Answers are grounded in this paper and cite their sources.
This study examines whether changes in reported sexual behavior in Lusaka, Zambia, are consistent with the evidence of declining HIV prevalence levels reported elsewhere. A comparison of findings from two cross-sectional surveys implemented in Lusaka between 1996 and 1999 shows statistically significant reductions in casual partnerships among men and women. Respondents with higher socioeconomic status were more likely to reduce casual sex. Condom use increased but the change was not statistically significant. Zambia has one of the highest levels of HIV prevalence in the world, with an estimated one-fifth of men and women between the ages of 15 and 49 years infected with HIV [1]. Recent research has shown declining HIV levels among 15–19-year-old women across Zambia and for 20–24-year-old urban women [2]. Because of issues related to the measurement of HIV levels [3–7], it is important to have evidence of behavioral changes that support findings of declining HIV prevalence [8]. This study uses data from two serial cross-sectional surveys on sexual behavior to determine whether changes in casual sex and condom use were reported in Lusaka between 1996 and 1999. The two surveys were representative of 15–49-year-old men and women living in Lusaka. Both surveys used three-stage probability samples (806 respondents were interviewed in 1996; 1211 in 1999), with 100 standard enumeration areas randomly selected in 1996, and 120 in 1999. The samples were drawn by the Central Statistical Office of Zambia. For both surveys, the response rate was 98%. The questionnaire used was based on the Partner Relations and Knowledge Attitude Behavior Practices Surveys implemented by the World Health Organization's Global Program on AIDS [9]. Experienced field supervisors from the Central Statistical Office were used, and enumerators were trained for one week before survey implementation. Same-sex interviews were conducted. Data were collected in October 1996 and June 1999. To determine socioeconomic status, an index of assets was created on the basis of the ownership of nine assets, including a bicycle, motorcycle, car, video player, radio, television, refrigerator, farm, and house. To measure sexual behavior, respondents were asked to describe their relationship with their last sexual partner. The response was coded as marital, regular, or casual. Respondents were then asked whether they used a condom in their last sexual intercourse. The survey also included questions on exposure to condom advertising carried out by the social marketing program. For example, respondents were asked if they had seen or heard advertising for Maximum brand condoms, and if so, through what sources. Weights were attached to match the age distributions of the two samples. Data from the two surveys were pooled and the chi-square test of independence [10] was used to determine statistical significance. All results presented in the Table 1 are weighted, but the number of cases shown are unweighted.Table 1: Percentage of women and men in Lusaka whose last partner was a casual partner and percentage who used a condom in last sex (with any partner): 1996–1999. Sample characteristics There were no significant differences in the socio-demographic characteristics of the 1996 and the 1999 samples. In each year, 51% of the respondents were men, 51% were married, and 19% had some primary education. Approximately one-third of respondents in each year came from households with up to one asset (not shown). Changes in casual sex Table 1 shows the percentage of sexually experienced women and men who reported that they had last sex with a casual partner in 1996 and 1999, by various independent variables. Overall, there was a significant reduction in reported casual partnerships among both women and men: approximately 7% of women reported having had last sex with a casual partner in 1996, compared with 3% in 1999; approximately 27% of men reported having had last sex with a casual partner in 1996, compared with 19% in 1999. Declines in casual sex were especially large for respondents who reported not having a partner at the time of the survey: 41% of women with no current partner reported having last sex with a casual partner in 1996, compared with 13% 1999; 73% of men who did not have a current partner in 1996 reported having last sex with a casual partner, compared with 49% in 1999. Significant declines in casual sex also occurred among those with higher socioeconomic status, including women with secondary or higher education, men with some secondary education, women with four or more assets, and men with two or three assets. Changes in condom use Table 1 also shows the percentage of sexually experienced women and men who reported that they had used a condom in last sex, by various independent variables. Women reported 17% condom use in 1996 and 21% in 1999, but the increase was not statistically significant. Men reported 24% condom use in 1996 and 28% in 1999, but the increase, again, was not statistically significant. There were, however, significant increases in condom use among women with a regular non-marital partner (from 33 to 50%), among men aged 30–34 years (from 7 to 20%), and among men who had been exposed to social marketing condom advertising through three or more media channels (44% in 1999, up from 30% in 1996). Our analysis of data from two rounds of representative household surveys of Lusaka has shown that there was a significant reduction in casual sex among both women and men between 1996 and 1999. Although condom use in last sex with any partner increased during this period, the increase was not statistically significant. Because casual sex has tended to be concentrated among Zambian men and women who are not in stable partnerships [11,12], these declines in casual sex reflect a substantial reduction in high-risk sexual activity in Lusaka between 1996 and 1999. Respondents with higher socioeconomic status backgrounds were more likely to reduce casual sex, a finding that is consistent with recent research showing a declining HIV prevalence among more educated Zambians [2]. These findings indicate the importance of intensifying efforts to promote behavior change among those with low socioeconomic status in Lusaka. Acknowledgements The author would like to thank Dominique Meekers, Muyiwa Oladosu, Kerry Richter and Nils Gade who commented on an earlier version of this paper, and Megan Klein and Cheryl Kolwicz who helped to edit it.
No takes yet. Share an insight, caveat, or question.
Sohail Agha (2002) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: