Background

In Botswana, HIV surveillance has been conducted since 1992 in public health facilities among pregnant women and men having signs of sexually transmitted infections (STIs). While the surveys provided valuable data on the prevalence of HIV in the country, they were limited in providing information on risk factors for contracting HIV. For many years, the World Health Organization (WHO) and the Joint United Nations Programme on AIDS (UNAIDS) have advocated for the adoption of second generation surveillance as a strategy for addressing some of the limitations of the HIV sentinel surveys. In particular, the importance of using behavioural data to inform and explain trends recorded in HIV infection was not emphasized and advocated for.

Attempts have been made in Botswana in response to this call. During the period 2000 and 2001, three behavioural surveys were conducted and yielding important population-based estimates of HIV/AIDS data on the knowledge, attitudes, behavior and practice of the population in Botswana:

1. The Botswana Multiple Indicator Survey (2000): That survey focused on women and interviewed 6,485 aged 15-49 in 14 districts (Gaborone, Francistown, Lobastse, Selebi-Phikwe, Southern, South East, Kweneng, Central, North-East, North West, Gantsi and Kgalagadi). The questionnaire used had several modules, but HIV related questions were limited and focused primarily on exposure and knowledge/attitudes.

2. The Sexual Behaviour of Young People in Botswana Survey (2001): The survey was focused on adolescents and young adults (10-24 years). Questionnaires covered four topic areas: knowledge, attitudes, sexual practices, and health seeking behaviour. A total of 4,328 interviews were completed [2100 (10-14 years); 900 (15-19 years); 900 (20-24 years) and 428 adult caregivers.

3. The Botswana AIDS Impact Survey (BAIS) (2001): The survey included men and women aged 10-64 years. It collected data on knowledge, attitudes, and sexual behaviour associated with the HIV/AIDS epidemic among 4,267 respondents and was the first national survey of its kind in Botswana. And as such, the survey provided a baseline for future surveys, monitoring changes in indicators, and programme evaluation.

A common observation from these three attempts is that awareness level on HIV/AIDS was high and so was condom use rate. However, accurate knowledge on the disease was still low and risky sexual behaviour common amongst the population. Precise estimates of sexual risk behaviours, AIDS knowledge and prevention behaviours, and other relevant indicators are required. Also required is the need to know more about recent developments in the social and behavioural sciences. To these ends, NACA has developed indicators to measure for behavioural change, and the measurement of these indicators would be most useful for the various intervention programmes that are in place or that will be put into place in the near future. Many of these indicators have yet to be introduced into national-level surveys in Botswana.

It is within this context that the second Botswana HIV/AIDS Impact Survey (BAISII) is proposed. This survey will be an important pilot test on the feasibility of including psychosocial factors in a national HIV behavioural survey of the Botswana population. It would be a nationally representative survey of persons 18 months to 64 years plus.