THE IMPACT OF A BUDDY SYSTEM ON THE SELF-CARE BEHAVIOURS OF WOMEN LIVING WITH HIV / AIDS IN BOTSWANA

A needs assessment done among HIV-positive (HIV+ve) people in Botswana in 2000 indicated that these people required social support. Based on these results, a buddy system for and by HIV+ve women was instituted in Botswana during 2002. This study examined the impact of the buddy system on the self-care behaviours of 116 HIV+ve women volunteers who used the services of COCEPWA (Coping Centre for People with Aids) during 2002. The convenience sample comprised 39 buddies who completed the buddy training programme, 39 patients assigned to the 39 buddies and 38 controls who lived in areas where the buddy programme did not operate. The results indicate that HIV+ve patients who had buddies showed improved self-care behaviours from April 2002 until November 2002 compared to the controls. These self-care behaviours encompassed informing a number of other people about their HIV+ve status, compliance with tuberculosis treatment, CD4 quantification and adherence to antiretroviral therapy. Although the differences were not always statistically significant, the patients showed greater improvements than the controls in all self-care behaviours. Thus the buddy system might have assisted and empowered the patients to achieve higher levels of self-care behaviours than the controls.


INTRODUCTION
During 2004, the Joint United Nations Programme on HIV/AIDS (UNAIDS) reported that globally an estimated 39.4 million people were living with HIV/AIDS, 4.9 million people became newly infected with HIV and 3.1 million people died of AIDS.Sub-Saharan Africa (SSA) is the worst hit region with 60.0% (25.4 million) of all people living with HIV/AIDS, although SSA has only 10.0% of the world's population (UNAIDS, 2004a:1).
Botswana is a middle-income SSA country with a per capita gross domestic product of US$3 300 and an estimated sustained growth rate of 9.0% per annum (UNAIDS, 2004b:1).HIV/AIDS affects 38.8% of Botswana's economically active population and the prevalence of HIV among pregnant women aged 15-49 years was 35.4% during 2000 (NACA, 2002:49;UNDP, 2000:7).These HIV/AIDS statistics imply that an estimated 330 000 people were living with HIV/AIDS in Botswana at the end of 2002 (UNAIDS, 2004b:1).Based on the estimated HIV prevalence of 35% in 2001, and the hypothesis that approximately 30% of the HIV positive (HIV+ve) population would have CD4 counts below 200 mm 3 , and that 10% of the population with CD4 counts above 200 mm 3 would have an AIDS-defining illness, it was estimated that the demand for anti-retroviral therapy (ART) would be about 110 000 people in Botswana in 2001(De Korte, Mazonde & Darkoh, 2004:2).
People who start using anti-retroviral drugs (ARVs) but who fail to adhere to their regimens will not derive the maximum benefits from ARVs, might develop full-blown AIDS, and spread HIV mutations that are resistant to ARVs, putting their sex partners and communities at risk of becoming HIV+ve with ARV-resistant strains.
Adherence to ARVs is influenced by cultural traditions and practices as well as by peer group influences and health care workers' attitudes (Ritzenthaler, 2005:1).
The Botswana government has instituted a number of services to address the HIV/AIDS epidemic, including supplying home-based care for terminally ill patients, providing prevention of mother-to-child transmission (MTCT) of HIV as well as voluntary counselling and testing (VCT) services.ARVs have been supplied in Botswana since 2001(Zuyderduin, 2004:7-8).Botswana's national ART programme is known as the MASA, meaning "a new dawn", programme.

The implementation of a buddy system in Botswana
Under the auspices of COCEPWA (Coping Centre for People with AIDS), a needs assessment was done among people living with AIDS/HIV (PLWA/H) in Botswana (Zuyderduin, Ehlers & Van der Wal, 2007:39).
Using the results of this Botswana needs assessment, supported by previous research reports emphasising the importance of social support to PLWA/H, 39 HIV+ve women volunteers were trained as buddies at the beginning of 2002 by COCEPWA.This training included basic HIV/AIDS knowledge, compliance with tuberculosis (TB) treatment, CD4 cell counts, ARV adherence and support skills.The buddies were paid a stipend of BWP50 per month and were expected to devote 4-6 hours weekly to their patients.Buddies were not expected to implement directly observed treatment of their patients' TB pills or ART on a daily basis.However, the buddies had to offer support and guidance about adherence to all treatment regimens.The buddies' strength lay in expressing care and concern for other women living with HIV.That is why the buddy programme stressed the aspect of care and support for and by women living with HIV.
Implementing and sustaining a buddy programme for and by HIV+ve women was an expensive exercise.Thus it became important to determine whether the buddy system had any impact on the self-care behaviours of HIV+ve patients, who were supported by trained buddies.

DEFINITION OF KEY TERMS
Adherence refers to the "degree to which a patient follows drug schedules; a synonym for compliance; implying a steady propensity to stick to a prescribed therapeutic regimen" (Murphy & Canales, 2001:176).
In this case adherence refers to the patients' compliance with their prescribed ARV and/or TB drug regimens.
Anti-retroviral drugs (ARVs) are pharmacological agents that act on retroviral infections, improving the symptoms, but not curing the disease.ARVs must be taken for the duration of the person's life.
Anti-retroviral therapy (ART) refers to the ARVs prescribed for a specific HIV+ve person.
A buddy refers to a friend or a companion.In this study a buddy was an HIV+ve woman who had successfully completed the buddy training programme and was assigned to tend to a specific HIV+ve woman (patient) in the community.
The buddy system refers to the system implemented by COCEPWA in Botswana, whereby trained buddies are assigned to specific HIV+ve women (patients).
Buddies need to provide information and support to their A patient is someone who uses specific services.In this report the term "patient" refers to an HIV+ve woman who had a specific buddy assigned to her, although the term "client" continues to be used within Botswana and COCEPWA contexts.
HIV/AIDS refers to diseases caused by the retrovirus, of the lentevirus class, with delayed onsets and protracted courses, without any cure (Kalichman, 1995:363).
Immune deficiency refers to a breakdown in parts of the immune system's functions, increasing the person's susceptibility to diseases, especially to TB (Kalichman, 1995:362).
MASA is a Setswana word meaning "new dawn" and is the name of the Botswana government programme providing ARVs and AIDS treatment and care to the people of Botswana.
People living with AIDS or HIV (PLWA/H) are individuals infected with the human immunodeficiency virus and are thus HIV+ve.
Self-care behaviours refer to behaviours that people "initiate and perform, within time frames, on their own behalf in the interests of maintaining life, healthful functioning, continuing personal development and well-being, through meeting known requisites for functional and developmental regulation" (Orem, 2001:522).In this study HIV+ve women's self-care activities refer specifically to the checking of their CD4 cell counts, using ARVs and completing TB prevention therapy.

PURPOSE, OBJECTIVES, ASSUMPTIONS AND HYPOTHESIS OF THE STUDY
The purpose of this study was to determine whether the buddy system would impact positively on the selfcare behaviours of HIV+ve women in Botswana.In order to identify these potential differences, data was gathered prior to the implementation of the buddy system (during April 2002) and again six months later (during November 2002) from 39 buddies, 39 patients and 38 controls.The specific self-care behaviours that were investigated included disclosure of HIV+ve status to others, periodic CD4 quantifications, compliance with TB treatment/prophylaxis and adhering to ARVs.
The objectives of the study were to: • identify the number of people to whom the par- The major assumption underlying this study was that the buddy system would impact positively on the selfcare behaviours of the patients.This was expected owing to the support and guidance the patients would have received from their buddies over a six-month pe-riod compared to the controls, who had no buddies.No major changes were expected in the behaviours of the buddies as they had been trained as buddies from the outset of the study.Including a control group excluded the possibility that changes in patients' self-care behaviours could be attributed to the time lapse of six months only and not necessarily to the buddies' interventions in their patients' self-care behaviours.
The hypothesis guiding this quasi-experimental design was that there would be significant changes in the patients' self-care behaviours from April to November 2002 compared with those of the controls.
The null hypothesis underlying the study was that there would be no significant changes in the patients' selfcare behaviours from April to November 2002 compared with those of the controls.

ETHICAL CONSIDERATIONS
The ethical issues pertinent to this study related to the fact that individual participants should not be harmed by the conduct, publicity and/or results of the research.
The research proposal was approved by the Research and Ethics Committee of the Department of Health Studies, Unisa.COCEPWA was fully informed of the nature and purpose of the research project, and monthly discussions were held with COCEPWA members throughout the data collection period.
Every willing participant signed a consent form, which was sealed in an envelope and placed in a container, separate from the container used for the anonymously completed structured interview schedules.In this way no signed consent form could be linked to any specific completed structured interview schedule, ensuring anonymity.The participants were assured that they would remain entitled to the health care services whether or not they participated in the research project, and that they could withdraw their participation at any stage without suffering any ill effects whatsoever.Every participant's level of HIV disclosure was respected.Each buddy signed a written agreement about her rights and responsibilities.A COCEPWA supervisor monitored the buddy-patient pairs and a telephone line was available for handling any queries.
No names were entered on the structured interview schedule.Each participant was assigned a unique number for later correlation of the two sets of data, obtained during April and November 2002.This information was kept locked up and available only to one researcher until its destruction after the approval of the research report.

RESEARCH DESIGN AND METHODS
A quasi-experimental descriptive survey was used to explore the impact of the buddy system on the selfcare behaviours of HIV+ve women in Botswana.

Population and sample
Only 39 women volunteered to be trained as buddies.
COCEPWA's policy is to assign only female patients to female buddies.Thus the population for this study comprised all HIV+ve women in Botswana.However, this population is unknown and invisible, without any existing census of these women.Each HIV+ve COCEPWA member was approached to participate in the study and to invite any other HIV+ve woman to participate, amounting to snowball sampling.All participants had to be older than 18, able to sign consent to participate in the study, and had to have disclosed their HIV+ve status to at least one other person.
During 2002, when the data was collected, the buddy programme was available only in selected towns in Botswana.Of the 116 participants, 39 volunteered to be trained as buddies, who were willing to be disclosed as being HIV+ve women.Subsequent to completing her training, each buddy was paired with a patient.The

Research instrument
Initially it was the intention to use a questionnaire to collect data as it could be completed in total privacy, ensuring the anonymity of the respondents (Breakwell, Hammond & Fife-Schaw, 2000:161).However, during the pretest phase, all the women requested the research assistant to remain nearby and to elaborate on some questions.The research assistants ended up entering the women's responses on the questionnaires as they were familiar with the questionnaire's layout.This amounted to conducting structured interviews.Every structured interview commenced by asking demographic questions, enabling the interviewer and the participant to become familiar with each other and facilitating the rest of the interview process.
The structured interview schedule, using Likert scales, comprised seven items requesting information about demographics and self-care behaviours.The items were selected based on a literature review, and on the four self-care behaviours selected for this study, namely disclosure of HIV+ve status to others, adherence to TB and ARV regimes and having CD4 cell counts done at regular intervals.

Validity and reliability
Validity refers to the extent to which an instrument "actually reflects the abstract construct being measured" (Burns & Grove, 2001:814).The employees of COCEPWA agreed that the items included in the structured interview schedule requested information about the interviewees' demographics, or their self-care behaviours relevant to disclosure of HIV+ve status, adhering to ART and/or TB treatment and having regular checks done of CD4 counts.Two researchers and a statistician also examined the structured interview schedule's items and declared that every item was relevant to self-care behaviours of HIV+ve women.During pretesting the instrument, the HIV+ve women requested that the English terminology be included in addition to the Setswana terminology.This was done to enhance the women's understanding of each question.
Reliability refers to the consistency with which an instrument measures a specific construct (Burns & Grove, 2001:809).The same structured interview schedule was used during the pretest before the actual data collection commenced, prior to the interaction of buddies and patients in April 2002, and again six months into the buddy-patient relationships in November 2002.The scores obtained for the four different types of self-care behaviours did not change significantly for the buddies and controls over the six-month period, although many scores changed for the patients.Thus the structured interview schedule was stable over the six-month pe-riod.Four trained interviewers conducted the structured interviews.The completed structured interview schedules were checked on a daily basis by the first author for completeness and accuracy of reported data.No differences could be attributed to specific interviewers, and therefore the instrument was assumed to have had inter-rater reliability.
The Kruskal-Wallis test is the most powerful non-parametric test for detecting differences between three or more groups (Burns & Grove, 2001:579, 589).The Kruskal-Wallis test results indicated that there were differences between the buddies, patients and controls in this study (Zuyderduin, 2004:Annexure VI), confirming the structured interview schedule's reliability.

Data collection process
The structured interview schedule was translated into Setswana.Four structured interviews were conducted with COCEPWA volunteers to pretest the instrument.
It took 45 minutes to conduct one interview and to write down the verbatim responses.The major recommendation of the women who participated in the pretest was that the English terms for HIV/AIDS issues also be included, as these were better known to respondents than the Setswana terms.This was done.

Analysis of data
The answers recorded on the structured interview schedules were transposed as numbers to allow for a com-

Demographic information
The ages of the 116 women ranged from 20 to 52 with an average age of 31.16 years (SD 7.18).The average age of the buddies was three years higher than that of the patients and controls.Of the participants, only 13.8% (n=16) were married, 3.4% (n=4) were widows, while 25.9% (n=30) were single and 56.9% (n=66) were unmarried but had partners.The number of children that the HIV+ve women had varied between 0 and 6 (M=1.85,SD=1.37).Twenty-six HIV+ve women had attended at least primary school (22.4%; n=116) and 56 (48.3%) had secondary school education up to Junior Certificate level, which is equivalent to three years of secondary schooling.Twenty-one HIV+ve women (18.1%) had completed their secondary education up to O Level (ordinary or school end certificate level).Post-secondary education, college or vocational training had been achieved by 11 (9.5%)HIV+ve women, and 2 (1.7%) HIV+ve women had been to university.
In April 2002, 67 HIV+ve women (57.8%; n=116) earned less than BWP200 per month.The patients seemed less well off, with nobody in this group earning more than BWP1 500 and 24 (60.5%)patients earning BWP200 or less.

HIV+ve women's self-care behaviours
To explore the self-care behaviours before the patients had been linked up with their buddies and to assess whether self-care behaviours improved six months into the buddy relationship, questions were asked about HIV testing and disclosure of positive results to others, compliance with TB treatment, CD4 cell count quantifications and adherence to ART.

Disclosure of HIV positive test results
The HIV+ve women were asked to state the number of people to whom they had disclosed their HIV+ve status in April 2002 and again in November 2002.
By April 2002, of the 116 HIV+ve women, 50% (n=58) had disclosed their HIV+ve status to between 1 and 5 people and 17.2% (n=20) had disclosed to 31 or more people.More buddies had disclosed to more people than the other groups because 11 buddies (28.2%) had disclosed to 31 or more people.
There was an increase in the number of people to whom the participants' HIV+ve status had been disclosed after six months; the frequency of the lowest category (1 to 5) was only 24.3% (n=27), and the frequency of the highest category (31 or more) had increased from 17.1% (n=19) to 30.6% (n=34).The buddies were more open about their HIV+ve status than the controls and the patients in April 2002.Over the six-month period, 17 of the 34 buddies (50%) remained in the same category, but the other 17 buddies (50%) moved to a higher category, meaning that they had disclosed their status to substantially more people by November 2002.

Buddies
Of the 38 controls, 42.1% (n=16) had disclosed their HIV+ve status to 1 to 5 people, 60.5% (n=23) remained in the same category and 39.5% (n=15) moved to a higher category, having disclosed their HIV+ve status to more people by November 2002.
The 39 patients' disclosure levels also increased, with 48.7% (n=19) moving to a higher category and 51.3% (n=20) remaining in the same category.The controls and patients still had more participants in the 1 to 5 group than the buddies, and fewer in the group of 31 or more people.
The Pearson chi-square test (value 17.168; p=.009, twosided) was used to compare the three groups' disclosure status.The difference was significant at the 1% level of significance.The buddies and the patients were more open about disclosing their HIV+ve status to others between the first and second surveys than the controls, with fewer buddies and patients having told five or fewer people compared to the controls.Having a buddy resulted in increased levels of disclosure for 51.3% (n=20) of the patients.Not having a buddy did not prevent the controls from disclosing their HIV+ve status to more people, but they did so at a slower pace than the patients.
These results suggest that participating in the buddy system encouraged disclosure of HIV+ve status to other people, but did not force the HIV+ve women to disclose if they were not ready to do so.It can be accepted that the buddy system impacted positively on the number of people the women informed about their HIV+ve status.Nevertheless, HIV+ve people in Botswana might not disclose their HIV+ve status to their loved ones and families because they might want to protect them from hearing such bad news, or because they want to protect themselves from being blamed for bringing AIDS into their homes (Tlou, 2002:655).patients.The difference was not significant (p=.707).

TB treatment compliance
Nine buddies (26.4%), 6 controls (15.8%) and 5 patients (12.8%) changed from low to high levels of selfcare behaviours between the two surveys.These differences were not significant at the 5% level of significance (p=.381).
Thus it can be concluded that being or having a buddy had no significant impact on the HIV+ve women's selfcare behaviours concerning TB treatment and prophylaxis.
Adherence to treatment or prophylaxis for TB is vital for the HIV+ve person's health and wellbeing.This is the case because TB is reportedly the leading cause of death among HIV+ve people, as more than 30% of HIV+ve people die as a result of TB (Clemens, 2003:97).
As the buddies enhanced the patients' adherence to TB treatment and prophylaxis, they could play a major role in decreasing the mortality and morbidity rates from

CD4 quantification
At the start of the study in April 2002, 66 (56.9%) out of the 116 HIV+ve women had checked their CD4 counts regularly and 50 (43.1%)had not done so.Of the 66 women who had checked their CD4 counts, 29 (74.4%) were buddies, 18 (47.4%)were controls and 19 (48.7%) were patients.This difference was significant at the 5% level of significance (p=0.026).
The responses in November 2002 indicated that CD4 count checks had increased in all three groups from 56.9% (n=66) to 82.1% (n=92), but the patients (89.7%; n=35) and buddies (88.2%; n=30) exceeded the controls (71.1%;n=27).In total, 92 out of 111 (82.1%) HIV+ve women checked their CD4 counts regularly and 19 (16.9%) did not do so by November 2002.The increase in the percentage of patients who checked their CD4 counts regularly was 80% (increased from 19 to 35 out of 39) patients but for the controls it was only 45% (increased from 18 to 27 controls out of 38) (Zuyderduin, 2004:229).The chi-square test provided no significant differences at the 5% level of significance (p=0.227).
It can be concluded that more patients than controls had CD4 counts done, although this difference was not statistically significant.Therefore it cannot be concluded that the buddy system necessarily impacted positively on the number of HIV+ve women who had regular CD4 counts done.
The optimum time to commence ART is before patients present with opportunistic infections or become generally unwell.A baseline CD4 count, when the person is diagnosed as HIV+ve, is essential for future compari-sons to judge the progress of the disease and the decline of the person's immune system.Patients with CD4 counts below 200 cells/mm 3 or with AIDS-defining illnesses should be started on ART (WHO, 2006:16).
Whether the HIV+ve person is on ART or not, it remains essential to have regular CD4 counts done to monitor the progress of the disease or the patient's reaction to ART (Anabwani & Jimbo, 2005:6-7).Thus it remains vital that every PLWA/H has regular CD4 cell counts done.Buddies can play an important part in reminding their patients to go for regular tests, to support them before, during and after the tests and to act appropriately once the CD4 results are known.

LIMITATIONS
All the participants were HIV+ve Botswana women aged 18 years or older who were members of COCEPWA.
Different results might have been obtained by studying a cross-section of the population of both sexes and different ages, and of people who were not COCEPWA members (but these constituted an "invisible population" during the data collection phase of this study).
There are no guarantees that the 116 women who vol- Interviewing other relevant people to validate each interviewee's report could have validated the research results.However, this was not done as such interviews with 'significant others' carried the risk of involuntary disclosure of the specific woman's HIV+ve status.This would have amounted to a breach of confidentiality, and was therefore not attempted.
Checking the women's verbal reports against their clini-cal data on their medical files could also have increased the validity of the findings, using triangulation of data sources.However, this could not be done as each woman interviewed was assured of anonymity and confidentiality.Rigorous steps need to be followed to obtain permission for using data on patients' files from Botswana's Medical Research Council.Such permission was not obtained for this study, and patients' files could thus not be used.

CONCLUSIONS
Based on the research results, the null hypothesis that the implementation of the buddy system had no impact on the self-care behaviours of patients was rejected.The hypothesis that the implementation of a buddy system improved patients' self-care behaviours of disclosing their HIV+ve status, adhering to TB treatment/prophylaxis, having regular CD4 counts done and adhering to ART was accepted at the .10significance level.
The buddy system translated theory into practice.By practising additional self-care behaviours, the HIV+ve women were able to participate more actively in shaping the conditions that influenced their health.The results of increased levels of disclosure, increased checks of CD4 counts and enhanced levels of self-care behaviours for TB and ARVs in the present study pointed towards the effectiveness of being a buddy and having a buddy to promote levels of self-care behaviours and possibly also enhanced levels of personal wellbeing.

RECOMMENDATIONS
HIV+ve women who adhere to their ART, and who are willing to disclose their HIV+ve status, should be re-

CONCLUDING REMARKS
The buddies, as members of their communities, are role models of positive living and have the skills to motivate their peers to be self-reliant and access HIV care services, when required."While the challenges are great, the buddy system of care and support for and by women living with HIV or AIDS has a future in a resource-limited country searching for culturally appropriate ways to reach the community with informational and emotional support" (Zuyderduin, 2004:306).
patients.COCEPWA refers to the Coping Centre for People with AIDS, a non-government organisation in Botswana that trains buddies and provides a social support network to HIV+ve women.(No VCT services are provided by COCEPWA, as these services are available free of charge at government clinics and hospitals in Botswana.)CD4 cell count is a measure indicating how much the human immunodeficiency virus has diminished a person's immune functions.Normal CD4 values range between 800 and 1 200 cells/mm 3 and patients with CD4 cell counts of lower than 200 cells/mm 3 are eligible for free ARVs in Botswana (Zuyderduin, 2004:171).
ticipants had disclosed their HIV+ve status during April and November 2002; • monitor the CD4 quantifications during April and November 2002; • obtain HIV+ve women's self-reports about their compliance with TB treatment or prophylaxis during April and November 2002; • determine whether the HIV+ve women adhered to their ART; and • identify whether differences in these behaviours occurred between the buddies, patients and controls from April 2002 to November 2002.
other 38 women were assigned to the control group, being neither buddies nor patients.Thus the sample comprised 116 women (39 buddies, 39 patients and 38 controls) in April 2002 but only 111 in November 2002 because five women died during this time.Four of the women who died used ARVs and one did not.

Four
research assistants, COCEPWA employees who were bilingual in English and Setswana, knowledgeable about HIV/AIDS and known to the COCEPWA patients, were trained to conduct the structured interviews.These interviews were conducted with the buddies, patients and controls during April 2002, prior to the assignment of buddies to patients, in order to gather baseline data, and again in November 2002, six months after the buddy-patient pairs had been established.The interviews were conducted at a time convenient for each participant in a private room at COCEPWA, as visits from COCEPWA staff members at patients' homes might have indicated their HIV+ve status to their family members, friends and/or neighbours.No interviewee agreed to being audiotaped.The first author checked every completed structured interview schedule immediately after its completion and any ambiguities were clarified with the interviewer and/or interviewee.
puter-based analysis, using the Statistical Package for the Social Sciences (SPSS) version 12, with the help of a statistician.The variables of interest were the scores of the buddies, patients and controls at baseline in April 2002, and six months later in November 2002, specifically pertaining to the disclosure of their HIV+ve status, the adherence to TB and ART regimens, and the testing of CD4 cell counts.
unteered to participate represented all women living with HIV in Botswana, because of the non-random (snowball) sampling method used.HIV+ve women who had not disclosed their status to others might have had different types of self-care behaviours which remained unidentified during this study.Two possible mistakes might have been made concerning the null hypothesis: rejecting a true null hypothesis, or accepting a false null hypothesis because the sample size of 116 was small and because the probability level of .10implied that there was a 10.0% risk of committing one of these errors(Burns & Grove,   2001:815).These statistical considerations prevent the generalisation of the research results to the population of HIV+ve women in Botswana.Different results might have been obtained by conducting in-depth qualitative interviews, but none of the HIV+ve women who participated in this study would have agreed to being tape-recorded.Focus group interviews were out of the question as this would have implied disclosing respondents' HIV+ve status to every participant in the focus group.
cruited and trained to become buddies to other PLWA/ H in Botswana.Trained buddies should participate in the training of future buddies.A longitudinal study should be conducted to monitor the potential impact of the buddy system on the selfcare behaviours of HIV+ve patients in Botswana.Based on these results, ongoing education and training should be provided to current and future buddies.Longitudinal HIV-related quality of life studies should be conducted over a number of years.Although ART can retard the disease progression and extend the patient's life, ART also has unpleasant side effects that could potentially erode the PLWA/H's quality of life.If effective remedies for specific unpleasant side effects could be substantiated by well-documented research, these remedies might enhance the person's quality of life, lengthened by ART.Future policy planners should consult PLWA/H, and especially trained buddies, about these people's needs.Future research should focus on the consequences of telling people about their HIV+ve status in Botswana and on ways in which PLWA/H in Botswana cope with the double stigma of being HIV+ve and suffering from TB. Research-based ongoing community education should be conducted in Botswana informing people about HIV/ AIDS and TB.