Household Survey of Behavioural Risks and HIV Sero Status in two districts in Botswana pptx

Free download from www.hsrcpress.ac.za
Household survey of behavioural risks and
HIV sero-status in two districts in Botswana
GN Tsheko
LW Odirile
K Bainame
M Segwabe
PS Nair
O Ntshebe
Free download from www.hsrcpress.ac.za
Research report prepared by the Human Sciences Research Council (HSRC) and the Nelson
Mandela Children’s Fund (NMCF) for the strategy of the WK Kellogg Foundation (WKKF) for
the care of orphans and vulnerable children (OVC) in Botswana, South Africa and Zimbabwe
in commemoration of the WKKF’s 75th anniversary.
Published by HSRC Press
Private Bag X9182, Cape Town, 8000, South Africa
www.hsrcpress.ac.za

First published 2007

ISBN 978-0-7969-2196-3

© 2007 Human Sciences Research Council

Copyedited by David Le Page
Typeset by Janco Yspeert
Cover design by Oryx Media
Cover photo: © Tessa Frootko Gordon/iAfrika Photos
Print management by Compress
Distributed in Africa by Blue Weaver
Tel: +27 (0) 21 701 4477; Fax: +27 (0) 21 701 7302

www.oneworldbooks.com
Distributed in Europe and the United Kingdom by Eurospan Distribution Services (EDS)
Tel: +44 (0) 20 7240 0856; Fax: +44 (0) 20 7379 0609
www.eurospangroup.com/bookstore

Distributed in North America by Independent Publishers Group (IPG)
Call toll-free: (800) 888 4741; Fax: +1 (312) 337 5985
www.ipgbook.com
Free download from www.hsrcpress.ac.za
Acknowledgements iv
Tables v
Abbreviations and acronyms vi
Executive summary vii
Chapter 1 Introduction 1
Prevalence of HIV/AIDS in Botswana 1
Background to the OVC project 1
Purpose of the BSS Survey 2
Objectives of the study 2
Chapter 2 Literature review 3
Behavioural risks for HIV/AIDS in Botswana 3
Chapter 3 Methodology 7
The original BAIS II Survey 7
The Botswana BSS methodology 8
Chapter 4 Findings: Central Serowe District 9
Demographic characteristics 9
HIV prevalence 10
HIV knowledge and attitudes 11
HIV risk behaviour 12
Substance abuse 13
Awareness of and access to social and medical services 14

Human rights and HIV/AIDS issues 17
Chapter 5 Findings: Kweneng West District 19
Demographic characteristics 19
HIV prevalence 20
HIV knowledge and attitudes 21
HIV risk behaviour 22
Substance abuse 23
Awareness of and access to social and medical services 23
Human rights and HIV/AIDS issues 26
Chapter 6 Conclusion 29
Discussion 29
Recommendations 30
References 33
CONTENTS
Free download from www.hsrcpress.ac.za
iv
This study was supported by the WK Kellogg Foundation and undertaken by the Masiela
Trust Fund OVC Research-Botswana Team under the umbrella of the Human Sciences
Research Council in South Africa. The Masiela Trust Fund OVC Research Team is indebted
to the Central Statistics Office in the Ministry of Finance and Development Planning for
allowing the team to use the BAIS II data (Central Statistics Office 2004).
We would also like to thank both Professors Leickness Simbayi and Karl Peltzer of the
Human Sciences Research Council for their advice and comments during the preparation
of this report.
ACKNOWLEDGEMENTS
Free download from www.hsrcpress.ac.za
v
Table 4.1: Demographic and basic social characteristics of Central Serowe District
Table 4.2: HIV prevalence in Central Serowe District by sex, school attendance, marital
status, and age group

Table 4.3: HIV prevalence in Central Serowe District by skills level
Table 4.4: Correct responses to questions on knowledge and misconceptions on HIV/
AIDS by sex in Central Serowe District
Table 4.5: Percentage of respondents who have had multiple sexual partners in the last
twelve months
Table 4.6: Percentage of respondents aged 10–64 years who have ever had alcohol in
their lifetime
Table 4.7: Awareness of social and medical services in the area by sex
Table 4.8: Awareness of social and medical services in the area by age group
Table 4.9: Accessing of social and medical services in the area by sex
Table 4.10: Accessing of social and medical services in the area by age group
Table 4.11: Type of support received by PLWHA
Table 4.12: Sources of support for PLWHA
Table 4.13: Type of support offered by individual community members
Table 4.14: Percentage of responses to some human rights issues pertaining to HIV/AIDS
Table 5.1: Demographic and basic social characteristics of Kweneng West District
Table 5.2: HIV prevalence in Kweneng West District by sex, school attendance, marital
status, and age group
Table 5.3: HIV prevalence in Kweneng West District by skills level
Table 5.4: Correct responses to questions on knowledge, misconceptions on HIV/AIDS
in Kweneng West District
Table 5.5: Percentage of respondents who have had multiple sexual partners in the last
twelve months
Table 5.6: Percentage of respondents aged 10–64 years who have ever had alcohol in
their lifetime
Table 5.7: Awareness of social and medical services in the area by sex
Table 5.8: Awareness of social and medical services in the area by age group
Table 5.9: Accessing of social and medical services in the area by sex
Table 5.10: Accessing social and medical services in the area by age group
Table 5.11: Type of support received by PLWHA

Table 5.12: Sources of support for PLWHA
Table 5.13: Percentage of responses to some human rights issues pertaining to HIV/AIDS
TABLES
Free download from www.hsrcpress.ac.za
vi
AIDS Acquired Immune Deficiency Syndrome
ARV anti-retroviral
BAIS Botswana AIDS Impact Survey
BSS Behavioural Risks and Sero-Status Survey
CBO community-based organisations
CSO Central Statistics Office
FBO faith-based organisations
HBC home-based care
HIV Human Immuno-deficiency Virus
HSRC Human Sciences Research Council
IPT Isoniazid Preventive Therapy
KABP knowledge, attitudes, behaviour and practices
NACA National AIDS Coordinating Agency
NGO non-governmental organisations
OVC orphans and vulnerable children
PLWHA people living with HIV/AIDS
PMTCT prevention of mother-to-child transmission
SADC Southern African Development Community
STI sexually transmitted infection
UNAIDS Joint United Nations AIDS Programme
UNICEF United Nations Children’s Fund
WHO World Health Organisation
ABBREVIATIONS AND ACRONYMS
Free download from www.hsrcpress.ac.za
vii

This report presents the findings of the Behavioural Risks and HIV Sero-Status Survey
(BSS) for the Central Serowe District and Kweneng West District in Botswana. The
purpose of the survey was to determine the knowledge, attitudes, sexual behaviours,
practices, prevention, care and support issues concerning HIV/AIDS among the
population in the Central Serowe District and Kweneng West District. Specifically, the
survey quantified HIV prevalence, sexual risk behaviours and other practices among
adults and children.
The archival research method that was employed as an existing national database
obtained from the Botswana AIDS Impact Survey of 2004 (BAIS II) was used for the
analysis of the behavioural risks and HIV sero-status for the Central Serowe District and
Kweneng West District. The BAIS II was carried out by the Central Statistics Office in
the Ministry of Finance and Development Planning from 12 February to 31 July 2004 in
all districts in the country. The target population for BAIS II was all household members
aged 10–64 years for the individual questionnaires, and individuals aged 18 months and
above for the HIV status biomarker. The questionnaire covered various issues such as HIV
knowledge and attitudes, awareness, availability and accessibility of social and medical
services. A community schedule was also administered to the target population, while
another workplace questionnaire was administered to three organisations in each district.
The national response rate from BAIS II for the household interviews was 93% (15 878
individuals), while 61% (15 161 individuals) submitted specimens for HIV testing. In
Central Serowe District and Kweneng West District, the response rates for interviews were
96.3% and 94.0% respectively, while 59.7% (833 individuals) and 60.2% (195 individuals)
respectively submitted specimens for HIV testing.
Secondary data analysis was conducted using the Statistical Package for Social Science
(SPSS). Only data from the two districts of Central Serowe and Kweneng West were
analysed for this report, and this largely involved using descriptive statistics.
Findings
HIV status and demographic data
HIV prevalence in Central Serowe was 18.5%. Prevalence was higher among females
(22.0%) than among males (14.2%). HIV was more prevalent among individuals living

together (33%) than among those who were never married (19%). An analysis of the HIV
prevalence by skill showed no difference between the unskilled and skilled workers, with
their rates ranging between 32.8% and 32.1% respectively. Adults (25 years and above)
were more affected at 29%, while prevalence for youth was 13.1%. The prevalence of
children aged 2–11 years and 12–14 years was relatively low (7.4% and 5% respectively).
HIV prevalence in Kweneng West was 10.8%. Prevalence was higher among females
(12.1%) than among males (10%). Prevalence was higher among individuals who were
living together (38%) than among those who were never married (15.2%). Prevalence was
highest among adults aged 25 years and above (19%), and lowest among children aged
2–11 years (1%).
HIV knowledge, attitudes and risk behaviour
Misconceptions about HIV/AIDS were found in both districts. For example, only 88.3% of
males and 87.9% of females in Central Serowe correctly identified that a healthy person
can have HIV. Misconceptions about getting infected with HIV through mosquito bites
EXECUTIVE SUMMARY
Free download from www.hsrcpress.ac.za
Household survey of behavioural risks and HIV sero status
viii
were high, as only 53.2% males and 54.6% females responded correctly to the question.
In Kweneng West, 61.3% of males and 61.7% of females correctly identified that a healthy-
looking person can have HIV. Regarding getting infected with HIV through mosquito
bites, 41.3% of males and 40.4% of females responded correctly to the question.
The proportion of respondents who reported having more than one sexual partner in
Central Serowe was highest among those who never married for both males and females
(31.5% and 10.4% respectively). Multiple sex partners were further found among males
who were married or living with partners (10.3%). Data also show that this behaviour was
most common among males in the age group 15–24 years (32.7%) and females in the age
group 15–24 years (8.0%).
In Kweneng West, multiple sex partners were observed among the never married couples
(28.6% males and 5.6% females; two males and one female) and those who were living

together (10.5% males and 6.3% females; four males and two females).This behaviour was
found mostly among males aged 15–24 years (n = 4) and females aged 25–49 years
(n = 3).
The results showed that 47.0% of males and 20.6% of females in Central Serowe had
taken alcohol in their lifetime. Drinking alcohol was common among youth and older
groups, especially males as observed in the age groups of 15–24 years (35.8% males and
22.9% females) and 25–49 years (66.7% males and 23.7% females) The same pattern of
males using more alcohol was also observed in Kweneng West District. In the age group
15–24 years, 35.7% of males and 12.0% of females used alcohol, whereas in the age group
25–49 years 52.0% of males and 17.4% of females engaged in this behaviour.
Social and medical services
Most of the respondents in Central Serowe District were aware of the social and medical
services in their community, with females generally showing more awareness than males;
for example, 75.8% of males and 83.5% of females were aware of the destitute care
programme. The age group 25–49 years generally showed high levels of awareness about
social and medical services in their community as compared to other age groups: 77%
for home-based care (HBC); 73% for prevention of mother-to-child transmission of HIV
(PMTCT); 80% for orphan care; and 84% for destitute care programmes. The existence
of Isoniazid Preventive Therapy (IPT) and organisations for people living with HIV/AIDS
(PLWHA) were the least known among the entire population in the Central Serowe
District. Despite the high awareness levels about social and medical services, data showed
low levels of access, with females accessing HBC services more than any other service
(at 6%). Males were accessing HBC and the destitute programme more than any other
services (5.4% for each).
The numbers of respondents who were aware of the social and medical services in
Kweneng West were far fewer than was found in Central Serowe. Generally, females
showed higher levels of awareness as compared to males. For example, 55.5% of females
and 40.5% of males were aware of the orphan care programme, while 51.8% of females
and 33.3% of males were aware of the HBC programme. Overall, the age group 25–49
years generally showed the highest levels of awareness about social and medical services

in their community as compared to other age groups: 67.6% for destitute care; 63.4%
for HBC; 66.2% for orphan care; and 38.0% for PMTCT of HIV. Awareness of both IPT
and anti-retroviral (ARV) drug treatment programmes, and organisations for PLHWA, was
lowest among the entire population in the Kweneng West Sub-district. Despite awareness
Free download from www.hsrcpress.ac.za
ix
about social and medical services that existed in Kweneng West Sub-district, data showed
low levels of access, with both males and females accessing the destitute programme
more than any other service, at 15.2% and 15.9% respectively.
PLWHA in both districts received different kinds of support, which was provided by both
government and civil society. The most common types of support received in Central
Serowe were counselling (27%), education (24%), HBC (13%), and food (6%). There was
also evidence that most of the services offered to PLWHA were from civil society (22.6%)
and government organisations (6.6%). The most common types of support received in
Kweneng West included money, food and education (each at 1.7%). There was evidence
that most of the services offered to PLWHA were provided by civil society organisations
(61.2%).
Recommendations
Given the findings of the study, the following are the actions required to address HIV/
AIDS in the Central Serowe District and Kweneng West District:
1. Information, education and communication (IEC) strategies should continue to
address both the basic facts and myths and/or misconceptions around the spread of
HIV/AIDS. It would also be helpful to assess the social influences surrounding the
spread of misconceptions and myths.
2. Behaviour-change strategies must embrace everyone, hence the need to implement
behaviour-change strategies in a wide range of settings to make them accessible.
These strategies should include issues of consistent and correct condom use, use
and abuse of alcohol, and informed decision-making.
3. There is a need for programmes that educate youth about the dangers of alcohol.
Findings from both districts point to the fact that youth start consuming alcohol at a

very young age, hence the need to intensify such programmes.
4. There is a need for advocacy campaigns to promote the availability of various HIV/
AIDS related services.
Executive summary
Free download from www.hsrcpress.ac.za
Free download from www.hsrcpress.ac.za
1
Introduction
Prevalence of HIV/AIDS in Botswana
According to the last population census, Botswana has a population of 1.7 million people
(Central Statistics Office 2001). In the Botswana 2003 Sentinel Survey conducted among
7 251 pregnant women attending ante-natal clinics in various parts of the country, the
overall HIV prevalence was 37.4%. The highest age-specific prevalence was observed
among those aged 25–29 years at 49.7% (NACA, 2003). While the prevalence in older age
groups was shown to be increasing, prevalence in the age group 15–19 years remained
fairly stable and ranged from 21–23% between 2002 and 2003 (NACA, 2003).
The most recent statistics from the Botswana AIDS Impact Survey II (BAIS II) (Central
Statistics Office 2004) indicated that the overall HIV prevalence in the general population
aged 18 months and older was measured at 19.8% for females and 13.9% for males,
with overall national prevalence at 17.1% (Central Statistics Office 2004). The BAIS
II also showed that the very young, aged 18 months to 4 years, had the lowest HIV
prevalence of 6.3%, while the age group 30–34 years had the highest HIV prevalence
of 40.2%. According to BAIS II, the respondents who were hardest hit by the epidemic
were those between 25 to 54 years old with prevalence that ranged from a low of 20.9%,
among those in the 50–54 age group, to a high of 40.2% in the 30–34 age group. Young
people under the age of 19 years and old people aged 65 and above had relatively low
prevalence (below 10%) (Central Statistics Office 2004).
Background to the OVC project
The Human Sciences Research Council (HSRC) together with its research partners within
the Southern African Development Community (SADC) region – University of Botswana

and Botswana Harvard Partnership in Botswana as well as the National Institute of Health
Research and Biomedical Research & Training Institute‘s Centre for International Health
and Policy in Zimbabwe – were commissioned by the WK Kellogg Foundation (WKKF)
to develop and implement a five-year intervention project on orphans and vulnerable
children (OVC), as well as families and households coping with an increased burden of
care for affected children, initially in Botswana, South Africa and Zimbabwe. The Masiela
Trust Fund was chosen as a grantmaker to work with researchers from the University
of Botswana. The Masiela Trust Fund in turn selected community-based organisations
(CBOs) and faith-based organisations (FBOs) to implement the appropriate intervention
programme at all the identified sites.
The project also looked at families and households coping with the increased burden
of care for affected children in Botswana, South Africa and Zimbabwe. The specific
objectives of the project were as follows:
• to assess the social conditions, health, development and quality of life of OVC;
• to identify family and household systems for coping with the burden of care for
OVC at family, ward, community, national and international level;
• to use the information obtained to build capacity in community-based systems for
sustaining care and support to vulnerable children and households over the long
term; and
CHAPTER 1
Free download from www.hsrcpress.ac.za
Household survey of behavioural risks and HIV sero status
2
• to use information obtained to improve HIV/AIDS awareness, advocacy and
policy-support programmes for the benefit of vulnerable children, families and
communities.
The overall philosophy behind the project was to empower communities to help, placing
a very strong emphasis on the sustainability of the project beyond donor funding.
Purpose of the BSS Survey
The purpose of the study was to determine knowledge, attitudes, behaviour and practices

(KABP), prevention issues, care programmes and human rights issues concerning HIV/
AIDS among the general population in Serowe/Palapye and Kweneng West Districts.
Objectives of the study
The specific objectives of the HIV BSS associated with the objectives of the project stated
above were:
• to quantify HIV prevalence in the two sites especially among children;
• to determine KABP with regard to HIV/AIDS; and
• to identify prevention issues and care programmes as well as human rights issues
concerning HIV/AIDS.
Free download from www.hsrcpress.ac.za
3
Literature review
HIV/AIDS continues to be the leading cause of death in most African countries, as well as
a major contributor to the increasing number of orphans in the world. It is estimated that
to date, more than 15 million children under the age of 18 have been orphaned by HIV/
AIDS worldwide. Research shows that in sub-Saharan Africa, AIDS has left vast numbers
of children without one or both parents (UNAIDS 2005). Botswana, like any other country
in this region, has not escaped the impact of HIV/AIDS. By the end of 2003, there were
42 000 registered orphans and the growing number of OVC in the country has been
acknowledged as a national crisis (NACA 2003).
Worldwide, HIV/AIDS continues to claim more and more lives. It is becoming increasingly
clear that many parents die, leaving their children orphaned. In this regard, it is important
to assess behavioural risks that encourage the spread of HIV/AIDS in order to develop
policies and guidelines for the establishment and implementation of relevant strategies
and interventions that ensure and consider the needs of OVC (UNICEF 2001; UNAIDS
2005).
Behavioural risks for HIV/AIDS in Botswana
HIV/AIDS cases in Botswana and other countries in sub-Saharan Africa, as well as
globally, result mainly from either unprotected sexual intercourse with an HIV-infected
person or the use of HIV-contaminated injection drug equipment. Presently, the most

effective way recommended to reduce the spread of HIV/AIDS is behaviour change and
this strategy is widely promoted internationally today (Behrman et al. 2003; Chimwaza &
Watkins 2004).
It is posited that different behaviours that can put one at risk include unprotected sex or
sex with multiple partners. Unprotected sex and inconsistent condom use, as well as lack
of knowledge on HIV/AIDS issues, have been strongly linked to risky behaviours that can
expose one to HIV/AIDS. Another commonly cited behavioural risk is drug and alcohol
abuse (Jernigan 2001). Alcohol has been noted as a factor that increases vulnerability to
HIV/AIDS.
It has been acknowledged that as long as an HIV/AIDS cure is not found, changing
behaviour practices remains the only hope of reducing the spread of HIV (ACHAP 2002).
Behavior change practices that are likely to lead to reduced infections include abstinence,
proper and consistent use of condoms, knowing one’s HIV status and reduction of
multiple sexual partners (Green et al. 2006).
Unprotected sex
Research has shown that any form of unprotected sex (anal, vaginal or oral) can transmit
HIV (Meehan 2004). In assessing the sexual behaviour of young people in Botswana it
was found that the youth did not acknowledge the risk of unprotected sex because they
believed that they could identify someone else with HIV/AIDS,
and did not believe that
they would contract HIV. The practice of youth engaging in unprotected sex has long
been noted. The numbers of young people who engage in unprotected sex with multiple
partners continues to grow (UNAIDS/WHO 2005).

CHAPTER 2
Free download from www.hsrcpress.ac.za
Household survey of behavioural risks and HIV sero status
4
Another factor that suggests that people still engage in unprotected sex is the fact
that many unmarried women fall pregnant, as identified in the Botswana 2003 second

generation surveillance, which reported that over 80% of antenatal attendees in the survey
were single mothers.
Inconsistent condom use
Proper and consistent use of condoms has been cited as an effective mode of HIV/
AIDS prevention. However, studies in Africa have shown that more people continue to
be infected as they ignore the use of condoms or engage in inappropriate condom use
(UNAIDS/WHO 2005).
It should be noted that condom use continues to be a method that is promoted in
Botswana to decrease the spread of HIV/AIDS. Research has shown that over the past
eight years peoples’ attitudes towards condom use appear to be becoming more positive.
In a study on the sexual behaviour of young people in Botswana (NACA 2004), the
large majority of the participants (88%) claimed that they would insist on condom use
every time they had sexual intercourse with either regular or irregular partners. However,
they believed it was impossible to use a condom every time one has sex because of
emergencies and circumstances which pressurise them to ignore condom use at times.
However, other studies have shown that young people, compared to any other group,
continue to engage in unprotected sexual intercourse. A segmentation study conducted
at the University of Botswana in 2003 showed that 60% of the party boys’ segment used
condoms consistently, indicating that 40% did not. The main reason cited for using
condoms in this segment was to avoid pregnancy and sexually transmitted infections, as
well as to avoid getting HIV. About 35% of the party boys also believed that condoms
were not particularly effective while more than average reported dissatisfaction and/or
problems with condoms (University of Botswana 2003).
Knowledge
Many researchers have long recognised the importance of knowledge in the fight against
HIV/AIDS. Knowledge is believed to empower individuals to make informed decisions as
well as to make plans to reduce risk (Carroll 1991). In 2002, the government of Botswana,
UNICEF, UNAIDS and Populations Services International conducted a study about sexual
information, knowledge, attitudes, practices and behaviour of young people. In this study
a total of 2 100 interviews were conducted with 10–24 year-olds and 428 caregivers.

This study revealed that youth from towns and big villages had more knowledge about
HIV/AIDS and sexually transmitted infections (STIs) as compared to those in rural areas.
However, the study also concluded that there were some misunderstandings regarding
changing behaviours and promoting positive attitudes. This is in agreement with previous
studies (see, for example, Kirby 1997), which indicated that even though people seem to
have basic knowledge about HIV/AIDS, behaviour has not changed much. Therefore, the
importance of connecting knowledge and behaviour and of identifying the missing link
between knowledge and behaviour change remain a challenge.
The Botswana 2003 second-generation surveillance survey concluded that youths in
Botswana were knowledgeable about HIV/AIDS issues. In this survey, a large number
of youths answered all knowledge questions on HIV/AIDS correctly. These included
questions on the modes of infection and transmission and knowledge about condom
Free download from www.hsrcpress.ac.za
Literature review
5
use. However, there were still some myths and misconceptions about how HIV/AIDS is
transmitted. Such myths need to be addressed in order to reduce the escalating spread of
HIV/AIDS.
Drug and alcohol abuse
There has been a growing recognition that drinking can expose an individual to HIV/
AIDS. Some of the documented consequences of drug and alcohol abuse include rape
and unprotected sex. It is believed that those under the heavy influence of drugs and
alcohol are likely to make uninformed choices (Wechsler et al. 2004).
In 2001, the Ministry of Health in Botswana conducted a rapid situation assessment
survey on substance abuse and drug trafficking and discovered that 56% of the sampled
population had experimented with alcohol. About 18% of the total were females who
used alcohol occasionally, and about 32% used it once a week, while 11% used it
occasionally during social events. This study noted that statistics of men who used alcohol
outnumber those of women and further indicated that males tended to start drinking at
an earlier age.

Research into how drug and alcohol abuse can affect people’s attitudes toward sex
has been conducted. In a study on the attitudes of college students and the use of
alcohol conducted by the Harvard School of Public Health in 2000 (Harvard School of
Public Health 2000), it was found that about 43% of college students engaged in high
risk drinking frequently. Of these, about 20% of women experienced unwanted sexual
advances. The study concluded that after heavy drinking, close to half of those who took
alcohol and drugs engaged in risky sexual behaviours.
Free download from www.hsrcpress.ac.za

Free download from www.hsrcpress.ac.za
7
Methodology
The original BAIS II Survey
The Botswana AIDS Impact Survey (BAIS II) (Central Statistics Office 2004) is a nationally
representative survey designed to identify factors – behaviour, knowledge, attitudes and
cultural practices – that are associated with the HIV epidemic amongst the population
aged 10–64 years. The survey was also conducted to measure the country’s population-
based estimate of HIV/AIDS prevalence amongst the population aged 18 months and
over.
Aims and objectives
The objectives of BAIS II were to:
• generate a nationally representative population-based estimate of HIV/AIDS
prevalence amongst the population aged 18 months to 64 years;
• identify and document those factors (behaviour, knowledge, attitudes, cultural
practices) that are associated with the prevention, infection and impact mitigation of
the HIV epidemic amongst the population aged 10–64 years; and
• establish core benchmarks against which successive progress on the impact of the
national response to HIV/AIDS can be measured.
BAIS II methodology
Sampling frame

The survey design was a two-stage design. The sampling frame in the first stage was
based on the 2001 population and housing census. This comprised the list of all
enumeration areas (EA) found in three geographical regions: cities and towns; urban
villages; and rural districts.
The sampling frame in the second stage was from the selected EAs. The selected EAs
served as primary sampling units (PSUs). Probability proportional to measure of size
(pps) method was used to sample these EAs. The sampling of households was carried out
systematically from the list of occupied households in the selected EAs.
Instruments
Questionnaires were the main tools for BAIS II. There was a questionnaire designed for
each of the following:
• households;
• individuals;
• workplace; and
• community.
The household questionnaire was administered only to those households which were
selected through the survey sampling scheme. The focus of the tool was on the education
and economic activity at household level. The individual questionnaire was administered
to eligible individuals from the household questionnaire. The focus of this tool was on
sexual behaviour, knowledge about HIV/AIDS, access to interventions, and attitudes
towards PLWHA. The workplace questionnaire was administered to a maximum of three
institutions in the selected EAs: private; parastatal; and government. The focus of this tool
CHAPTER 3
Free download from www.hsrcpress.ac.za
Household survey of behavioural risks and HIV sero status
8
was on HIV/AIDS policy issues. The community schedule was administered to community
members in the selected EAs. This tool focused mainly on availability of resources and
support for PLWHA.
The BAIS II instruments were pre-tested in and around Gaborone in November 2003.

Field supervisors were trained for three weeks while enumerators were trained for two
weeks. Data collection was conducted between February and July 2004.
During BAIS II a total of 7 612 households were interviewed out of the targeted
8 275, generating a response rate of 92%. A total of 16 992 persons were eligible for the
individual interviews, and 15 878 were successfully interviewed, with a response rate of
93%. A total of 24 756 were eligible for submitting specimens for HIV testing, and 15 161
provided a specimen for HIV testing, producing a response rate of 61% (for more details
see BAIS II Survey report and BAIS II popular report [Central Statistics Office 2004]).
In BSS the focus was mostly on the individual level responses and to some extent on the
biomarker. The analysis of data for the biomarker covered persons aged 18 months and
older, while the analysis for the individual responses covered persons aged 10–64 years in
respect of their HIV knowledge, attitudes and behaviours.
The Botswana BSS methodology
Research method
The archival research method was employed in the present study, and an existing
database from the Botswana AIDS Impact Survey of 2004 (Central Statistics Office 2004)
was used.
Sample
BSS focused on only two districts: Central Serowe (of which Palapye is one of the urban
villages) and Kweneng West. These are the two research sites for the OVC project.
Data that were collected for BAIS II using the four questionnaires and biomarker were
organised and analysed to meet the objectives of BSS.
The total population for Central Serowe was 170 864 (53.7% female, 46.3% male) and for
Kweneng West 46 231 (59.8% female, 40.2% male). There were 634 households sampled
in Central Serowe and 170 households sampled in Kweneng West. The demographic
characteristics are presented in Section 4.1.
Data analysis
Secondary data analysis was conducted on the data using the Statistical Package for Social
Science (SPSS) – mostly descriptive statistics such as frequencies of occurrences were
analysed. Associations between variables were tested using chi-square tests. The 95%

confidence level was used to determine significance of the associations.
Free download from www.hsrcpress.ac.za
9
Findings: Central Serowe District
Demographic characteristics
The demographic and other basic social characteristics of the study sample in Central
Serowe District are shown in Table 4.1 below. Altogether, a total of 2 173 individuals were
covered in the district – 54.9% of whom were female, and 45.1% were male. Forty-five per
cent of the sample were adults (25+ years), while 28.0% were children aged 2–11 years,
and 19.8% were young adults or youth. The majority of the sample (57.1%) was ‘never
married’, while 32.6% were ‘living together’.
Table 4.1: Demographic and basic social characteristics of Central Serowe District
Sex Number %
Male 980 45.1
Female 1 193 54.9
Total 2 173 100.0
Age group
Children aged 2–11 581 28.0
12–14 (children) 147 7.1
15–24 (youth) 411 19.8
25 and above (adults) 934 45.1
Total 2 073 100.0
Marital status
Living together
a
486 32.6
Ever married
b
155 10.4
Never married 852 57.1

Total 1 493 100.0
Relationship to head of household
Head 607 27.9
Spouse 136 6.3
Son/daughter 611 28.1
Stepchild 5 0.2
Grandchild 414 19.1
Parent 14 0.6
Grandparent 9 0.4
Brother/sister 108 5.0
Nephew/niece 140 6.4
Son/daughter-in-law 8 0.4
Parent-in-law 1 0.0
Other relative 74 3.4
Not related 46 2.1
Total 2 173 100.0
Source: Central Statistics Office 2004
Notes: a) Includes those who were married and those who were cohabiting
b) Includes those who were separated, divorced and widowed
CHAPTER 4
Free download from www.hsrcpress.ac.za
Household survey of behavioural risks and HIV sero status
10
Information on household composition and the relationship of the members to the head
of the household was captured. Results show that 27.9% of the sample comprised heads
of the household, 28.1% were sons/daughters and 19.1% were grandchildren.
HIV prevalence
Table 4.2 below shows HIV prevalence by sex, school attendance, marital status and age
group. The overall prevalence of HIV in Central Serowe District was 18.5%. It was higher
among females (22.0%) than males (14.2%). The prevalence was highest among those with

primary education (31.0%), followed by those who had acquired a diploma (27.3%) and
those with secondary and certificate level of education (26.4% and 26.2% respectively).
Similarly, prevalence was higher among those who were living together (28.0%) followed
by those who ever married (23.8%). Prevalence was the highest in the adult group (25+
years) at 29.0% and lowest in the 12–14 age group at 4.9% (see Table 2).
Table 4.2: HIV prevalence in Central Serowe District by sex, school attendance, marital status, and
age group
Sex N HIV Prevalence
Male 508 14.2
Female 632 22.0
Total 1 140 18.5
Educational status
Primary 200 31.0
Secondary 246 26.4
Certificate 42 26.2
Diploma 22 27.3
Degree 12 25.0
Don’t know 618 10.4
Total 1 140 18.5
Marital status
Living together 271 28.0
Ever married 80 23.8
Never married 508 18.7
Total 859 22.1
Age group
Children aged 2–11 272 7.4
12-14 (children) 82 4.9
15–24 (youth) 252 13.1
25 and above (adults) 524 29.0
Total 1130 18.5

Source: Central Statistics Office 2004
Table 4.3 on page 11 shows HIV prevalence in the district by the skill level of
respondents – skilled, semi-skilled or unskilled worker. Table 3 shows that the prevalence
of HIV was highest among unskilled workers at 32.8%, followed by skilled workers at
32.1%.
Free download from www.hsrcpress.ac.za
Findings: Central Serowe District
11
Table 4.3: HIV prevalence in Central Serowe District by skills level
Variable N HIV prevalence
Skilled workers 84 32.1
Semi-skilled workers 165 21.8
Unskilled workers 122 32.8
Not stated 769 14.0
Total 1 140 18.5
Source: Central Statistics Office 2004
HIV knowledge and attitudes
Table 4.4 shows responses in relation to knowledge and attitudes about HIV and AIDS.
Most of the respondents (88.0%) knew that a healthy person could have HIV (88.3% of
males and 87.9% of females) and that condoms could reduce HIV transmission if used
correctly and consistently (89.9% of sample; 88.0% of males and 91.4% of females). It was
also observed that most respondents (90.1%) knew that they could reduce their chances
of getting HIV if they had only one sexual partner who was uninfected and who did not
have other sexual partners (89.2% of males and 90.9% of females), and that HIV could be
transmitted from mother to child (90.5% of sample; 86.5% of males and 93.7% of females).
Table 4.4: Correct responses to questions on knowledge and misconceptions on HIV/AIDS by sex in
Central Serowe District
Question Correct
response
Male

N = 342
Female
N = 428
Total
N = 770
n% n % n%
Is it possible for a healthy looking
person to have the AIDS virus?
Yes 302 88.3 376 87.9 678 88.0
Can people reduce their chances of
getting HIV/AIDS by using a condom
correctly every time they have sex?
Yes 300 88.0 391 91.4 691 89.9
Can people reduce their chances of
getting HIV/AIDS by having only one
uninfected sexual partner who has no
other partners?
Yes 305 89.2 388 90.9 693 90.1
Can HIV/AIDS be transmitted from a
mother to a child?
Yes 296 86.5 401 93.7 697 90.5
Do you think that a person can get
infected with HIV/AIDS through
mosquito bites?
No 182 53.2 233 54.6 415 54.0
Can a person get infected with HIV/
AIDS by sharing a meal with a person
who has HIV/AIDS?
No 194 56.7 262 61.2 456 59.2
Can people get HIV/AIDS because of

witchcraft?
No 261 76.3 349 81.5 610 79.2
Source: Central Statistics Office 2004
Free download from www.hsrcpress.ac.za
Household survey of behavioural risks and HIV sero status
12
Over half of the respondents (54.0%; 53.2% of males and 54.6% of females) correctly
reported that a person could not get infected with HIV/AIDS through mosquito bites.
59.2% of participants said a person could not get infected with HIV by sharing a meal
with a person who had AIDS (56.7% of males and 61.2% of females) while 79.2% of the
participants responded that a person could not get infected with HIV through witchcraft
(76.3% of males and 81.5% of females).
HIV risk behaviour
The study investigated the number of sexual partners in the past twelve months among
the respondents. Table 4.5 below captures results for respondents who reported multiple
sexual partners by age, educational attainment, marital status, and religion in the twelve
months prior to the study.
Table 4.5: Percentage of respondents who have had multiple sexual partners in the last twelve months
Variable Male Female Total
N%N%N%
Age
10–14 1 100 0 0 1 100
15–24 17 32.7 6 8.0 23 18.1
25–49 24 19.4 11 6.4 35 11.9
50+ 2 9.1 0 0 2 5.6
Total 44 22.1 17 6.3 61 13.3
Educational status
Non formal 1 25.0 1 50.0 2 33.3
Primary 7 15.2 3 3.9 10 8.1
Secondary 18 25.0 8 6.6 26 13.4

Higher 12 34.3 3 9.7 15 22.7
Total 38 24.2 15 6.5 53 13.6
Marital status
Never married 34 31.5 13 10.4 47 20.2
Living together
a
9 10.3 3 2.4 12 9.8
Ever married
b
000 000
Total 43 21.6 16 6.2 59 13.3
Religion
Christianity 29 21.2 13 5.8 42 11.6
Badimo 6 25.0 1 7.7 7 18.9
No religion 8 22.9 2 9.1 10 17.5
Total 43 21.6 16 6.2 59 12.9
Source: Central Statistics Office 2004
Notes: a) Includes respondents who were cohabiting and those who were married
b) Includes respondents who were separated, divorced and widowed
Free download from www.hsrcpress.ac.za
Findings: Central Serowe District
13
The pattern of the responses shows that the most reproductively and economically active
age groups, 15–24 years and 25–49 years, demonstrated more of this risky behaviour than
the other age groups among both sexes. Table 4.5 indicates that 18.1% of respondents
aged 15–24 years (32.7% of males and 8.0% of females) and 11.9% of respondents aged
25–49 years (19.4% of males and 6.4% of females) reported having had more than one
sexual partner in the past twelve months.
Analysis of multiple sexual partners in the last twelve months by education levels
shows that those who had acquired higher educational levels recorded high numbers

(22.7%; 34.3% of males and 9.7% of females), followed by those had a secondary school
education (13.4%; 25.0% of males and 6.6% of females). Having more than one partner
in the last twelve months was most common among respondents who were never
married (20.2%; 31.5% of males and 10.4% of females), followed by those who were
living together (9.8%; 10.3% of males and 2.4% of females). Table 4.5 further shows that
having more than one sexual partner in the last twelve months was most common among
those who believed in Badimo (ancestors) (18.9%; 25.0% of males and 7.7% of females),
followed by those who had no religion (17.5%; 22.9% of males and 9.1% of females)
and was lowest among those who were Christians (11.6%; 21.2% of males and 5.8% of
females).
Substance abuse
Table 4.6 below presents the results for respondents aged 10–64 years who had ever had
alcohol in their lifetime. The table also reflects the total sample for each age group. The
results show that overall, 269 respondents (173 males and 96 females) said they had ever
taken alcohol. This represents 32.7% of the sample. Of this group, there were more males
(47.0%) than females (20.6%) who have taken alcohol in their life.
Table 4.6: Percentage of respondents aged 10–64 years who had ever had alcohol in their lifetime
Age group Males (N = 368) Females (N = 465)
n % Total n % Total
10–14 4 6.6 61 0 0 70
15–24 44 35.8 123 30 22.9 131
25–49 100 66.7 150 51 23.7 215
50+ 25 73.5 34 15 30.6 49
Total 173 47.0 368 96 20.6 465
Source: Central Statistics Office 2004
The results show that 6.6% of the 10–14 year-old males had already taken alcohol in their
life, while none of the females in the same category reported having ever taken alcohol.
The results further show that 35.8% of males and 22.9% of females aged 15–24 years
had taken alcohol in their lifetime. This is evidence that underage drinking occurs in this
district and that some people started drinking alcohol at a very early age. The numbers

of people who had ever taken alcohol in their life increased with age for both males and
females: for example, 66.7% of males aged between 25–49 years and 73.5% of males aged
50 and above reported taking alcohol. This was similar among females where 23.7% of
those aged between 25–49 years and 30.6% of those aged 50 and over reported taking
alcohol. The numbers were higher among males than among females.
Free download from www.hsrcpress.ac.za
Household survey of behavioural risks and HIV sero status
14
Awareness of and access to social and medical services
Respondents were asked if they were aware of the social and medical services available
in their area. Results show that the majority of the respondents were aware of all the
named services, with females consistently being more aware of the services than their
male counterparts (see Table 4.7 below).
Table 4.7: Awareness of social and medical services in the area by sex
Programme Males (N = 360) Females (N = 449) Total (N = 809)
n% n % n %
Home-based care 225 62.5 340 75.7 565 69.8
Orphan care 245 68.1 358 79.7 603 74.5
PLWHA support 146 40.6 232 51.7 378 46.7
Destitute programme 273 75.8 375 83.5 648 80.1
ARV programme 177 49.2 264 58.8 441 54.5
PMTCT programme 205 56.9 321 71.5 526 65.0
IPT programme
a
155 43.1 230 51.2 385 47.6
Source: Central Statistics Office 2004
Note: a) Isoniazid Preventive Therapy (IPT) is a programme for people who are HIV positive but have not been diagnosed with
tuberculosis. The medication they receive is intended to prevent the development of tuberculosis.
Large majorities of both males and females were aware of services such as HBC (69.8%;
62.5% of males and 75.7% of females), orphan care (74.5%; 68.1% of males and 79.7%

of females) and the destitute programme (80.1%; 75.8% of males and 83.5% of females).
Respondents showed the least awareness of PLWHA support, IPT and ARV treatment
programmes.
When comparing age groups, the results show that all of the age groups surveyed were
aware of the social and medical services in their area to some extent (see Table 4.8
below).
Table 4.8: Awareness of social and medical services in the area by age group
Age
group
HBC Orphan
care
PLWHA
support
Destitute
programme
ARV PMTCT IPT
N%N %N % N % N % N%N%
10–14 49 38 76 59 23 18 80 63 43 34 45 35 25 20
15–24 177 72 176 72 126 51 202 82 134 55 170 69 125 51
25–49 274 77 283 80 184 52 298 84 216 61 261 73 195 55
50+ 65 81 68 85 45 56 68 85 48 60 50 63 40 50
Source: Central Statistics Office 2004
Free download from www.hsrcpress.ac.za
Findings: Central Serowe District
15
Table 4.8 further shows that more adults than youth were aware of these services.
Awareness of these services among youth aged 15–24 years was quite high, especially for
the HBC and orphan care programmes (both 72%) and the destitute programme (82%).
Awareness of the PLWHA support, IPT and ARV programmes was lowest for all
age groups.

Despite the figures showing awareness about these services, access was very low as
shown in Table 4.9 below. It is not surprising that only one male accessed the PMTCT
programme, as it is mostly believed that women are the ones who benefit from the
programme.
Table 4.9: Accessing of social and medical services in the area by sex
Service Males (N = 296) Females (N = 377) Total (N = 673)
n% n % n %
Home-based care 13 4.4 24 6.0 37 5.3
Orphan care 16 5.4 22 5.5 38 5.5
PLWHA support 9 3.0 8 2.0 17 2.4
Destitute programme 16 5.4 36 9.0 52 7.5
ARV programme 2 0.7 16 4.0 18 2.6
PMTCT programme 1 0.3 19 4.8 20 2.9
IPT programme 2 0.7 8 2.0 10 1.4
Source: Central Statistics Office 2004
Table 4.10 below provides accessing of social and medical services by age group. The
table shows that very few children younger than 15 years had accessed HBC (3.2%),
orphan care (7.4%) destitute (10.5%) and PLWHA support (1.1%) programmes.
Table 4.10: Accessing of social and medical services in the area by age group
Age
group
(years)
HBC Orphan
care
PLWHA
support
Destitute
programme
ARV PMTCT IPT
N%N% N %N % N %N%N%

10–14 3 3.2 7 7.4 1 1.1 10 10.5 0 0 0 0 0 0
15–24 9 4.2 14 6.5 7 3.2 14 6.5 2 .9 10 4.6 1 0.5
25–49 20 6.3 12 3.8 8 2.5 20 6.4 12 3.8 7 2.2 9 2.9
50+ 5 7.0 5 7.0 1 1.4 8 11.3 4 5.6 3 4.2 0 0
Total 37 5.3 38 5.5 17 2.4 52 7.5 18 2.6 20 2.9 10 1.4
Source: Central Statistics Office 2004
In the community survey, respondents were asked about the type of support that was
received by PLWHA in their community (see Table 4.11 on page 16).