This article was downloaded by: [Eindhoven Technical University] On: 17 November 2014, At: 22:28 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
AIDS Care: Psychological and Socio-medical Aspects of AIDS/HIV Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/caic20
Social networks and mental health among people living with human immunodeficiency virus (HIV) in Johannesburg, South Africa a
Willis Omondi Odek a
Department of Sociology, King's College, University of Aberdeen, Aberdeen, UK Published online: 01 Apr 2014.
To cite this article: Willis Omondi Odek (2014) Social networks and mental health among people living with human immunodeficiency virus (HIV) in Johannesburg, South Africa, AIDS Care: Psychological and Socio-medical Aspects of AIDS/HIV, 26:8, 1042-1049, DOI: 10.1080/09540121.2014.902421 To link to this article: http://dx.doi.org/10.1080/09540121.2014.902421
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
AIDS Care, 2014 Vol. 26, No. 8, 1042–1049, http://dx.doi.org/10.1080/09540121.2014.902421
Social networks and mental health among people living with human immunodeficiency virus (HIV) in Johannesburg, South Africa Willis Omondi Odek* Department of Sociology, King’s College, University of Aberdeen, Aberdeen, UK
Downloaded by [Eindhoven Technical University] at 22:28 17 November 2014
(Received 5 August 2013; accepted 4 March 2014) People living with human immunodeficiency virus (PLHIV) in developing countries can live longer due to improved treatment access, and a deeper understanding of determinants of their quality of life is critical. This study assessed the link between social capital, operationally defined in terms of social networks (group-based and personal social networks) and access to network resources (access to material and non-material resources and social support) and health-related quality of life (HRQoL) among 554 (55% female) adults on HIV treatment through South Africa’s public health system. Female study participants were involved with more group-based social networks but had fewer personal social networks in comparison to males. Access to network resources was higher among females and those from larger households but lower among older study participants. Experience of social support significantly increased with household economic status and duration at current residence. Social capital indicators were unrelated to HIV disease status indicators, including duration since diagnosis, CD4 count and viral load. Only a minority (13%) of study participants took part in groups formed by and for predominantly PLHIV (HIV support groups), and participation in such groups was unrelated to their mental or physical health. Personal rather than group-linked social networks and access to network resources were significantly associated with mental but not physical health, after controlling for sociodemographic characteristics. The findings of limited participation in HIV support groups and that the participation in such groups was not significantly associated with physical or mental health may suggest efforts among PLHIV in South Africa to normalise HIV as a chronic illness through broad-based rather than HIV-status bounded social participation, as a strategy for deflecting stigma. Further research is required to examine the effects of HIV treatment on social networking and participation among PLHIV within both rural and other urban settings of South Africa. Keywords: social capital; people living with HIV; South Africa; mental health; health-related quality of life
Background With improved treatment access, people living with human immunodeficiency virus (PLHIV) in developing countries can live longer and even attain life expectancy similar to that of the general population (Roehr 2013; van Sighem et al., 2010). There is, therefore, need for a deeper understanding of determinants of quality of life among PLHIV in developing countries. The concept of quality of life is used to refer to the assessment of both objective state of societies (populationlevel) and subjective state of individuals (individual-level) (Rapley, 2003). At the individual level, quality of life is defined in generic terms and in reference to specific outcomes, such as health. The World Health Organisation (WHO, 1997, p. 1) describes overall quality of life as a multidimensional construct that is affected by a “person’s physical health, psychological state, level of independence, social relationships, personal beliefs and their relationship to salient features of their environment”. The United States Centres for Disease Control and Prevention (CDC, 2000) considers health-related quality of life *Emails: [email protected]; [email protected] © 2014 Taylor & Francis
(HRQoL) to include aspects of overall quality of life that can clearly be shown to affect physical or mental health. Wilson and Cleary (1995, 1996) and Brenner, Curbow, and Legro (1995) developed conceptual models, which identify determinants of HRQoL at proximal and distal levels. The proximal factors include biological and physiological conditions and captures such variables as disease symptoms, general bodily functioning and psychological state. Distal factors are generally external to health care and include social characteristics such as socio-economic status, social integration and social support. Social capital may be an external factor that influences quality of life, especially among PLHIV. This paper seeks to expand understanding of the link between social capital and HRQoL among people on HIV treatment through South Africa’s public health system. South Africa has an estimated 5.6 million PLHIV with women comprising nearly two-thirds of them. The estimated HIV prevalence in young people aged 15–24 years in the country is 12.7% and 4.0% among young females and males, respectively. Nearly a million (36%) of the people in need of HIV treatment in South Africa
Downloaded by [Eindhoven Technical University] at 22:28 17 November 2014
AIDS Care have access, primarily through the public health system (Republic of South Africa [RSA] (2010; UNAIDS, 2012). South Africans on HIV treatment through the public health system are mostly unemployed, with an estimated 40% of them falling in the lowest socio-economic status quintile (Cleary et al., 2011; Wouters et al., 2010). Social capital is a contested concept both conceptually and in empirical application (Hawe & Shiell, 2000; Macinko & Starfield, 2001; Muntaner & Lynch, 2002; Woolcock, 2010). Portes (1998) attributed the first systematic contemporary theorisation of social capital to a French sociologist, Pierre Bourdieu. Social capital in Bourdieu’s view consists of actual or potential resources that individuals can appropriate from networks in which they are members (Bourdieu & Wacquant, 1992). Putman, Leonardi, and Nanetti (1993, p. 167) defined social capital as “features of social organisation such as trust, norms, and networks, that can improve the efficiency of society by facilitating coordinated actions”. Putman (1995, p. 665) observed that social capital is higher in areas where people have “connections with the life of their communities”, or civic engagement. Putman’s perspective constitutes the social cohesion or communitarian school of social capital as contrasted to Bourdieu’s networks school (Kawachi, Subramanian, & Kim, 2008). For the present study, social capital is conceptualised from a social networks point of view and is defined as networks of social relations and their associated resources. Social capital can have spin-offs at an individual or group level (Ostrom & Ahn, 2003). Social capital is thought to be relevant to health outcomes through pathways of social support, social influence, social engagement, persons-to-person contact and access to resources (Berkman & Glass, 2000; Kawachi et al., 2008). Social relations, however, do not occur in a vacuum. In the particular case of HIV, the greatest challenge to social relations may be that of stigma. Stigma, defined by Goffman (1963, p. 13) as “an attribute that is deeply discrediting”, varies by whether the condition attracting negative stereotyping or prejudice is visible or concealable. People with a visible condition suffer stigma from being discredited, while those with a concealable condition, such as living with HIV (for some people), suffer stigma from their knowledge of possessing a discreditable identity and face the challenge of how to manage that information (Goffman, 1963). Goffman (1963) and Pachankis (2007) note that stigma associated with concealable conditions can have psychological and behavioural implications such as suspiciousness and social avoidance, which can compromise social relations and subjective well-being. Previous studies in Southern Africa have reported relationships between social capital and risks for HIV and adherence to antiretroviral treatment among PLHIV
1043
(Campbell et al., 2013; Pronyk et al., 2008a, 2008b). However, there has been limited attention to the potential role of social capital in the adjustment to living with HIV. This paper addresses this gap, and further examines if social capital varies by gender and is related to objective indicators of HIV disease status, such as CD4 count, viral load and duration since diagnosis, among PLHIV. Methods Study setting This study was conducted among the PLHIV and receiving treatment at two public health facilities in Gauteng Province, South Africa, namely, Chris Hani-Baragwanath and Tembisa Hospitals. Chris Hani-Baragwanath Hospital is South Africa’s largest national referral hospital while Tembisa is a provincial hospital. Study design The study was of a cross-sectional analytical design. Study participants were followed retrospectively for changes in treatment indicators (CD4 count and viral load) since enrolment into treatment. Additional crosssectional data on HRQoL and social capital were collected. Eligibility criteria The study inclusion criteria were as follows: (1) (2) (3) (4) (5) (6)
Two or more years of enrolment in HIV treatment; Aged between 18 and 64 years at first diagnosis with HIV; Availability of longitudinal data on HIV treatment performance; Living within the study hospitals’ catchments; Ability to provide additional data by self rather than through a proxy; and Able to freely consent to participate in the study.
Recruitment Chris Hani-Baragwanath Hospital and Tembisa Hospital had three and two designated clinic days per week, respectively, for patients on antiretroviral treatment, with approximately 100 patients seen on each of the days. The recruitment process involved general announcements to patients at the waiting area on the designated clinic days to invite those who were willing to take part in the study. Participants who came forward after the announcements received in-depth information on the study following a standardised information and consenting process. Interviews were conducted at the hospitals in the privacy of
1044
W.O. Odek
an office with consenting participants. The study’s recruitment process and interviews with consenting participants were conducted from April to November 2010. Ethics clearance
Downloaded by [Eindhoven Technical University] at 22:28 17 November 2014
The University of Witwatersrand’s Human Research Ethics Committee, South Africa, reviewed and approved the study. Sampling The overall objective of the study, which guided the sampling strategy, was to compare HRQoL, based on the validated Medical Outcomes Study 36-Item Short Form (SF-36) (Quality Metrics Incorporated) between PLHIV in formal employment and those not formally employed. The study sample calculated based on this objective was 568 and the sample attained was 554 (55% female).
education, community health and professional groups). Respondent’s own intensity of group participation was assessed in three categories: (i) an ordinary member (scored 1); (ii) an active member (scored 2) and (iii) a leader (scored 3). Following from an application by Pronyk et al. (2008b), a group participation intensity score was calculated for each respondent as follows: Group participation intensity ¼ b1ðg1 Þ þ b2ðg2 Þ þ bi ðgi Þ þ . . . ði ¼ 1, . . .11Þ where βi is the multiplier (range 1–3) for intensity of membership of group gi. The group participation intensity score could range from zero (0) to 33. Personal social networks were measured as the number of close friends with whom the respondent felt at ease to talk with on personal matters or could call upon for help or advice. The number of close friends ranged from none (0) to 21 in the study sample. Access to social network resources
Data collection Primary data on HRQoL, social capital, sociodemographic characteristics, including household size, marital status, duration of living at current location, current age, education, household economic status and employment status were collected using a structured questionnaire administered by five trained research assistants. Secondary treatment history data were obtained from patient files and an electronic patient database. With the exception of social capital, detailed information on the measurement of key variables, including HRQoL, can be found elsewhere (Odek, 2013). HRQoL was measured using the validated Medical Outcomes Study 36-Item Short Form (SF-36) (Ware, 2000) and is represented in the analyses by physical functioning and mental health subscale scores. According to Ware and Kosinski (2001), physical functioning and mental health subscales of the SF-36 instrument provide the purest measures of physical and mental health, respectively. Measurement of social capital Social networks The indicators of social capital for this study are social networks and access to social network resources. Social network measurement captured both group-based and personal social networks (Grootaert, Narayan, Jones, & Woolcock, 2004). Group-based social networks were assessed at the household level. Each study participant was asked about membership of their household members in 11 different kinds of groups that were either formal or informal (i.e., religious, welfare, neighbourhood, leisure, economic, HIV support, political, labour,
Access to network resources was categorised into two elements: (1) access to material and non-material resources; and (2) experience of social support. Study participants were asked about the sources from which they could obtain help in eight different situations of need, such as borrowing money and being taken to hospital when ill (International Social Survey Programme, 2001). Respondents were asked to give the forename of a person they could turn to in the first instance if they needed help in each of the identified situations and their relationship to that person. The responses were collapsed into a dichotomous variable with “no source of help” as 0 and “family or non-family source of help” as 1. These dichotomous responses were summed up to derive an access to network resources score with a range from 0 to 8. Experience of social support was measured as the number of people with whom the respondent had shared matters of personal importance in the previous 12 months. This number ranged from 1 to 11 in the study sample. Figure 1 depicts the hypothesised relationship of these social capital indicators and other background characteristics with HRQoL.
Analysis Data were analysed with SPSS Statistics 18 software. Simple linear regressions were conducted to examine the relationship between social capital indicators and study participants’ background characteristics. Multiple linear regressions controlling for study participants’ background characteristics, were used to examine the relationship between social capital and HRQoL. A 95%
AIDS Care
1045
Socio-demographic characteristics - Age, education, marital status, household size, gender, economic status, etc
Social networks - Group social networks - Group participation intensity - Personal social networks Health-related quality of life - Physical functioning - Mental health
Downloaded by [Eindhoven Technical University] at 22:28 17 November 2014
Social capital Access to network resources - Material and non-material resources - Experience of social support
HIV treatment indicators - Duration since diagnosis - CD4 count - Viral load
Figure 1. A conceptual framework of the relationship between social capital and HRQoL.
statistical significance level was set for all analyses. Diagnostic checks, including multicollinearity and model fit were conducted.
Results
households, but lower among older study participants. Experience of social support significantly increased with household economic status, duration at current residence and education. Both social networks and access to network resources indicators were unrelated to HIV disease status indicators (Table 2).
Table 1 shows study participants’ background characteristics, stratified by gender. Significant gender differences were found in formal employment status, age, length of residence at current location, marital status and household size. Female study participants had been diagnosed with HIV for a mean of one year longer than males, while male study participants had a significantly lower CD4 count than their female counterparts. The study further assessed the relationship between social capital and study participants’ sociodemographic characteristics and HIV disease status indicators. Female study participants were involved with more groups but had fewer personal social networks in comparison to males. Older study participants had more personal social networks but group participation did not vary significantly by age. Access to network resources was higher among female study participants and those from larger
Social capital and HRQoL Table 3 presents a univariate analysis of HRQoL indicators, while Table 4 shows simple (unadjusted) and multiple (adjusted) linear regression coefficient estimates for the association between social capital and HRQoL, represented by the SF-36 physical functioning and mental health subscale scores. Social capital indicators had no significant association with physical functioning in both bivariate and multivariable analysis. Personal social networks, access to material and nonmaterial resources and experience of social support were significantly associated with better mental health after controlling for study participants’ sociodemographic and economic characteristics (Table 4). The present study also analysed the link between participation in groups formed by and for predominantly
1046
W.O. Odek
Table 1. Sociodemographic and economic characteristics and HIV disease status indicators, stratified by gender. Gender Characteristics
Male (n = 251)
Female (n = 303)
p Value
38.6%
48.5%
.020
127 (50.6%) 124 (49.4%) 42.7 (7.8)
154 (50.8%) 149 (49.2%) 39.0 (8.19)
.957
AIDS Care: Psychological and Socio-medical Aspects of AIDS/HIV Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/caic20
Social networks and mental health among people living with human immunodeficiency virus (HIV) in Johannesburg, South Africa a
Willis Omondi Odek a
Department of Sociology, King's College, University of Aberdeen, Aberdeen, UK Published online: 01 Apr 2014.
To cite this article: Willis Omondi Odek (2014) Social networks and mental health among people living with human immunodeficiency virus (HIV) in Johannesburg, South Africa, AIDS Care: Psychological and Socio-medical Aspects of AIDS/HIV, 26:8, 1042-1049, DOI: 10.1080/09540121.2014.902421 To link to this article: http://dx.doi.org/10.1080/09540121.2014.902421
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
AIDS Care, 2014 Vol. 26, No. 8, 1042–1049, http://dx.doi.org/10.1080/09540121.2014.902421
Social networks and mental health among people living with human immunodeficiency virus (HIV) in Johannesburg, South Africa Willis Omondi Odek* Department of Sociology, King’s College, University of Aberdeen, Aberdeen, UK
Downloaded by [Eindhoven Technical University] at 22:28 17 November 2014
(Received 5 August 2013; accepted 4 March 2014) People living with human immunodeficiency virus (PLHIV) in developing countries can live longer due to improved treatment access, and a deeper understanding of determinants of their quality of life is critical. This study assessed the link between social capital, operationally defined in terms of social networks (group-based and personal social networks) and access to network resources (access to material and non-material resources and social support) and health-related quality of life (HRQoL) among 554 (55% female) adults on HIV treatment through South Africa’s public health system. Female study participants were involved with more group-based social networks but had fewer personal social networks in comparison to males. Access to network resources was higher among females and those from larger households but lower among older study participants. Experience of social support significantly increased with household economic status and duration at current residence. Social capital indicators were unrelated to HIV disease status indicators, including duration since diagnosis, CD4 count and viral load. Only a minority (13%) of study participants took part in groups formed by and for predominantly PLHIV (HIV support groups), and participation in such groups was unrelated to their mental or physical health. Personal rather than group-linked social networks and access to network resources were significantly associated with mental but not physical health, after controlling for sociodemographic characteristics. The findings of limited participation in HIV support groups and that the participation in such groups was not significantly associated with physical or mental health may suggest efforts among PLHIV in South Africa to normalise HIV as a chronic illness through broad-based rather than HIV-status bounded social participation, as a strategy for deflecting stigma. Further research is required to examine the effects of HIV treatment on social networking and participation among PLHIV within both rural and other urban settings of South Africa. Keywords: social capital; people living with HIV; South Africa; mental health; health-related quality of life
Background With improved treatment access, people living with human immunodeficiency virus (PLHIV) in developing countries can live longer and even attain life expectancy similar to that of the general population (Roehr 2013; van Sighem et al., 2010). There is, therefore, need for a deeper understanding of determinants of quality of life among PLHIV in developing countries. The concept of quality of life is used to refer to the assessment of both objective state of societies (populationlevel) and subjective state of individuals (individual-level) (Rapley, 2003). At the individual level, quality of life is defined in generic terms and in reference to specific outcomes, such as health. The World Health Organisation (WHO, 1997, p. 1) describes overall quality of life as a multidimensional construct that is affected by a “person’s physical health, psychological state, level of independence, social relationships, personal beliefs and their relationship to salient features of their environment”. The United States Centres for Disease Control and Prevention (CDC, 2000) considers health-related quality of life *Emails: [email protected]; [email protected] © 2014 Taylor & Francis
(HRQoL) to include aspects of overall quality of life that can clearly be shown to affect physical or mental health. Wilson and Cleary (1995, 1996) and Brenner, Curbow, and Legro (1995) developed conceptual models, which identify determinants of HRQoL at proximal and distal levels. The proximal factors include biological and physiological conditions and captures such variables as disease symptoms, general bodily functioning and psychological state. Distal factors are generally external to health care and include social characteristics such as socio-economic status, social integration and social support. Social capital may be an external factor that influences quality of life, especially among PLHIV. This paper seeks to expand understanding of the link between social capital and HRQoL among people on HIV treatment through South Africa’s public health system. South Africa has an estimated 5.6 million PLHIV with women comprising nearly two-thirds of them. The estimated HIV prevalence in young people aged 15–24 years in the country is 12.7% and 4.0% among young females and males, respectively. Nearly a million (36%) of the people in need of HIV treatment in South Africa
Downloaded by [Eindhoven Technical University] at 22:28 17 November 2014
AIDS Care have access, primarily through the public health system (Republic of South Africa [RSA] (2010; UNAIDS, 2012). South Africans on HIV treatment through the public health system are mostly unemployed, with an estimated 40% of them falling in the lowest socio-economic status quintile (Cleary et al., 2011; Wouters et al., 2010). Social capital is a contested concept both conceptually and in empirical application (Hawe & Shiell, 2000; Macinko & Starfield, 2001; Muntaner & Lynch, 2002; Woolcock, 2010). Portes (1998) attributed the first systematic contemporary theorisation of social capital to a French sociologist, Pierre Bourdieu. Social capital in Bourdieu’s view consists of actual or potential resources that individuals can appropriate from networks in which they are members (Bourdieu & Wacquant, 1992). Putman, Leonardi, and Nanetti (1993, p. 167) defined social capital as “features of social organisation such as trust, norms, and networks, that can improve the efficiency of society by facilitating coordinated actions”. Putman (1995, p. 665) observed that social capital is higher in areas where people have “connections with the life of their communities”, or civic engagement. Putman’s perspective constitutes the social cohesion or communitarian school of social capital as contrasted to Bourdieu’s networks school (Kawachi, Subramanian, & Kim, 2008). For the present study, social capital is conceptualised from a social networks point of view and is defined as networks of social relations and their associated resources. Social capital can have spin-offs at an individual or group level (Ostrom & Ahn, 2003). Social capital is thought to be relevant to health outcomes through pathways of social support, social influence, social engagement, persons-to-person contact and access to resources (Berkman & Glass, 2000; Kawachi et al., 2008). Social relations, however, do not occur in a vacuum. In the particular case of HIV, the greatest challenge to social relations may be that of stigma. Stigma, defined by Goffman (1963, p. 13) as “an attribute that is deeply discrediting”, varies by whether the condition attracting negative stereotyping or prejudice is visible or concealable. People with a visible condition suffer stigma from being discredited, while those with a concealable condition, such as living with HIV (for some people), suffer stigma from their knowledge of possessing a discreditable identity and face the challenge of how to manage that information (Goffman, 1963). Goffman (1963) and Pachankis (2007) note that stigma associated with concealable conditions can have psychological and behavioural implications such as suspiciousness and social avoidance, which can compromise social relations and subjective well-being. Previous studies in Southern Africa have reported relationships between social capital and risks for HIV and adherence to antiretroviral treatment among PLHIV
1043
(Campbell et al., 2013; Pronyk et al., 2008a, 2008b). However, there has been limited attention to the potential role of social capital in the adjustment to living with HIV. This paper addresses this gap, and further examines if social capital varies by gender and is related to objective indicators of HIV disease status, such as CD4 count, viral load and duration since diagnosis, among PLHIV. Methods Study setting This study was conducted among the PLHIV and receiving treatment at two public health facilities in Gauteng Province, South Africa, namely, Chris Hani-Baragwanath and Tembisa Hospitals. Chris Hani-Baragwanath Hospital is South Africa’s largest national referral hospital while Tembisa is a provincial hospital. Study design The study was of a cross-sectional analytical design. Study participants were followed retrospectively for changes in treatment indicators (CD4 count and viral load) since enrolment into treatment. Additional crosssectional data on HRQoL and social capital were collected. Eligibility criteria The study inclusion criteria were as follows: (1) (2) (3) (4) (5) (6)
Two or more years of enrolment in HIV treatment; Aged between 18 and 64 years at first diagnosis with HIV; Availability of longitudinal data on HIV treatment performance; Living within the study hospitals’ catchments; Ability to provide additional data by self rather than through a proxy; and Able to freely consent to participate in the study.
Recruitment Chris Hani-Baragwanath Hospital and Tembisa Hospital had three and two designated clinic days per week, respectively, for patients on antiretroviral treatment, with approximately 100 patients seen on each of the days. The recruitment process involved general announcements to patients at the waiting area on the designated clinic days to invite those who were willing to take part in the study. Participants who came forward after the announcements received in-depth information on the study following a standardised information and consenting process. Interviews were conducted at the hospitals in the privacy of
1044
W.O. Odek
an office with consenting participants. The study’s recruitment process and interviews with consenting participants were conducted from April to November 2010. Ethics clearance
Downloaded by [Eindhoven Technical University] at 22:28 17 November 2014
The University of Witwatersrand’s Human Research Ethics Committee, South Africa, reviewed and approved the study. Sampling The overall objective of the study, which guided the sampling strategy, was to compare HRQoL, based on the validated Medical Outcomes Study 36-Item Short Form (SF-36) (Quality Metrics Incorporated) between PLHIV in formal employment and those not formally employed. The study sample calculated based on this objective was 568 and the sample attained was 554 (55% female).
education, community health and professional groups). Respondent’s own intensity of group participation was assessed in three categories: (i) an ordinary member (scored 1); (ii) an active member (scored 2) and (iii) a leader (scored 3). Following from an application by Pronyk et al. (2008b), a group participation intensity score was calculated for each respondent as follows: Group participation intensity ¼ b1ðg1 Þ þ b2ðg2 Þ þ bi ðgi Þ þ . . . ði ¼ 1, . . .11Þ where βi is the multiplier (range 1–3) for intensity of membership of group gi. The group participation intensity score could range from zero (0) to 33. Personal social networks were measured as the number of close friends with whom the respondent felt at ease to talk with on personal matters or could call upon for help or advice. The number of close friends ranged from none (0) to 21 in the study sample. Access to social network resources
Data collection Primary data on HRQoL, social capital, sociodemographic characteristics, including household size, marital status, duration of living at current location, current age, education, household economic status and employment status were collected using a structured questionnaire administered by five trained research assistants. Secondary treatment history data were obtained from patient files and an electronic patient database. With the exception of social capital, detailed information on the measurement of key variables, including HRQoL, can be found elsewhere (Odek, 2013). HRQoL was measured using the validated Medical Outcomes Study 36-Item Short Form (SF-36) (Ware, 2000) and is represented in the analyses by physical functioning and mental health subscale scores. According to Ware and Kosinski (2001), physical functioning and mental health subscales of the SF-36 instrument provide the purest measures of physical and mental health, respectively. Measurement of social capital Social networks The indicators of social capital for this study are social networks and access to social network resources. Social network measurement captured both group-based and personal social networks (Grootaert, Narayan, Jones, & Woolcock, 2004). Group-based social networks were assessed at the household level. Each study participant was asked about membership of their household members in 11 different kinds of groups that were either formal or informal (i.e., religious, welfare, neighbourhood, leisure, economic, HIV support, political, labour,
Access to network resources was categorised into two elements: (1) access to material and non-material resources; and (2) experience of social support. Study participants were asked about the sources from which they could obtain help in eight different situations of need, such as borrowing money and being taken to hospital when ill (International Social Survey Programme, 2001). Respondents were asked to give the forename of a person they could turn to in the first instance if they needed help in each of the identified situations and their relationship to that person. The responses were collapsed into a dichotomous variable with “no source of help” as 0 and “family or non-family source of help” as 1. These dichotomous responses were summed up to derive an access to network resources score with a range from 0 to 8. Experience of social support was measured as the number of people with whom the respondent had shared matters of personal importance in the previous 12 months. This number ranged from 1 to 11 in the study sample. Figure 1 depicts the hypothesised relationship of these social capital indicators and other background characteristics with HRQoL.
Analysis Data were analysed with SPSS Statistics 18 software. Simple linear regressions were conducted to examine the relationship between social capital indicators and study participants’ background characteristics. Multiple linear regressions controlling for study participants’ background characteristics, were used to examine the relationship between social capital and HRQoL. A 95%
AIDS Care
1045
Socio-demographic characteristics - Age, education, marital status, household size, gender, economic status, etc
Social networks - Group social networks - Group participation intensity - Personal social networks Health-related quality of life - Physical functioning - Mental health
Downloaded by [Eindhoven Technical University] at 22:28 17 November 2014
Social capital Access to network resources - Material and non-material resources - Experience of social support
HIV treatment indicators - Duration since diagnosis - CD4 count - Viral load
Figure 1. A conceptual framework of the relationship between social capital and HRQoL.
statistical significance level was set for all analyses. Diagnostic checks, including multicollinearity and model fit were conducted.
Results
households, but lower among older study participants. Experience of social support significantly increased with household economic status, duration at current residence and education. Both social networks and access to network resources indicators were unrelated to HIV disease status indicators (Table 2).
Table 1 shows study participants’ background characteristics, stratified by gender. Significant gender differences were found in formal employment status, age, length of residence at current location, marital status and household size. Female study participants had been diagnosed with HIV for a mean of one year longer than males, while male study participants had a significantly lower CD4 count than their female counterparts. The study further assessed the relationship between social capital and study participants’ sociodemographic characteristics and HIV disease status indicators. Female study participants were involved with more groups but had fewer personal social networks in comparison to males. Older study participants had more personal social networks but group participation did not vary significantly by age. Access to network resources was higher among female study participants and those from larger
Social capital and HRQoL Table 3 presents a univariate analysis of HRQoL indicators, while Table 4 shows simple (unadjusted) and multiple (adjusted) linear regression coefficient estimates for the association between social capital and HRQoL, represented by the SF-36 physical functioning and mental health subscale scores. Social capital indicators had no significant association with physical functioning in both bivariate and multivariable analysis. Personal social networks, access to material and nonmaterial resources and experience of social support were significantly associated with better mental health after controlling for study participants’ sociodemographic and economic characteristics (Table 4). The present study also analysed the link between participation in groups formed by and for predominantly
1046
W.O. Odek
Table 1. Sociodemographic and economic characteristics and HIV disease status indicators, stratified by gender. Gender Characteristics
Male (n = 251)
Female (n = 303)
p Value
38.6%
48.5%
.020
127 (50.6%) 124 (49.4%) 42.7 (7.8)
154 (50.8%) 149 (49.2%) 39.0 (8.19)
.957
