity increased with decreasing age at first intercourse, particularly in the Philippines, where women who started sexual activity before 17 years of age were approximately seven times more likely to be HSV-2 positive than those initiating sexual activity at 21 years of age or older. In both countries, HSV-2 seroprevalence was higher among women reporting more than one lifetime sexual partner. Within each category of age at first intercourse in Brazil, HSV-2 seropositivity was significantly higher among women with two or more lifetime sexual partners (data not shown).
Age-adjusted risk factors for HSV-2 positivity in Brazil (Table 2), but not in the Philippines, included a history of casual partners, no reported condom use, a history of smoking for more than 20 years, and a report of a husband who occasionally or frequently had other sexual partners. In the Philippines, long-term hormonal contraceptive use (≥4 years) was a significant HSV-2 risk factor, but not in Brazil. In the Philippines, women had a higher HSV-2 seroprevalence if they were uncertain whether their husbands had other sexual partners, or if they were cohabiting with their husbands. In neither country was HSV-2 seropositivity significantly associated with residence, parity, or HPV DNA positivity. No participants reported a history of genital herpes. Other factors examined that were not statistically significant in either country included educational status, a history of a cesarean section, anal intercourse, number of pregnancies, abnormal gynecologic diagnosis, total number of household amenities, and a history of vaginal douching, hormonal contraceptive use, smoking, or a miscarriage (data not shown).
In the multivariate analyses (Table 3), younger age at first intercourse was the only factor significantly associated with HSV-2 seropositivity among women in both Brazil and the Philippines. More than one lifetime sexual partner, a report of a husband who occasionally or frequently had other sexual partners, urban or semi-urban residence, and a report that condoms had never been used were significant multivariate risk factors for HSV-2 seropositivity in Brazil, but not in the Philippines. In the Philippines, women who reported long-term hormonal contraceptive use or uncertainty about whether their partner had other sexual partners showed a significantly higher risk for HSV-2 seropositivity. A nonsignificant increase in HSV-2 risk also was observed in the Philippines among women who reported having had two or more lifetime sexual partners.
To determine whether the higher HSV-2 seroprevalence in Brazil than in the Philippines reflects differences in the prevalence of factors associated with HSV-2 seropositivity, stratified analyses were used to compare sociodemographic and sexual behavior characteristics of the participants (Table 4). Women from Brazil had an older average age and a lower median age at first sexual intercourse. Brazilian women more frequently reported having had three or more sexual partners (13.8%) than Filipino women (1.9%). More Brazilian than Filipino women reported a history of casual sexual partners, using hormonal contraceptives for more than 4 years, having smoked for 20 or more years, and cohabiting with their husbands.
Because of the notable differences in both HSV-2 seroprevalence and distribution of risk factors among Brazilian and Filipino women, HSV-2–attributable fractions were calculated for risk factors associated with selected sexual behaviors. 16 In 40.3% of the HSV-2–seropositive cases in Brazil and 3.7% of the cases in the Philippines, HSV-2 seropositivity can be attributed to having three or more lifetime sexual partners. In 29.8% of the HSV-2 cases in Brazil, and 9.1% of the cases in the Philippines, HSV-2 seropositivity can be attributed to a husband who has other sexual partners occasionally or frequently.
To determine whether the sexual behavior of males contributes to HSV-2 seropositivity among their female partners, the behavior characteristics among husbands of the participating women were investigated (Table 5). The husbands of Brazilian women generally reported a history of more high-risk sexual behavior than that reported by the Filipino husbands. The Brazilian men surveyed also reported a higher prevalence of cohabiting with their wives and of never being circumcised than Filipino men. The odds ratios of HSV-2 seropositivity were significantly higher among Brazilian women with male partners who reported a history of two or more lifetime sexual partners, anal intercourse, or cohabiting with their partners. Filipino women with husbands who reported a history of anal intercourse had a significantly increased risk of HSV-2 seropositivity, as did those whose husbands reported previous contact with prostitutes, although this increase in HSV-2 risk was not statistically significant. Other factors examined that were not statistically significant included husband’s age, age at first intercourse, education attainment, and smoking duration.
HSV-2 seroprevalence was found to be four times greater among middle-age women in São Paulo, Brazil (42%), than in Manila, the Philippines (9.2%). This prevalence of type-specific HSV-2 antibodies among Brazilian women with a mean age of 52.4 years is similar to the 39% to 42% range among women with mean ages of 39 to 47 years from studies in Costa Rica, 17,18 and among control subjects in a multicenter cervical cancer study in Colombia, Costa Rica, Mexico, and Panama, 19 yet lower than the 60% among control subjects (mean age, 48 years) in a cervical cancer study in Colombia. 20 This 42% HSV-2 seroprevalence found in São Paulo is the same as that among low-income, 27-year-old women in São Paulo (42%), 21 yet higher than that among similar women with a higher income (31%),21 and among younger women from Campinas, Brazil (23%). 22
In Asian countries, the 9.2% HSV-2 seroprevalence among Filipino women with a mean age of 46.6 years in the current study is consistent with the low levels of HSV-2 seropositivity of 7% and 14% among pregnant women in Japan and Taiwan, respectively, 4 and among women with a mean age of 49 years in Osaka, Japan (15%), 23 yet lower than the 29% among control subjects with a mean age of 52 years in a cervical cancer study in Sichuan Province, China. 24
Despite the use of different HSV-2 type-specific serologic assays in these various studies, these data are consistent with the notion that population-based HSV-2 seropositivity is greater in South and Central America than in Asia. The current data are based on women selected from public hospitals and clinic centers. The participating women were representative of the hospital populations from which they were recruited, but may not have been representative of a population-based sample of middle-age women.
The higher HSV-2 seroprevalence among Brazilian women is consistent with the higher incidence of invasive cervical cancer in Brazil (31/100,000 women) than in the Philippines (22/100,000 women). 25 The lack of an association between HSV-2 seropositivity and HPV DNA positivity may result from the different ascertainment measurements of these markers, with HSV-2 antibodies representing a cumulative measure of exposure and HPV DNA indicating chronic or current genital HPV infection. The estimated adult rate of HIV in 1997 is seven times higher in Brazil (0.43%) than in the Philippines (0.06%), 26,27 strengthening the notion HSV-2 seropositivity may be a useful indicator of HIV risk or an actual risk factor for HIV transmission.
Age was not a significant HSV-2 risk factor in either country, probably because 77.5% of the participants were older than 40 years. Other studies have shown that HSV-2 seroprevalence may reach a plateau or tend to decrease at older ages. 1,28 Earlier age at first intercourse was a significant HSV-2 risk factor in both countries, which is consistent with some studies 17,29 but not others. 30,31
Other markers of sexual behavior (more than one lifetime sexual partner and a report of a husband with other sexual partners) were significant multivariate HSV-2 risk factors in Brazil, but not in the Philippines. These results contrast with findings suggesting that in areas of high STD prevalence, traditional sexual behavior risk factors for HSV-2 may be obscured because of the higher levels of exposure with each sexual encounter. 29,32
In the Philippines, the lack of statistical significance between HSV-2 seropositivity and the total number of sexual partners is probably explained by the large variance of these estimates, as related to the low overall HSV-2 seropositivity (9.2%) and the small proportion of women who reported more than one sexual partner (10.3%). Nondifferential underreporting of sexual behavior, possibly because of cultural differences in reporting between Filipino and Brazilian women, may have attenuated the true measures of association. However, Filipino women reported less high-risk sexual behavior, consistent with their lower HSV-2 seropositivity than that among Brazilian women. Because of the lower HSV-2 seroprevalence in the Philippines, it is possible that each additional sexual partner may not incur an increased risk of acquiring HSV-2 infection among Filipino women.
The results from this study indicate that Brazilian women had a higher HSV-2 seropositivity if they reported having a husband with other sexual partners. This is consistent with findings that having concurrent sexual partners or partner networks increases the risk for STD and HIV. 33 This finding may reflect the higher HSV-2 seroprevalence in Brazil, indirectly indicating that the risk of HSV-2 acquisition is dependent not only on whether a woman’s husband has other sexual partners, but also on the likelihood that he will acquire HSV-2 infection from these sexual partners.
Living in an urban or semi-urban residence was a significant HSV-2 risk factor in Brazil, but not in the Philippines. Urban or semi-urban and rural residents in Brazil had a similar prevalence of high-risk sexual behavior (data not shown). In Brazil, urban residence may be a determinant of HSV-2 risk because of the higher HSV-2 prevalence in urban centers, or may be a proxy of other risk behaviors that more directly increase the risk of HSV-2 seropositivity (e.g., travel). 34
Data obtained from husbands of the female participants indicate that Brazilian men reported a higher prevalence of high-risk sexual behavior than the Filipino men surveyed. Women from both countries were more likely to be HSV-2 seropositive if their husband reported a history of anal intercourse. It is noteworthy that clinically confirmed circumcision was highly prevalent (97%) among the Filipino men and comparatively rare (8%) among the Brazilian men interviewed, particularly because of recent data indicating that uncircumcised men have an increased risk of HIV. 35 The investigation to determine whether circumcision may offer protection against male-to-female HSV-2 transmission was limited because of the small sample size.
The higher HSV-2 seroprevalence in Brazil than in the Philippines appears to be largely because of differences in women’s reported histories of their sexual behavior and that of their husbands. The current data indicate that STD risk may differ by geographic region, likely depending on regional differences in sexual norms and population-based STD prevalence. A given sexual behavior may confer a different risk of HSV-2 acquisition based on the HSV-2 prevalence in a population. Public health interventions to reduce the incidence of HSV-2 and other STDs among women should be tailored to the specific population and implemented to reduce high-risk sexual behavior among men, particularly in areas such as Brazil. As safe and effective HSV-2 vaccines and microbicides become available, data on the population prevalence of HSV-2 will be an important guide for future prevention efforts.
Using recently developed type-specific HSV-2 serologic assays, these data further understanding concerning the magnitude of endemic HSV-2 and HSV-2 risk factors in Brazil and the Philippines. Given the improvement in HSV-2 detection and characterization, previous studies of HSV-2 should be reviewed and may need to be reassessed using updated serologic assays. The current results are consistent with observations that HSV-2 seroprevalence data may be used as a marker of past sexual behavior in epidemiologic studies, 30 further demonstrating their relevance for the direct comparison of different population groups.
1. Fleming DT, McQuillan GM, Johnson RE, et al. Herpes simplex virus type 2 in the United States, 1976 to 1994. N Engl J Med 1997; 337: 1105–1111.
2. Corey L. The current trend in genital herpes: progress in prevention. Sex Transm Dis 1994; 21: S38–S44.
3. Nahmias AJ, Lee FK, Beckman-Nahmias S. Sero-epidemiological and -sociological patterns of herpes simplex virus infection in the world. Scand J Infect Dis Suppl 1990; 69: 19–36.
4. Nahmias AJ, Lee FK, Keyserling HL. The epidemiology of genital herpes. In: Stanberry LR, ed. Genital and Neonatal Herpes. New York: John Wiley and Sons, 1996: 93–108.
5. Dickerson MC, Johnston J, Delea TE, White A, Andrews E. The causal role for genital ulcer disease as a risk factor for transmission of human immunodeficiency virus: an application of the Bradford Hill criteria. Sex Transm Dis 1996; 23: 429–440.
6. Severson JL, Tyring SK. Relation between herpes simplex viruses and human immunodeficiency virus infections. Arch Dermatol 1999; 135: 1393–1397.
7. Dobbins JG, Mastro TD, Nopkesorn T, et al. Herpes in the time of AIDS: a comparison of the epidemiology of HIV-1 and HSV-2 in young men in northern Thailand. Sex Transm Dis 1999; 26: 67–74.
8. Ashley R, Cent A, Maggs V, Nahmias A, Corey L. Inability of enzyme immunoassays to discriminate between infections with herpes simplex virus types 1 and 2. Ann Intern Med 1991; 115: 520–526.
9. Ashley RL, Wald A. Genital herpes: review of the epidemic and potential use of type-specific serology. Clin Microbiol Rev 1999; 12: 1–8.
10. Corey L, Wald A. Genital herpes. In: Holmes KK, Mardh PA, Sparling PF, et al, eds. Sexually Transmitted Diseases. New York: McGraw-Hill, 1998:285–312.
11. Eluf-Neto J, Booth M, Muñoz N, Bosch FX, Meijer CJ, Walboomers JM. Human papillomavirus and invasive cervical cancer in Brazil. Br J Cancer 1994; 69: 114–119.
12. Ngelangel C, Muñoz N, Bosch FX, et al. Causes of cervical cancer in the Philippines: a case–control study. J Natl Cancer Inst 1998; 90: 43–49.
13. Muñoz N, Castellsagué X, Bosch FX, et al. Difficulty in elucidating the male role in cervical cancer in Colombia, a high-risk area for the disease. J Natl Cancer Inst 1996; 88: 1068–1075.
14. Bosch FX, Castellsagué X, Muñoz N, et al. Male sexual behavior and human papillomavirus DNA: key risk factors for cervical cancer in Spain. J Natl Cancer Inst 1996; 88: 1060–1067.
15. Ashley RL. Type-specific antibodies to HSV-1 and -2: review of methodology. Herpes 1998; 5: 33–38.
16. Bruzzi P, Green SB, Byar DP, Brinton LA, Schairer C. Estimating the population-attributable risk for multiple risk factors using case–control data. Am J Epidemiol 1985; 122: 904–914.
17. Oberle MW, Rosero-Bixby L, Lee FK, Sanchez-Braverman M, Nahmias AJ, Guinan ME. Herpes simplex virus type 2 antibodies: high prevalence in monogamous women in Costa Rica. Am J Trop Med Hyg 1989; 41: 224–229.
18. Stone KM, Zaidi A, Rosero-Bixby L, et al. Sexual behavior, sexually transmitted diseases, and risk of cervical cancer. Epidemiology 1995; 6: 409–414.
19. Hildesheim A, Mann V, Brinton LA, Szklo M, Reeves WC, Rawls WE. Herpes simplex virus type 2: a possible interaction with human papillomavirus types 16/18 in the development of invasive cervical cancer. Int J Cancer 1991; 49: 335–340.
20. de Sanjosé S, Muñoz N, Bosch FX, et al. Sexually transmitted agents and cervical neoplasia in Colombia and Spain. Int J Cancer 1994; 56: 358–363.
21. Weinberg A, Canto CL, Pannuti CS, Kwang WN, Garcia SA, Zugaib M. Herpes simplex virus type 2 infection in pregnancy: asymptomatic viral excretion at delivery and seroepidemiologic survey of two socioeconomically distinct populations in Sao Paulo, Brazil. Rev Inst Med Trop Sao Paulo 1993; 35: 285–290.
22. Carvalho M, de Carvalho S, Pannuti CS, Sumita LM, de Souza VA. Prevalence of herpes simplex type 2 antibodies and a clinical history of herpes in three different populations in Campinas City, Brazil. Int J Infect Dis 1999; 3: 94–98.
23. Hashido M, Lee FK, Nahmias AJ, et al. An epidemiologic study of herpes simplex virus type 1 and 2 infection in Japan based on type-specific serological assays. Epidemiol Infect 1998; 120: 179–186.
24. Peng HQ, Liu SL, Mann V, Rohan T, Rawls W. Human papillomavirus types 16 and 33, herpes simplex virus type 2, and other risk factors for cervical cancer in Sichuan Province, China. Int J Cancer 1991; 47: 711–716.
25. Ferlay J, Parkin DM, Pisani P. GLOBOCAN. vol. 1. Cancer Incidence and Mortality Worldwide. International Agency for Research on Cancer Cancerbase No. 3. CD-ROM. Lyon: IARC, 1998;
26. UNAIDS. Brazil. Epidemiological fact sheet on HIV/AIDS and sexually transmitted diseases. UNAIDS/WHO Epidemiological Fact Sheet. June 1–12, 1998.
27. UNAIDS. Philippines. Epidemiological fact sheet on HIV/AIDS and sexually transmitted diseases. UNAIDS/WHO Epidemiological Fact Sheet June 1–12, 1998.
28. Boucher FD, Yasukawa LL, Bronzan RN, Hensleigh PA, Arvin AM, Prober CG. A prospective evaluation of primary genital herpes simplex virus type 2 infections acquired during pregnancy. Pediatr Infect Dis J 1990; 9: 499–504.
29. Wald A, Koutsky L, Ashley RL, Corey L. Genital herpes in a primary care clinic: demographic and sexual correlates of herpes simplex type 2 infections. Sex Transm Dis 1997; 24: 149–155.
30. Cowan FM, Johnson AM, Ashley R, Corey L, Mindel A. Antibody to herpes simplex virus type 2 as serological marker of sexual lifestyle in populations. BMJ 1994; 309: 1325–1329.
31. Obasi A, Mosha F, Quigley M, et al. Antibody to herpes simplex virus type 2 as a marker of sexual risk behavior in rural Tanzania. J Infect Dis 1999; 179: 16–24.
32. Lewis LM, Bernstein DI, Rosenthal SL, Stanberry LR. Seroprevalence of herpes simplex virus type 2 in African American college women. J Natl Med Assoc 1999; 91: 210–212.
33. Potterat JJ, Rothenberg RB, Muth SQ. Network structural dynamics and infectious disease propagation. Int J STD AIDS 1999; 10: 182–185.
34. Decosas J, Kane F, Anarfi JK, Sodji KD, Wagner HU. Migration and AIDS. Lancet 1995; 346: 826–828.
35. O’Farrell N, Egger M. Circumcision in men and the prevention of HIV infection: a “meta-analysis” revisited. Int J STD AIDS 2000; 11: 137–142.