Social Communications and AIDS Population Behaviour Changes in Uganda Compared to Other Countries

Published by the Center for AIDS Development, Research and Education (CADRE) in 2004, this study investigates the impacts of social communications on population behaviour changes and HIV prevalence declines in Uganda. Casual sex and HIV prevalence had dramatically declined in Uganda between the late-1980s and mid-1990s, a trend that was not observed in Uganda's neighbouring countries suffering from high HIV incidence. The authors hypothesised that horizontally communicating HIV-related matters through social networks would have positive impacts on population behaviour and HIV prevalence. Comparisons were made between Uganda and its neighbouring countries (Kenya, Malawi, South Africa, Tanzania, Zambia, and Zimbabwe) to discern the impacts of HIV-related communication patterns.



Evaluation/Research Methodologies:

Two main sources of data were the population-based demographic and HIV behavioural data collected among men and women in Uganda and six other sub-Saharan countries through (1) Demographic and Health (DHS) surveys, and (2) Knowledge, Attitudes and Behaviours (KAPB) surveys. These surveys were administered in the mid- to late-1990s. The authors found high compatibility between individual surveys even though they were conducted in separate time points across the countries and the questionnaires had slight cross-national variations. In addition, the study also used separate KAPB survey data collected in Uganda in 1989 and 1995 to supplement the six-country DHS and KAPB surveys.

Each of the seven countries implemented public health and media-based AIDS awareness and information programmes. The surveys collected data on: knowledge of AIDS, communication channels for learning about AIDS, personal assessment of risk, reported sexual behaviour and behaviour change by age, sex, and place. Communication channels for acquiring AIDS knowledge were classified into three groups: mass (radio, TV, newspaper, pamphlet), institutional (religious, school, health clinic), personal (friend/relative, community, workplace).



Key Findings/Impact:

Uganda stood out in the patterns of AIDS communication. In Uganda, personal networks of friends and family were dominant channels of AIDS information. Eighty-two percent of women in Uganda heard of AIDS through personal networks, compared to 40-65% in other countries. Mass media dominated as the primary channel of AIDS communication in all other countries. The latter was the case for Uganda in 1985, but personal communication had replaced mass media as the dominant communication channel by 1995.



Another communication component, the knowledge of someone with AIDS or who had died from AIDS, also displayed a very characteristic pattern in Uganda. In Uganda, 91.5% of men and 86.4% of women knew someone with AIDS or who had died from AIDS, compared to 68-71% in Zambia, Kenya, and Malawi, and below 50% in Tanzania, Zimbabwe, and South Africa, despite much higher AIDS prevalence in these latter countries (except Tanzania). The authors suggest that horizontal communication of HIV infection and AIDS death through social networks in Uganda likely had preventive and deterrence effects for HIV transmission and unsafe sexual contact. Among 15-24 year old men in Uganda, 19.7% of those who knew someone with AIDS started using condoms, compared to 4.9% of those who did not know someone with AIDS. A similar difference in AIDS preventive practices was found among older men.



Social communication of AIDS matters was challenged by age and gender barriers. Knowledge of someone with AIDS was significantly lower in the younger ages, 15-19, in comparison to older ages, 20-54, in all countries but Uganda. The knowledge gap between the two age groups was only 2 percentage points in Uganda, comparing positively to the five other countries (no data was available for South Africa) where the gap hovered at 17% - 27% points. Gender barriers to communicating AIDS within couples were still significant in Uganda. Almost four times more men than women perceived themselves at low risk when their partner perceived themselves at high risk. The authors suggest that rather than the individual or partnership, it may be community-level social communications that are significant in influencing behavioural norms.



By modeling the relationship between social communication and epidemic stages, the authors found that a high level of horizontal social communication, such as that seen in Uganda, would allow the majority of people to know someone with AIDS early on in an epidemic, before the peak of HIV incidence. Conversely, if social communication is very restricted, even very high HIV prevalence and mortality could lead to very limited numbers of people knowing someone with AIDS. In the latter case, the behavioural impact of social communication would be much less significant in reducing HIV infections because risk awareness (through personally knowing someone with AIDS) is realised only after peak HIV incidence.



Although inconclusive, the authors also point out the possible interactions between vertical communication (formal interventions through mass media and institutional communication) and horizontal communication (personal and social networks) in AIDS prevention. By comparing the timings of public health campaigns and voluntary HIV testing in Uganda, the authors suggest that vertical communications (social marketing of condoms and increased public health reporting) alone might not have resulted in the observed behaviour change; a more plausible explanation is that vertical and horizontal communications operated in a mutually reinforcing manner, making Ugandans much more receptive to public health campaigns.

Editor's note, October 5 2018: This document is no longer available online.