UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Stigma and confidentiality as barriers to uptake of HIV counseling and testing for health workers in… Khan, Rabia 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2013_fall_khan_rabia.pdf [ 27.69MB ]
JSON: 24-1.0165635.json
JSON-LD: 24-1.0165635-ld.json
RDF/XML (Pretty): 24-1.0165635-rdf.xml
RDF/JSON: 24-1.0165635-rdf.json
Turtle: 24-1.0165635-turtle.txt
N-Triples: 24-1.0165635-rdf-ntriples.txt
Original Record: 24-1.0165635-source.json
Full Text

Full Text

	 ?	 ?STIGMA AND CONFIDENTIALITY AS BARRIERS TO UPTAKE OF HIV COUNSELING AND TESTING FOR HEALTH WORKERS IN 3 PUBLIC HOSPITALS IN FREE STATE PROVINCE, SOUTH AFRICA:	 ?A MIXED-METHODS STUDY	 ?	 ?by	 ?	 ?	 ?Rabia Khan	 ?	 ?	 ?B.Sc., McMaster University, 2008	 ?B.A., McMaster University, 2010	 ?	 ?	 ?	 ?A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF 	 ?THE REQUIREMENTS FOR THE DEGREE OF 	 ?	 ?	 ?MASTER OF SCIENCE	 ?	 ?	 ?in	 ?	 ?	 ?THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES	 ?	 ?	 ?(Population and Public Health)	 ?	 ?	 ?	 ?THE UNIVERSITY OF BRITISH COLUMBIA	 ?(Vancouver) 	 ?	 ? October, 2013    ? Rabia Khan, 2013 	 ? ii	 ?Abstract  Background: The HIV and AIDS epidemic has created a human resource crisis that ?has replaced financial issues as the most serious obstacle to implementing national treatment plans? (WHO 2006a: 20). To retain the existing health workforce, international guidelines promote priority access to health services for health workers (HWs) through occupationally-based HIV counseling and testing (HCT) services. Such services have been implemented in South Africa (RSA), however recent evidence suggests their uptake is low. Objective: To identify barriers and facilitators to uptake of HCT services by HWs in three hospitals in Free State province, RSA.    Methods: This mixed-methods study analyzed a portion of a self-administered survey and focus groups interviews (FGIs) to explore participants? attitudes and behaviours related to HIV in the workplace, why HIV services may be underutilized and participants? recommendations to improve the service. Results: In total, 978 HWs participated in the survey and 38 participated in the FGIs. Among survey respondents, 38.9% indicated a fear that confidentiality will not be maintained as the reason for not using OHS-based HIV services. 38.5% HWs perceive there is HIV stigma in the workplace. Six themes were identified from the FGIs, including location for testing, privacy, confidentiality, gossip, stigma and facilitators. FG participants perceived doctors? and nurses? experience with HIV in the workplace differs from other HWs, supported by multivariate analyses indicating patient-care HWs (PCHWs) have higher odds of perceiving confidentiality is not maintained in the OHS (adjusted ORs = 2.3; 95% CI 1.8-3.2) and perceiving HIV stigma in the workplace (adjusted OR = 2.4; 95% CI 1.8-3.2) when compared to non-PCHWs. FG participants 	 ? iii	 ?also identified the need for in-service training on a range of topics related to HIV and expressed a desire to form HIV support groups to address negative attitudes toward HIV/AIDS in the workplace.  Conclusions: Fear of breaches in confidentiality and HIV stigma were identified as the primary barriers to uptake of occupationally-based HCT by HWs. Overcoming these barriers require educating HWs on policies and guidelines that govern HIV in the workplace, implement measures to ensure confidentiality is maintained and addressing HIV stigma through stigma reduction interventions.      	 ? iv	 ?Preface  All research described in this dissertation was conducted under the approval of the University of British Columbia Research Ethics Boards (H10-00360, H10-01879, H12-00942) and by approval of the Free State province Department of Health, South Africa.   	 ? v	 ?Table of Contents Abstract ................................................................................................................................... ii	 ?Preface.................................................................................................................................... iv	 ?Table of Contents .................................................................................................................... v	 ?List of Tables ......................................................................................................................... vii	 ?List of Figures ...................................................................................................................... viii	 ?List of Acronyms .................................................................................................................... ix	 ?Acknowledgements................................................................................................................. xi	 ?Dedication.............................................................................................................................. xii	 ?Chapter 1: Introduction .......................................................................................................... 1	 ?1.1 Literature Review ....................................................................................................................... 1	 ?1.1.1 Human Resources for Health and the HIV and AIDS Epidemic...................................................1	 ?1.1.2 HIV Counseling and Testing in RSA ............................................................................................6	 ?1.1.3 Barriers to Uptake of HCT in RSA .............................................................................................12	 ?1.2 Rationale for the Current Study ............................................................................................... 18	 ?1.3 Objectives.................................................................................................................................. 19	 ?1.4 Methods .................................................................................................................................... 19	 ?Chapter 2: Analysis of ?HIV/AIDS and Occupational Health? Section of a Self-Administered Questionnaire ........................................................................................................................ 23	 ?2.1 Aim................................................................................................................................................ 23	 ?2.2 Method ...................................................................................................................................... 23	 ?2.2.1 Participants and Sampling...........................................................................................................23	 ?2.2.2 Procedure ....................................................................................................................................24	 ?2.2.3 Instrument and Measures ............................................................................................................24	 ?2.2.4 Data Analysis ..............................................................................................................................28	 ?2.3 Results....................................................................................................................................... 30	 ?2.3.1. Sociodemographic Characteristics .............................................................................................31	 ?2.3.2 General Perceptions Regarding HIV and AIDS ..........................................................................32	 ?2.3.3 Consequences of having HIV in the Workplace: ........................................................................34	 ?2.3.4 Perceptions of HIV Stigma in the Workplace: ............................................................................34	 ?2.3.5 Knowledge of the OHS: ..............................................................................................................36	 ?2.3.6 Perception of the OHS: ...............................................................................................................37	 ?2.3.7 Potential Reasons for not using the HCT Service: ......................................................................37	 ?2.3.8 Perceptions of Confidentiality in the OHS:.................................................................................38	 ?2.4 Summary of Findings................................................................................................................ 40	 ?Chapter 3: Focus Groups Interviews (FGIs)......................................................................... 42	 ?3.1 Aim............................................................................................................................................ 42	 ?3.2 Method ...................................................................................................................................... 42	 ?3.2.1 Sampling and Study Sites............................................................................................................42	 ?3.2.2 Focus Groups ..............................................................................................................................43	 ?	 ? vi	 ?3.2.3 Introductions and Informed Consent ...........................................................................................43	 ?3.2.4 Interview Guide ..........................................................................................................................44	 ?3.2.5 Focus Group Interview Data Analysis ........................................................................................45	 ?3.3 Results....................................................................................................................................... 47	 ?3.3.1 Theme 1: Locations for Testing ..................................................................................................48	 ?3.3.2 Theme 2: Privacy ........................................................................................................................49	 ?3.3.3 Theme 3: Gossip .........................................................................................................................52	 ?3.3.4 Theme 4: Confidentiality ............................................................................................................54	 ?3.3.5 Theme 6: Stigma .........................................................................................................................57	 ?3.3.6 Theme 6: Facilitators ..................................................................................................................67	 ?3.4 Summary of Results .................................................................................................................. 76	 ?Chapter 4: Discussion............................................................................................................ 79	 ?4.1 Accessing HCT Services............................................................................................................ 79	 ?4.2 Preference for Private Physician............................................................................................... 80	 ?4.3 Privacy and Confidentiality ...................................................................................................... 81	 ?4.4 Stigma ....................................................................................................................................... 85	 ?4.5 The Role of Professional HWs .................................................................................................. 90	 ?4.6 Recommendations Arising from Study Results ........................................................................ 93	 ?4.6.1 Privacy and Confidentiality.........................................................................................................94	 ?4.6.2 Stigma .........................................................................................................................................96	 ?4.6.3 Role Modeling ............................................................................................................................99	 ?4.6.4 Support Groups .........................................................................................................................100	 ?Chapter 5: Conclusion......................................................................................................... 102	 ?5.1 Summary................................................................................................................................. 102	 ?5.2 Limitations and Strengths of the Study .................................................................................. 103	 ?5.3 Directions for Future Research............................................................................................... 105	 ?Bibliography........................................................................................................................ 106	 ?Appendix A: Occupational Health Survey .......................................................................... 118	 ?Appendix B: Sample Frame ................................................................................................ 127	 ?Appendix C: Occupation Coding Based on Sampling Frame and PCHW vs. Non-PCHW 128	 ? 	 ? vii	 ?List of Tables  Table 1 Dependent variables included in analysis???????????????..26  Table 2 Total population, targeted quota per hospital (15% sampling frame) and achieved sample by occupation (PCHW/non-PCHW)??????????????..31  Table 3 Sociodemographic characteristics of survey participants and HWs at the study hospitals????????????????????????????..32  Table 4 Results of general perceptions regarding HIV/AIDS?????????.??33  Table 5 Association between sociodemographic characteristics and perceived HIV stigma in the workplace?????????????????????????.35  Table 6 Multivariate logistic regression demonstrating adjusted association between sociodemographic characteristics and perceptions of HIV stigma in the workplace???????????????????????????...36  Table 7 Association between sociodemographic characteristics and perceptions regarding confidentiality in the OHS?????????????????????.38  Table 8 Multivariate logistic regression of the adjusted association between sociodemographic characteristics and perception that confidentiality is not maintained in the OHS???????????????????..............39  Table 9 FGs with number of participants in each group????????????....47  Table 10 Summary of FGI results??????????????????????76     	 ? viii	 ?List of Figures  Figure 1 Proposed process of stigmatization among HWs and relationship to other barriers?????????????????????????...88  Figure 2 Relationship between barriers to uptake of HCT and professional HWs??????????????????????????...89 	 ? ix	 ?List of Acronyms   AHP(s)    Allied health professional(s) AIDS     Acquired Immunodeficiency Syndrome  ARV(s)    Antiretroviral(s) ART     Antiretroviral Treatment CICT     Client Initiated Counseling and Testing  EAP     Employee Assistance Program FG(s)     Focus group(s) FGI(s)     Focus group interview(s) HCT     HIV counseling and testing  HIV     Human Immunodeficiency Virus HRH     Human Resources for Health HSC     Health and Safety Committee HW(s)     Health worker(s) IC     Infection Control ILO     International Labour Organization  IPT     Isoniazid Preventive Therapy LRT     Likelihood Ratio Test MLR     Multivariate Logistic Regression NGOs     Non-governmental organizations OH     Occupational Health OHS     Occupational health service  P#     Participant PCHW(s)    Patient-care health worker(s) PEP     Post-exposure prophylaxis PICT     Provider-initiated counseling and testing  PLWHA    People living with HIV and AIDS PSI     Population Services International R     Rabia, interviewer  	 ? x	 ?SA     South Africa SPSS     Statistical package software for social science STI(s)     Sexually Transmitted Infection(s) T#     Transcript number  TB     Tuberculosis VCT     Voluntary Counseling and Testing  WHO     World Health Organization 	 ? xi	 ? Acknowledgements Over the years, my father has often quoted T.S. Elliot?s, ?In my beginning is my end?in my end is my beginning? to delineate time as a continuum. However, this quote does not fully capture Elliot?s vision as an ellipsis omits the hidden gems in between this first and last line of Elliot?s, ?Four Quartets?. Within the ellipsis lies, ??Our only health is the disease, if we obey the dying nurse, whose constant care is not to please, but to remind of our, and Adam's curse, and that, to be restored, our sickness must grow worse?? Through this verse, I see the mortality of this thesis - from the project itself, to the community of people who have helped shape it, and to those who have supported me through it and I realize that these ?acknowledgements? are neither an end nor a beginning.  My beginning is not a line in this thesis but a journey that reflects the love of my mother in her sincere support of all of my endeavors. I can never fully acknowledge what it has meant to have the life and love you have given me. I also want to acknowledge my father for his wisdom as it manifests in his ability to reason and ability to write, for without the two, I would never have desired to complete my ???. To my other mom and dad, thank you for letting me go and for being role models for what hard work and dedication can truly accomplish. To my Bhai and Bhabi - your patience and kindness are reasons why you were chosen to keep my sanity while I willingly gave it away.    Like a mixed-methods approach, Elliot?s ?Four Quartets? are four different poems put together after they were written, to form one cohesive product. What is rarely acknowledged is that that vision was not Elliot?s himself.  Thus, I would be remiss if I did not humbly acknowledge my committee as the architects that laid the foundation to this work. Dr. Annalee Yassi and Dr. Jane Buxton ? you have exemplified for me what it means to be strong women in research. In your own unique ways each of you have helped me define who I aspire to be within my career. Dr. David Moore and Dr. Asta Rau, your feedback and support have been invaluable and my thought process has been forever changed because of your input. Of special mention, Michelle Engelbrecht, Andre Janse van Rensburg, and all our South African colleagues, thank you for providing me the opportunity to seamlessly integrate into a team and for giving me the social context greatly required in a project like this. The largest acknowledgement of this work lies with the great many participants that were courageous enough to share their stories, give me time and help me to understand and be voice to their concerns.   Finally, the ?Four Quartets? could seamlessly integrate because all four sections exist under the major theme of a person?s relationship with time. During my Masters, a supportive community, without whom this thesis would not be possible, has been instrumental in shaping my time at this school. My mentors, Dr. Gary Poole and Dr. Charlyn Black: beyond the classroom, you have shaped my understanding of research as a process and you have embodied the virtues of being good people. Beth Hensler ? no words can capture the gratitude I feel for your generosity of spirit, emotional aptitude, and never-ending altruism. I can never thank you enough. My friends, Laura Dale, Kelly Garton, Abdullah Moussa, William Hall, Caitlin Frame and Alden Blair ? thank you for accepting me for me and I know we have a lifetime of friendship ahead. To Catherine Brown, Aleem Kherani, Kevin Skoblenick and Dr. Andrew Lui ? I want to acknowledge the time you have given me to go and the space you have given me to grow.  In the middle of our journey together, here ends one piece, and begins a new chapter. 	 ? xii	 ?Dedication   For all the times I can?t be eloquent or profound, this thesis is dedicated to the missing link in my never-ending thought spiral.  	 ? 1	 ?Chapter 1: Introduction  This thesis will outline barriers and facilitators to providing human immunodeficiency virus (HIV) counseling and testing services for health workers (HWs) within hospital occupational health services (OHS) in Free State Province, South Africa.  The first chapter will provide an overview of the HIV and AIDS epidemic and its effect on the health workforce. The consequences of this epidemic on human resources for health, legislation that seeks to improve access to services for HWs including HIV counseling and testing (HCT), and barriers to HCT uptake will be discussed based on current literature. This chapter will conclude with the rationale for this study, its objectives and study methods.   1.1 Literature Review 1.1.1 Human Resources for Health and the HIV and AIDS Epidemic  The World Health Organization (WHO) defines HWs as    ?all people whose main activities are aimed at enhancing health.  They include the  people who provide health services ? such as doctors, nurses, pharmacists, laboratory technicians ? and management and support workers such as financial officers, cooks, drivers and cleaners.  Worldwide, there are 59.8 million health workers.  About two thirds of them (39.5 million) provide health services; the other one third (19.8 million) are management and support workers.  Without them, prevention and treatment of disease and advances in health care cannot reach those in need (WHO Fact Sheet #302, April 2006).   Collectively, these HWs make up the health workforce and are considered human resources for health (HRH).  Today, nearly all countries in the world face health workforce challenges (Chen et al., 2004). This global ?crisis? in HRH amounts to shortages estimated at 4.3 million HWs worldwide; doctors, nurses and midwives comprise 2.4 million of this shortage (WHO, 2006b).  Unfortunately, the HRH crisis disproportionately affects countries that bear the highest disease burden. For instance, while 24% of the global disease burden is borne in the African region, it 	 ? 2	 ?has access to only 3% of the world?s HWs and less than 1% of the financial resources available globally (WHO, 2006a).  The consequence for the African region is that "the dire shortage of HWs?is among the most significant constraints to achieving the three health-related Millennium Development Goals: to reduce child mortality, improve maternal health, and combat HIV and AIDS and other diseases, such as tuberculosis and malaria" (WHO, 2006a: 19).  In fact, ?a shortage of human resources has replaced financial issues as the most serious obstacle to implementing national treatment plans? (WHO, 2006a: 20; Kober & Van Damme, 2004; Van Damme et al., 2006).   Human Resources for Health (HRH) and the Republic of South Africa?s (RSA) Health Workforce  As opposed to the other African nations, RSA is not considered to be a country with ?critical? HRH shortages based on WHO global standards. Where the WHO's minimum guideline for the number of nurses per population is 120/100 000 (WHO, 2008); in 2008, RSA's ratio1 was 436/100 000, well above the minimum guideline. However, when considering only the public sector, in 2001, this ratio was 120.3/100 000, but declined to 116.6/100 000 in 2008. Shortages are even more evident in the unequal distribution of HWs. The WHO (2008) value (116.6 per 100 000) contains huge disparities provincially, with the Northern Cape (155.0) and KwaZulu-Natal (136.3) above the minimum guideline. However, Free State province (94.5) and North West province (81.1) remain far below the minimum.  Furthermore, shortages of nurses in the public sector are exacerbated by vacant posts: in 2008, approximately 40.3% (19 277 of 47 834) of professional nurse posts in the public sector were not filled, along with vacancy figures of 34.4% in the case of medical practitioners and pharmacists (van Rensburg, 2012: 400-409).  Additionally, while RSA?s per capita physician 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?1 Corresponds to the number of professional nurses in public and non-public sectors. 	 ?	 ? 3	 ?ratio is much higher than other sub-saharan countries (where ratios are between 3 to 5 per 100 000 population) (George et al., 2012; Strachan, Zabow & Van der Spuy, 2011; Wadee & Khan, 2007; WHO, 2006a), the number of doctors in RSA (77/100 000 of the population) is much lower than other middle-income countries, such as Brazil (185/100 000) and Mexico (180/100 000) (van Rensburg, 2012: 387-389). Thus, while the WHO may not categorize RSA as a country in ?critical? shortage, there are critical issues facing its health workforce (Human Resources for Health South Africa - HRH Strategy for the Health Sector: 2012/13 ? 2016/17: available at http://www.doh.gov.za/docs/reports/2013/Human_resources.pdf; here in referred to as HRH South Africa, 2011). The HIV and AIDS Epidemic and its Impact on RSA?s Health Workforce   In RSA, HIV and AIDS is the leading cause of morbidity and mortality, which has led to extraordinary HRH challenges (Colvin et al., 2010). First, the epidemic creates stressful working conditions due to increased patient loads. The Human Sciences Research Council and National School of Public Health at the Medical University of South Africa provided a report to the Department of Health entitled The Impact of HIV/AIDS on the Health Sector (2002), to assess the extent to which the HIV/AIDS epidemic affects the South African health system.  The report demonstrates that heavy patient loads, staff shortages and low salaries have led to 33.8% of HWs experiencing low morale.  As a consequence, 16.2% of HWs had to be treated for stress-related illnesses and of these 63.9% had to take sick leave, resulting in high absenteeism. This report supports global trends that in areas with high HIV epidemics, difficult working conditions lead to low morale, burnout and absenteeism (WHO, 2006a). In the decade since this report came out, job satisfaction among HWs remains low (Bester & Engelbrecht, 2009).  A 2009 survey showed that 23.1% of community service doctors would likely leave RSA due to working conditions in 	 ? 4	 ?the public sector (Wolvaardt, 2010).  Factors reported as affecting attrition include HIV and AIDS, working conditions, workload in the public sector, relationship with management in the public sector, morale in the workplace, risk of contracting tuberculosis (TB), and personal safety (Matsoso & Stratchan, 2011).   Second, the HIV and AIDS epidemic places a dual burden on HWs, who have to support a growing patient population, while additionally managing themselves and their colleagues as part of this population when they become people living with HIV and AIDS (PLWHA). While very few statistics exist for the exact number of HIV positive HWs, Shiasana et al. (2002) estimate that 15.7% of HWs in public and private facilities in Free State, Mpumalanga, KwaZulu-Natal and North West provinces are also PLWHA. Younger HWs (ages 18-35 years old) had the highest estimates, with a HIV prevalence rate of 20% - the majority of who are women.  Connelly et al. (2007) demonstrated similar findings in a study that determined the prevalence of HIV infection in South Africa?s public health system by surveying 2032 professional and support staff employed by two hospitals in Gauteng province. Their results showed that prevalence of HIV among these HWs was 11.5%. When stratified by occupation, nurses had a higher than average rate at 13.7% (Connelly et al., 2007).  Prevalence by age was highest for the 25-34 cohort at 15.9%.  They concluded that 1 in 7 nurses in this cohort was HIV positive (Connelly et al., 2007). Moreover, the occupational risk of TB (Menzies et al., 2007; Joshi et al., 2006), including multi-drug resistant and extremely-drug resistant mycobacteria species, poses further serious risk to health workers.  One study (Connelly et al., 2007) suggests that despite the fact that health workers are at higher risk from blood-borne and airborne infections because of their risk of occupational exposure, they are still less likely to be tested for HIV than the population-at-large.  	 ? 5	 ?  Finally, the HIV and AIDS epidemic has forced health professionals and policy makers to consider non-conventional models of service delivery and care. For instance, in order to meet the needs of a population crippled by the HIV and AIDS epidemic, the national government introduced an anti-retroviral therapy (ART) program in 2003, which changed the nature and volume of clinical responsibilities of HWs. When this program was first introduced, 31.1% of professional posts were vacant in the public health sector (SAIRR, 2006).  The roll out and subsequent scale-up of this program, coupled with increased patient loads, added a burden to an already weakened health workforce in the public sector. However, it should be noted that the benefit of the scale up of ART for HWs may be the potential for improvements in their own health. Scale up of ART has meant prevention of occupationally acquired HIV infection, through availability of post-exposure prophylaxis (PEP). Outside of the OHS, treatment within the community, through access to anti-retrovirals (ARVs) for HWs that are PLWHA may also have an impact not yet measured within RSA. However, the impact of being on treatment can only make significant gains if both the proportion of HWs receiving and adhering to ART keeps up with the pace of ART service rollout (Makombe et al., 2007). Thus, while treatment and care options continue to become available to the public, in order to survive this epidemic the healthcare system is in need of innovative solutions targeted at the health workforce required to support these programs.    Innovative solutions are particularly relevant as, according to the International Labour Organization (ILO) report HIV/AIDS and Work: Global Estimates, Impact and Response, (ILO 2004) in the absence of increased access to treatment an estimated 74 million workers will be lost to the workforce due to HIV by the year 2015.  This report makes it evident that increasing 	 ? 6	 ?the number of workers is not sufficient. In a recent study of HWs job satisfaction and intention to leave, Blaauw et al. (2013) demonstrated that despite better working conditions2 than Malawi or Tanzania, South African health professionals showed the lowest level of satisfaction and highest turnover intentions, due to low morale.  This is supported by previous literature that demonstrates in RSA, the lowest level of job satisfaction exists among public sector nurses (Pillay R, 2009, Kekana et al., 2007 and Uys et al., 2004). With 16% of the health workforce expected to leave due to illness or death based on the HIV and AIDS epidemic (Shisana et al., 2002), retention and care of remaining staff becomes critical to preventing further shortages. In order to retain and support existing workers, reinforcing and scaling up of existing services that target the specific needs of HWs who may themselves be PLWHA must be considered. 1.1.2 HIV Counseling and Testing in RSA International Recommendations on HCT for HWs in Healthcare Settings  The WHO, ILO and the International Organization for Migration recognize that ?although HWs are at the frontline of national HIV programs, they often do not have adequate access to HIV services themselves? (WHO, 2006b).  This observation, along with the increased risk to HWs of developing TB, has lead to the creation of the Joint WHO ILO UNAIDS policy guidelines on improving health worker access to prevention, treatment and care services for HIV and TB (28 April 2010; herein known as ?WHO/ILO guidelines?).  The WHO/ILO guidelines (2010) are an ?evidence-informed? policy for the provision of improved access to HIV and TB prevention, treatment, care and support for HWs, with the underlying goal to improve health for HWs and to strengthen the existing workforce through retention.  In order to facilitate this goal, the WHO/ILO guidelines (2010) are in favor of providing priority access by policy (not simply 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?2 Described based on a health system that has better resources	 ?	 ? 7	 ?practice) to programs that enable HWs to obtain prevention, treatment and care services ahead of the general public. The guidelines clarify that priority access does not mean priority over other priority groups (such as pregnant women), however, it does imply that HWs should not have to wait with the population-at-large in order to receive ??convenient, accessible, free, confidential, gender-sensitive and non-stigmatizing? services.    The WHO/ILO guidelines (2010) recommend that the site for providing priority access should be staff clinics in the workplace, with the caveat that this should only be done if confidentiality can be maintained, and a holistic program is offered.  Access to an onsite, comprehensive occupational health (OH) programme was seen as favorable from a cost and integration perspective.  In fact, a study by Moodley and Bachman (2002) that surveyed hospitals in RSA showed that hospitals that provide primary health care and the management of chronic illness as part of their OHS did not experience logistical or financial problems as a result. The WHO/ILO guidelines (2010) state that ?offering annual testing (including voluntary counseling and testing [VCT] and TB screening) in staff clinics as part of [an] existing annual check-up of general health is recommended by international organizations as good practice, and supported by evidence as a preferred approach.?    National Policies on HCT for HWs in Healthcare Settings  In keeping with international guidelines, RSA is an example of a country that has introduced national policies aimed at providing their health workforce with services that prevent occupational exposure to blood-borne and airborne diseases (like HIV and TB), and to prioritize access to programs that encourage not only prevention, but access to diagnosis, treatment and care.  	 ? 8	 ?First, RSA?s Employee Health and Wellness Strategic Framework for Public Service3 (2008) creates a national strategy to integrate health and safety within the workplace, with the goal of creating a healthy workforce by increasing productivity and enhancing service delivery. The objectives of this national policy is to ??facilitate the development of strategies, mechanisms and interventions by government departments and provincial administrations for the implementation of HIV&AIDS and TB Management, Health and Productivity Management; Safety, Health, Risk and Quality Management; and Wellness Management in the Public Service.?  It is based on current national priorities, as guided by the burden of disease in the South African worker population (Matsoso et al., 2011).    Second, the National Strategic Plan on HIV, STIs and TB, 2012? 2016  (NSP) outlines tangible objectives to achieve the Strategic Framework?s goals. These include reducing new HIV infections by at least 50%, initiating at least 80% of eligible patients on ARVs, reducing the number of new TB infections and reducing self-reported stigma related to HIV and TB by at least 50% (National Department of Health, 2012). The NSP is clear that all high risk workplaces should have occupational health policies on TB and HIV that are confidential, free from discrimination and offer routine medical screening and testing of employees, respiratory infection control, treatment, sick leave and psychosocial support.  Additionally, the NSP states that all wellness programs should address HIV, sexually transmitted infections (STIs) and TB in an integrated manner, aligning with national standards. Integration of HIV and TB services is also supported by international guidelines. The WHO (2009) describes how HCT may act as an 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?3	 ?This framework is influenced by the World Health Organization Global Plan of Action on Workers Health 2008-2017, the International Labour Organization?s Decent Work Agenda in Africa 2007-2015 and the recommendations of the report of the WHO?s Commission on Social Determinants of Health.	 ?	 ? 9	 ??entry point? in TB infection control (IC).  These guidelines outline that the package of TB IC4 depends on the HIV status of HWs, in addition to general administrative, ventilation and respiratory protective equipment (WHO, 2009). Accessing HIV testing within OHS has an additional benefit for TB care, as knowledge of a HWs HIV positive status may allow HWs working in areas with high TB risk to have the option of relocating to a low-risk area within the workplace (Buregyeya et al., 2012).   Thus, knowledge of HIV status is considered critical for both the prevention and treatment goals for both HIV and TB within the workplace. Once an individual has been tested for HIV, prevention can be reinforced and referral made to available treatment, care and support services. Furthermore, it has been argued that a potential added benefit is that increasing the number of people who know their HIV status through expanded access to HCT may also result in a decrease in HIV-related stigma, leading to a ?normalization? of the HIV epidemic (van Rooyen et al., 2009, 165; Jewkes, 2006; De Cock, 2002).  Thus, providing HCT to HWs may be beneficial to meeting the goals of RSA?s National Strategic Plan and overall, aid in the strengthening of RSA?s health workforce. After all, HIV testing is the gateway to treatment (Namakhoma et al., 2010). However, in order to provide such services, statement 5 of the WHO/ILO guidelines (2010), stipulates that implementation of ??programs for regular, free, voluntary and confidential counseling and testing for HIV and TB?? should happen in ??conjunction with health workers representatives?.  This makes RSA an ideal country to implement HCT for HWs, as RSA?s Employee Health and Safety Act (1993) legislates that all workplaces have one elected health and safety representative per 20-100 workers (depending on risk levels). Furthermore, the 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?4 According to the WHO (2009) guidelines, the TB IC package for HWs includes regular screening for TB, access to HIV care and ART, as well as Isoniazid Preventive Therapy (IPT). Additionally, HIV infected HCWs working in areas with high TB risk should ideally have the option to be relocated to a low-risk area in the health facility. 	 ? 10	 ?Health and Safety Act (1993) outlines that a Health and Safety Committee (HSC), comprising these elected representatives, management appointed representatives (representing less than elected members) and specialists (who are ex-officio members) must be created.  The purpose of the HSC, and the representatives specifically, is to identify hazards, write reports and act as a representative voice of HWs to the HSC.   In summary, RSA?s legislation and national strategy mirrors international guidelines and standards that advocate for access to services for workplace based HIV and TB programs. HCT for HWs is an example of a service that is in keeping with the NSP. Thus, RSA provides an appropriate setting to study the impact of this service for HWs delivered in health care settings.   Workplace-Based HCT Services in Healthcare Settings  While national and international guidelines are promoting HCT for HWs, there have been few intervention evaluations and even fewer ?model? programs to demonstrate the efficacy of HCT services for HWs in healthcare settings.  However, a claim for the efficacy of such services in occupational health settings can be made. In a cluster-randomized trial conducted in Zimbabwe, Corbett et al. (2006) randomly assigned businesses to either on-site rapid HIV testing, or vouchers for off-site voluntary counseling and testing (VCT) at a chain of freestanding centers.  The mean uptake of VCT was over 51% for the site service (within the OH) of the business arm, compared to only 19% for the off-site arm. The result was attributed to the workplace site being seen as convenient and accessible. Similarly, Namakhoma et al. (2010) demonstrated that short distances and perceived quality of service were the most cited factors for choosing testing and treatment facilities by health workers in Malawi.  However, it is not clear if these results generalize to a health care facility. 	 ? 11	 ?The most direct example for providing HCT services for HWs in health settings was presented in a study by Uebel et al. (2007), where staff care programs were described in 3 hospitals (2 in RSA, and one in Botswana).  While two of the programs were on-site, one was off site, near the hospital.  Overall, all three programs were viewed as successfully able to implement HIV testing and treatment for HWs.  ?Success? was subjectively described as observed improvements in morale, which was attributed to staff observing clinical improvement in their colleagues and becoming aware of where to access treatment. Successful treatment of staff also contributed to an anecdotal observation5 of decrease in stigma.  Furthermore, Uebel et al. (2007) attributed programmatic success to the work of trusted staff members that acted as counselors, clinicians, and advocates. The authors observed that confidentiality was a major barrier to initial implementation. However, caution should be exercised in interpreting the results of this study as: 1) ARVs were included as part of a comprehensive occupational program and; 2) no measures of observed concepts, such as ?stigma? and ?morale? were captured.  Thus, the effect of HCT, independent of providing ARVs, cannot be assessed.  In turn, scale up of HCT services requires further investigation, especially in locations where only HCT is currently provided and not treatment. In this context, the potential barriers to uptake of HCT services become increasingly important. Barriers to uptake may demonstrate that HCT may not be viable in the absence of treatment options.  However, the value of studying HCT alone cannot be undermined, because HCT is often referred to as the gateway to treatment (Namakhoma et al., 2010), whether provided onsite or not.     	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?5 It should be noted that there were no measure of decrease stigma in the study	 ?	 ? 12	 ?1.1.3 Barriers to Uptake of HCT in RSA While the number of public health and non-governmental sites for HCT in RSA has increased, uptake of HCT remains low (van Rooyen et al., 2009). While no studies on the proportion of HWs in RSA who have been tested for HIV could be found, Tarwaireyi et al. (2003) published a study that examined the proportion of HWs who had undergone VCT6 for HIV in three rural districts in Zambia. The study found that only 12% of HWs had gone for VCT and that 77% did not want an HIV test at all.  The reasons given for not testing included not being ready to go for VCT, not being able to cope with the results, not having the courage to go and more than 50% feeling that they did not need to be tested.  This study concluded that while VCT programs have been targeted at the larger community, HWs face many of the same issues as the communities they service and they require specialized programs with strong counseling components. This study is in keeping with literature that demonstrates individual factors within the wider community including fear of testing positive for HIV, fear of the ramifications of a positive test and fear of stigmatization (Namakhoma et al., 2010; Day et al., 2003; Kalichman and Simbayi, 2003) are the most common reasons why an individual chooses to not get tested. This also supports the belief that in the age of readily accessible treatment in the form of ART, what we are really fighting is not an epidemic of HIV and AIDS, but rather the epidemic of HIV and AIDS stigma (Rankin et al., 2005).  Finally, self-perception that one?s risk of HIV infection is low (Shiasana et al., 2005) may also contribute to low uptake of HCT. While several individual level reasons for lack of uptake of HCT services exist, other institutional and structural level issues may also affect individual behavior.  In RSA, system factors that limit uptake may include lack of trust in the health care system, fear of 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?6 Voluntary counseling and testing is the primary from of HCT. HCT is the general term for HIV testing without explicating the type of testing. 	 ?	 ? 13	 ?discrimination by the HW (van Dyk and van Dyk, 2003) and perception of a lack of confidentiality of the service itself (Obermeyer & Osborn, 2007). Lack of confidentiality can be perceived to have life-changing consequences, as fear of violence is also a barrier to testing and disclosure of HIV status to sexual partners (Jewkes, 2006; Maman et al., 2001; van Rooyen et al., 2009). Thus, real and perceived stigma and discrimination are among the largest barriers to implementation of HCT services in high-prevalence countries  (Deacon et al., 2009: 105-123).  HIV Stigma  In 2003, stigma was internationally recognized as a great barrier to HIV efforts, when it was highlighted as the theme for World AIDS Day.  The impact of stigma in undermining HIV and AIDS programming efforts has been well documented since the early 1990s and gained this international recognition with the work of Jonathan Mann in 1996.  His main thesis is that stigma ??exacerbates the effects of the epidemic and reduced the efficacy of AIDS programming efforts? (Mann, 1996).  Research now confirms that stigma and discrimination reduce the impact of prevention programs, inhibit treatment uptake and adherence, exacerbate the psycho-social effects of the HIV-infection and reduce the quality of life of people living with HIV and AIDS (for a review of stigma in RSA, see Deacon et al., 2005 and Deacon and Stephney, 2007). A notable South African study in Cape Town demonstrated that expressing stigmatizing attitudes against PLWHA was associated with delayed testing and treatment (Kalichman and Simbayi, 2003).  Other studies support this finding, where stigma has inhibited or delayed HIV treatment or testing services or affected ART adherence (Namakhoma et al., 2010; Cameron, 2007; Doherty et al., 2006; Hutchinson and Mahlalela, 2006; MacPhail et al., 2008; Uebel, 2004).  Thus, it has been suggested that in the era of readily available treatment in RSA, which has the largest public health ARV programme in the world, the HIV and AIDS epidemic has 	 ? 14	 ?shifted from a fight against low resources to one where stigma associated with HIV may be a greater impediment to treating HIV (Cameron, 2005). The most common definition of disease stigma owes to the earliest theoretical work of Goffman (1963), who stated that ??people who possess a characteristic defined as socially undesirable [such as HIV infection] acquire a ?spoiled identity??? who?s consequence is ??social devaluation and discrimination.?  Although this overarching definition of stigma persists, research has required differentiating various kinds of stigma in order to further understand what it is, how it operates and to measure the amount of it within a system.  Some types of stigma include instrumental stigma (fear of infection from casual transmission), symbolic stigma (moralistic shaming and blaming), felt/internal/perceived stigma (a stigmatized person?s internal sense of shame and fear of persecution) and expressed/external/experienced stigma (violence, discrimination or ostracism directed at one individual) (Herek, 2002). A person may also protect his or her identity as a person living with HIV and AIDS to protect their family and friends, known as courtesy stigma.  It is important to realize that while forms of stigma have been identified, stigma is a process or series of processes contingent on social norms and contexts (Aggleton & Parker, 2002). Thus, it is important to note that these typologies of stigma are not separate entities, but different dimensions of the same phenomenon.  One major area of research in barriers to uptake of HCT services focuses on both stigma and discrimination.  Often, the two concepts are used interchangeably, although they are not the same.  Discrimination can be seen as a form of external/experienced/expressed stigma but is not encompassing of all typologies of stigma. It should be noted that stigma can affect individuals negatively even in the absence of active discrimination, thus investigating stigma in the context of discrimination solely is problematic (Jennings et al., 2002).  	 ? 15	 ?British sociologist Graham Scambler (2009) provided the insight that actual discrimination does not need to occur for stigmatized people to suffer from the results of knowing their status.  For instance, felt/internal/perceived stigma motivates individuals to pass as members of the non-stigmatized majority.  However, in order to pass as a member of a non-stigmatized group the individuals change their behavior to fit in. When perceived stigma is internalized (i.e. negative judgments are accepted) and the individual makes judgments about how and how much stigma may be experienced (perceived/anticipated/felt stigma), this internal stigma greatly increases psychological distress. In turn, this distress restricts behavior and limits the opportunities of the individual experiencing felt stigma due to a feeling of vulnerability that may be persistent in the absence of overt or enacted stigma (Scambler, 2009). As a consequence, felt stigma might lead to an unwillingness to seek help and access resources (POLICY Project, 2003).  In fact, Malcolm et al. (1998) state that perceived or felt stigma often precedes rather than results from the experience of stigma. Thus, establishing trust between the HW and the health system, which represents their community, may be critical to removing this barrier to health service provision. Confidentiality   Lack of confidentiality has been identified as a key factor in unwillingness of patients to disclose HIV status to health professionals (Greeffe et al., 2008) and a barrier to uptake of HCT. Confidentiality is defined as ?the practice of keeping harmful, shameful, or embarrassing patient information within proper bounds? (Purtilo, 1999). The concept of ?confidentiality? in medical practice originated in ancient times, where its importance relates to its perceived value at the level of society and the individual (van Rensburg, 2012: 665).  The specific value of confidentiality to society is in sustaining trust within health professional-patient relationship.  	 ? 16	 ?Confidentiality in medicine ensures: 1) that respect is given to a person?s desire for privacy and, 2) acknowledges the vulnerability of patients seeking medical care (Kling, 2010). Confidentiality improves health care by allowing an individual or patient to trust a health professional with personal information (Kling, 2010). Professionalism in the form of assured confidentiality creates trust. However, the concept of professionalism is more complex today than in the past (Freidson, 2001) and thus, professional accountability is even more difficult to assess.  At the level of the individual this trust establishes a quality of relationship between health provider and recipient, such that it becomes safe to be vulnerable.  As infectious diseases become a topic of increasing concern within public health, there has been a change from absolute confidentiality to a relative confidentiality that takes into consideration public health notification requirements.  It is now broadly agreed that confidentiality is not absolute and that there are limits, e.g. when maintaining confidentiality conflicts with the rights of others, or would result in harm to the patient.  Within RSA?s Employee Health and Wellness Strategic Framework for the Public Service (2008), out of eleven core principles for implementing this strategy, the two principles that are the cornerstone for the Employee Health and Wellness (EH&W) programs, practices, and services are confidentiality and ethical behavior. With respect to confidentiality, the policy states:  Employees utilizing the EH&W programme are assured of confidentiality, except in cases of risk to self and others or in terms of legislation. Strict confidential records and information outside the personnel records should be maintained at all times. The Departments will ensure the confidentiality of all medical and personal information of employees. The employees within the Public Service have a constitutional right to confidentiality about their HIV and AIDS status, in particular, should an employee reveal his/her HIV status, this information shall not be disclosed to any other party including other employees, union member or management. Consent for disclosure should be in writing.  (Employee Health and Wellness Strategic Framework for the Public Service, 2008)   While confidentiality is assured, it is not absolute and thus has been very contentious in relation 	 ? 17	 ?to HIV infection. Officially sanctioned breaches in confidentiality may include reporting to public health authorities, or the shared confidentiality expected between providers caring for the same patient.  However, shared confidentiality does not include unofficial breaches such as revealing the results of an HIV test to other HWs not involved directly in a particular patient?s care.   The issue of confidentiality becomes increasingly complex when considering the practical way in which a health system operates. The growth of medical teams has necessitated sharing medical knowledge about patients within referrals.  Within hospitals, these referrals take place in complex institutions, which utilize computers and information networks. While computers make access to information readily available, they also increase questions on confidentiality of information and magnify privacy concerns when questions on who has access to information about patients and employees arise (van Rensburg, 2012: 666). Thus, assurances of confidentiality have to operate within a health system that is practical, where consequences for breaches are delineated and where professional conduct prevails, in order to establish trust in health service provision.    Stigma and Confidentiality in Healthcare Settings  In health care settings, lack of confidentiality and stigma are known to reduce access to health services (Uebel, 2004).  Stigma (in the form of gossip or discrimination) may lead to breaches in confidentiality, just as breaches in confidentiality may lead to stigma (i.e. internal stigma).  While the literature has identified these two areas as barriers to health service provision for patients with HIV/AIDS, there is a relative paucity of literature on the role stigma and confidentiality play in preventing access to healthcare for HWs specifically within occupational settings. As Uebel et al. (2007) observe, ?The fear of disclosure to colleagues within the small 	 ? 18	 ?community atmosphere of a hospital, if confidentiality is not protected, can make the barriers seem insurmountable?. The effects of this stigma, when compounded by the burnout experienced by many HWs, contributes to enormous barriers for HWs accessing HIV care. Thus, as access to HCT services begin to be promoted in OHS for HWs, it becomes critical to answer three main questions:  1) To what extent are stigma and lack of confidentiality commonly perceived as existing within healthcare facilities by HWs?  2) How and to what extent do stigma and a lack of confidentiality contribute to the utilization of HCT by HWs?  3) What can be done to reduce HIV related stigma and improve professionalism with respect to breaches of confidentially by HWs while expanding OHS services for HIV?  1.2 Rationale for the Current Study The WHO/ILO/UNAIDS (2010) guidelines argue that both active health and safety committees (HSC) and policy level commitments by national governments must be in place when services for HWs such as HCT become available within an OHS. Pelonomi Regional Hospital, a public hospital in Free State province, RSA, is a hospital with a strong HSC, strong management support, an active OHS and national, provincial and academic support (Yassi et al., 2009). Thus, this hospital would appear to be an ideal site to offer confidential HCT within its OHS.  However, between January and May 2011, the OHS nurses at Pelonomi conducted an audit of HWs use of OHS. They found that among the 1900 HWs at this hospital, hepatitis B immunization was given to 568 people, primary care that includes annual medical exams were given to 405 people and various other services were heavily used (Myeko et al., 2012).  However, only 121 individuals used the HCT service (Myeko et al., 2012). This service is 	 ? 19	 ?monitored as part of the overall HIV services offered at this OH unit. As underutilization of these services was identified, despite Pelonomi hospital appearing to have all facets necessary for successful implementation of HCT service within its OHS, this prompted the question of why this was occurring. In order to address this question and improve the uptake and quality of these services, the purpose of this study is to elucidate barriers and facilitators to the use of occupationally based HCT programs for HWs.   1.3 Objectives The main objectives of this study are to: 1) Elucidate major barriers to uptake of occupationally-based HCT by HWs within three tertiary care hospitals in Free State Province, RSA  2) Propose possible facilitators to overcome the identified barriers to subsequently improve HCT services 1.4 Methods  A mixed methods approach was applied to assess barriers and facilitators to the use of occupationally based programs for HCT among HWs. Data was analyzed from a substantial section of a self-administered questionnaire and from focus group interviews (FGIs). Data analysis involves triangulation of results/findings across these quantitative and qualitative data sources, in order to increase validity, provide context for understanding broad-brush quantitative findings, and enhance the possible interconnected nature of the results and findings.   Self-Administered Questionnaire (Chapter 2)  The survey analysis described in this study was conducted on a subsection of a larger occupational health and safety survey conducted by the Department of Health in Free State province, the University of Free State and the global health research team at the University of 	 ? 20	 ?British Columbia. The survey was distributed to a 20% quota sample of HWs at Pelonomi, Universitas and Bongani hospitals. The portions of the self-administered questionnaire that were analyzed in this study pertain to the sociodemographic characteristics of HWs who completed the survey and questions on a broad range of possible reasons for not using HIV services provided by the occupational health service (OHS) at the three study sites. Based on this data, HWs general perceptions of HIV and AIDS, consequences of having HIV in the workplace, and perceptions of HIV stigma in the workplace are reported. In addition, with specific reference to the OHS, HWs perceptions of the OHS staff and service, including confidentiality of the OHS, are also reported. Focus Groups (Chapter 3)  Once preliminary results from the questionnaire were analyzed (i.e. once descriptive statistics were calculated), five FGIs (n=39 HWs) were conducted using with a semi-structured interview guide. The focus groups aimed to contextualize findings and see if other barriers, and possible facilitators to overcome barriers, could be identified. As the questionnaire treated HWs as a group, FGIs were employed, rather than individual semi-structured interviews, to try to determine collective views on barriers to HCT service. Recruitment was open to all HWs, advertised through third party recruitment and by word of mouth. All sessions were conducted in English, audio-recorded and transcribed verbatim.  Rationale for Mixed-Methods Approach  It is appropriate for this study to employ a mixed-methods approach as it tries not only to identify barriers, but also to understand how barriers such as ?stigma? and ?confidentiality? effect utilization of HCT programs within OHS, in order to provide recommendations to overcome these barriers. There is an increasing recognition that combining quantitative and qualitative 	 ? 21	 ?approaches through mixed-methods inquiry improves both the research process and the interpretation of results (Moffatt, White, Mackintosh, & Howel, 2006).  As HCT exists within a HIV service provided in an OH unit for HWs within a hospital setting, the provision of HCT itself may be viewed as a ?complex intervention? (Campbell et al., 2000; Craig et al., 2008). Coupled with studying barriers that include conceptual components, such as ?trust? or ?stigma?, which are themselves difficult to evaluate using quantitative approaches alone, a mixed-methods approach helps understand both the intervention (HCT) and its relation to factors that influence its effectiveness. However, detecting the presence of barriers such as ?stigma? and ?confidentiality? without first assessing possible differential explanation for lack of uptake (including such simple explanations as a lack of knowledge of the existing service), make quantitative analysis a necessary first step in this study.  This study begins with a self-administered survey to empirically detect the presence or absence of factor(s) (i.e. stigma, confidentiality, perceptions of staff, etc.) determined a priori that are hypothesized to influence HWs access to HCT services.  As the survey captures a sample that closely approximates the demographic characteristics of the health workforce at the three hospitals, the survey generates evidence regarding the factors that are most relevant to underutilization of HCT services by the HW population at these institutions.  However, the questionnaire cannot conclude that that these are the only factors that act as ?barriers?, or, within barriers such as stigma or confidentiality, explain why these barriers are a problem (e.g. what types of stigma may be an issue). Thus, FG results will help contextualize the barriers identified in the survey. In addition, FGs of individual HWs may identify additional barriers and subsequently the reasons for why HWs subsequently chose not use the service in reference to these factors.	 ?	 ? 22	 ?Data Analysis  Results from the quantitative survey were analyzed using descriptive and comparative statistical methods in SPSS to identify which hypothesized factors were perceived to be barriers to the use of OHS among respondents. Transcripts from semi-structured FGIs were transcribed verbatim and analyzed thematically to generate a rich understanding of the possible reasons behind HWs choice to not use the OHS service. 	 ? 23	 ?Chapter 2: Analysis of ?HIV/AIDS and Occupational Health? Section of a Self-Administered Questionnaire  2.1 Aim  To identify barriers to HWs uptake of occupationally-based HCT at Pelonomi Regional Hospital, Universitas Academic Hospital and Bongani Regional Hospital.  2.2 Method  The analysis described in this chapter was conducted on a subsection of a larger occupational health and safety survey conducted as part of an occupational health and safety project by the Department of Health, the University of Free State and the global health research team at the University of British Columbia. 	 ?2.2.1 Participants and Sampling  The survey was distributed to a purposive sample of HWs at three large public tertiary care hospitals in the Free State province, RSA, in early 2012. A stratified random sampling method was not possible due to time and resource constraints and the lack of access to individual HWs identification and contact information. As a result, the original survey used a non-random quota sampling method (Bryman, 2012), to ensure that the sample reflects the HW population at the aforementioned hospitals with respect to the distribution of various occupations, namely doctors, nurses, allied health professionals (AHPs), administrative staff and support staff. To achieve this, an a priori sampling frame was created using hospital employment data7 to determine the number of HWs from each occupational category that would need to be surveyed in order to achieve a 20% sample of the 5080 HWs from these hospitals (Appendix B). A target 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?7 Data was accessed through the PERCEL database	 ?	 ? 24	 ?sample size of 20% of HWs at these hospitals was selected based on what the primary investigators felt would be achievable based on the recruitment process described below. 2.2.2 Procedure   Two members of the research team distributed questionnaires in person by visiting departmental units within each of the hospitals. HWs within these units were then approached individually, informed about the purpose and nature of the study, and invited to complete the survey. Those who agreed were provided with a copy of the questionnaire. Informed consent was assumed to have been provided by participants if they voluntarily submitted a completed questionnaire and signature on a form (which was provided at the end of the questionnaire). Any HWs who expressed interest in participating were provided a copy of the questionnaire. Both English and SeSotho speaking HWs were included in the survey. Participants were given a small package of snacks (i.e. a granola bar, juice and a bag of chips) as a token of appreciation for participating. The recruitment continued until all of the surveys required for each quota had been distributed. Approximately one-third more surveys then required by the quota were handed out to increase the likelihood of achieving the targeted 20% sample.  2.2.3 Instrument and Measures   The survey was developed through discussion and iterative revision among multiple individuals in the research team. The questionnaire was piloted with a sample of participants, to ensure comprehension and clarity of the survey questions. The English instrument was revised based on comments from respondents in the pilot test. The survey was then translated into SeSotho using a backward translation method and then reviewed and revised to ensure that the correct meaning of questions was retained in the translated questionnaire.  	 ? 25	 ?The final questionnaire consisted of three sections: (1) sociodemographic information about participating HWs; (2) questions related to the HWs workplace experience and views on health and safety and (3) questions regarding HIV/AIDS and the OHS. The present analysis was conducted on questions from sections 1 and 3 of this survey. A sample of these sections of the questionnaire is presented in Appendix A. 	 ?Independent Variables of Interest from ?Section 1? of the Survey (Sociodemographic Characteristics) The first section of this survey collected basic sociodemographic information on participants, which were used as independent variables and covariates in the subsequent analysis. Each question pertained to a specific characteristic that was coded into discrete categories. The following questions were used in the present analysis: i. At which hospital do you work? For this question, participants selected from 3 categories: Universitas, Pelonomi, or Bongani. Participants who indicated working at more than 1 of these hospitals were coded as ?multiple?.  ii. What is your age? For this question, participants selected from 6 categories: <20; 20-29; 30-39; 40-49; 50-59 and >60.  iii. What is your sex? For this question, participants selected female or male.  iv. What is your race? For this question, participants selected from 5 categories8: Asian, black, coloured, white and other. v. What is your occupation? For this question, participants selected from 16 categories, or if their occupation was not listed, wrote in the appropriate response (see Appendix C for ?occupational categories?). 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?8 As these are the legally designated groups in the RSA. 	 ? 26	 ?Dependent Variables of Interest from ?Section 3? of the Survey (Barriers to Uptake of HCT)  The third section of the questionnaire aimed to identify potential barriers to uptake of HCT services in the OHS units of the study sites. The questions included in this study have been grouped into 7 domains for the purposes of the current analysis. The specific questions included within each domain are described in Table 1.  Table 1: Dependent variables included in analysis Domain Domain Description Questions included 1 General perceptions regarding HIV/AIDS ? What proportion of your co-workers do you think would want to know their HIV status? ? What proportion of your co-workers do you think would do the following if they were told they had HIV o Would change their sexual behaviour? o Would change their work practices? o Would want to get treatment as soon as possible? ? If a co-worker found out that they have HIV, who do you think they are most likely to share that information with? 2 Consequences of HIV in the workplace ? If someone?s boss found out they had HIV, what do you think would happen to them? 3 Perceptions of HIV stigma in the workplace ? Do you think there is HIV stigma in the workplace? 4 Knowledge of the OHS ? Do you know where the OHS unit or services are located? ? Do you know the operating times of the OHS unit? ? Do you know if HIV treatment is available at the OHS? ? Do you know if TB treatment is available at the OHS? 5 Perceptions of the OHS ? Do you think that health care workers use the HIV/AIDS programme in the OHS unit effectively? ? Regarding OH practitioners: o The OH practitioners are well-trained to offer HIV counseling and testing? o The OH practitioners encourage people to use the OHS unit for HIV/AIDS services? 6 Potential reasons for not using the HCT services ? If you and your co-workers do not use the OHS for HIV services, tell us how likely it is that each of the factors below explains why? 7 Perceptions of confidentiality in the OHS ? Do you think confidentiality is maintained in the OHS unit?  	 ? 27	 ? For questions addressing the first two domains, participants were asked to reflect on their co-workers9 perceptions of HIV/AIDS and the associated services at the OHS unit. The investigators? rationale for not asking HWs directly about their preferences or actions was due to the ethical and human rights considerations in asking individuals to discuss their status and personal views on HIV/AIDS. The remainder of the questions asked participants to reflect on their own knowledge and perceptions.  In addition, each of these questions used one of three response types:  i. The first response-type utilized a 4-point Likert scale. For instance, on the question ?What proportion of your co-workers do you think would want to know their HIV status?? participants rated their response as ?very few?; ?some?; ?most? or ?all?. Alternatively, for the question ?Do you think HWs use the HIV/AIDS programme in the OHS unit effectively?? participants chose from the responses ?not at all?; ?not often?; ?most of the time? and ?always?. ii. The second response-type used a dichotomous scale to indicate a positive, negative, or neutral response. For instance, to the question ?Do you know the operating times of the OHS unit?? participants indicated ?yes?, ?no? or ?unsure?. iii. The third response-type presented a list of options to participants. For example, on the question of potential reasons for not using the HIV programme at the OHS, participants selected from a list of 5 reasons: (i) ?not given time during the work day to go to the OHS unit?, (ii) ?concern that the worker will have to pay for the service?, (iii) ?see no point if anti-retroviral treatment is not provided?, (iv) ?fear that confidentiality will not be maintained?, and (v) ?not applicable?. 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?9	 ?Defined within the survey as ?people in your occupational group, or on your unit?	 ?	 ? 28	 ?2.2.4 Data Analysis All statistical analyses were conducted using the Statistical Package Software for Social Science (SPSS) version 20 (IBM Corp, Armonk, NY). Descriptive statistics were used to summarize each independent and dependent variable. Specifically, univariate analyses in the form of frequencies and percentages of each response category were calculated for all questions in section 1 and for all questions in section 3 using Likert or dichotomous scale responses. In addition, for all questions in section 3 that presented a ?list of options?, the frequency and percentage of participants selecting each item in the list were calculated.  Recoding of Independent and Dependent Variables Further analyses were conducted to determine the association between the sociodemographic characteristics and 2 specific barriers to uptake of HCT that have previously been identified in the literature: (i) perceptions of HIV stigma in the workplace and (ii) perceptions of confidentiality in the OHS. As data from both of these dependent variables are ordinal in nature, the most appropriate analysis in this case would be multiple ordinal regression. However, examination of the frequencies for all combinations of the independent variable for both of these variables demonstrated that for some response categories, there were empty cells with no observations, which could potentially lead to an unstable model (Field, 2009). To address this, the response categories for both dependent variables were instead recoded as binary responses to facilitate multivariate logistic regression.10 Specifically, the responses ?not at all? and ?not often? were recoded as ?no? and the responses ?most of the time? and ?always? were recoded as ?yes?. For consistency, the binary codings were used throughout the univariate, bivariate and multivariate analyses of these variables. 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?10 The observed frequencies were re-calculated based on the recoded data, which confirmed that there were no empty response categories.	 ?	 ? 29	 ?In addition, as the number of observations in some categories of the independent variables age, race and occupation were too small to facilitate bivariate or multivariate analsysis11, these categories were also collapsed to allow for these independent variables to be included in these analyses. The specific coding changes were: i. In the variable age, the categories of <20 and >60 were collapsed into the categories 20-29 and 50-59 respectively. Thus, this variable was recoded as <30, 30-39, 40-49, and >50.  ii. In the variable race, the categories ?asian? and ?other? were collapsed into a single category. However, there were still an insufficient number of observations in this collapsed category. Thus, these observations were removed from the analyses involving race.12 Subsequently, the race category was recoded as black, white and coloured.  iii. For the variable occupation, each of the 16 categories and all written responses were initially coded according to the sample frame occupations (i.e. doctor, nurse, AHP, administrative staff or support staff). However, there were still an insufficient number of physicians and AHPs to facilitate multivariate analyses using these categorizations. Thus, the categories were further collapsed into patient-care health workers (PCHWs, corresponding to doctors, nurses and AHPs) and non-PCHWs (corresponding to administrative and support staff), based on existing literature (HCJ van Rensburg, 2012: 410). The codings for the occupation variable are detailed in Appendix C.  Bivariate and multivariate analyses: Bivariate analyses were conducted using Pearson?s chi-square tests, to determine crude associations between each dependent variable (perceptions of stigma and confidentiality) and all independent variables (occupation, age, race, sex and hospital), with the threshold for a 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?11 This was determined using the same process of reviewing frequency tables as described for the dependent variables related to confidentialty and stigma 	 ?12 This resulted in a total of 6 (0.6%) of the observations related to race being removed from the analyses. 	 ?	 ? 30	 ?significant result set at p < 0.05. Missing data points were recorded and these observations were excluded from the bivariate analyses for the variable in question. Non-missing data from the participant were included in other bivariate analyses as appropriate. The frequency and percentage of responses for all bivariate combinations of the independent and dependent variables, as well as the p-values for each chi-square test, are presented in section 2.3, ?Results?. Finally, multivariate logistic regression (MLR) analyses13 were carried out to determine the relationship between the aforementioned sociodemographic variables and each dependent variable, after controlling for the other covariates. The multivariate model was generated by initially including all independent variables that were found to be significantly associated with the dependent variables in the bivariate analysis, as well as all two-way interactions.14 Interaction terms followed by covariates were removed one at a time if they did not contribute significantly to the model based on the results of the likelihood ratio test (LRT). This process continued until further elimination of any variable in the model produced a significant LRT result at a threshold of p<0.05, indicating the variable contributed significantly to the variation in the dependent variable observed in the sample. The odds ratio and 95% confidence interval (95% CI) were calculated for each covariate in the final model.  2.3 Results   A total of 978 HWs completed the survey, representing 19.4% of the workforce at the three hospitals sampled. This response rate was just below the planned quota based on the a priori 20% sampling frame. Physicians were under-represented in the sample across all sites, 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?13 As there were no continuous independent variables in the model, the dependent variable was binary and all observations were independent, the assumptions of logistic regression were met. 	 ?14 It should be noted that participants with missing data on any dependent or independent variable in the model were entirely removed from multivariate logistic regression analysis, as is the default in SPSS. Although this results in a reduction in the total sample size used in these analyses, it ensures that the same participants are used throughout the analysis, allowing comparisons to be made between the various independent variables in the model. The total number of cases removed corresponded to < 5% of the total sample. 	 ?	 ? 31	 ?with only 11.7% of all physicians in the population participating. While the total number of nurses (18.8%) and support staff (19.7%) in the sample were just below the targeted quota, the number of AHPs (24.5%) and administrative staff (21.1%) were above the 20% target (Table 2).   Table 2: Total population, targeted quota per hospital (15% sampling frame) and achieved sample by occupation (PCHW/non-PCHW)  Occupation Total Population Targeted quota (20%) Achieved quota (%)    Doctors 450 90 53 (11.8%)    Nurses 2052 410 386 (18.8%)    AHPs 302 61 74 (24.5%)    Admin Staff 975 195 206 (21.1%)    Support Staff 1301 260 256 (19.7%) Total HWs 5080 1016 978 (19.4%)15  2.3.1. Sociodemographic Characteristics   The remaining sociodemographic characteristics of participants are summarized in Table 3. As can be seen from this table, the sociodemographic profile of participants closely resembles the proportion of these characteristics observed in the workforce of the three hospitals, suggesting the sample reflects the population on these variables of interest. Approximately half (52.5%) of participants were PCHWs; three quarters (76.6%) were female; one quarter were in the 30-39 and 40-49 age cohorts (25.8% and 26.2% respectively), with 13.6% being under the age of 30 and 34.3% over the age of 50. In addition, workers from Universitas, Pelonomi and Bongani represented 45.6%, 35.0 and 17.9% of participants, respectively. A total of 70.8% were black, 19.5% were white and only 8.6% were coloured.       	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?15 Includes 3 missing values where an occupation was not provided	 ?	 ? 32	 ? Table 3: Sociodemographic characteristics of survey participants and HWs at the study hospitals  Characteristic Sample frequency (n) Sample % Population % Occupation      Non-PCHWa 462 47.4 44.8   PCHWb 513 52.6 55.2 Age      <30 133 13.6 9.9   30-39 252 25.8 25.3   40-49 256 26.2 28.8   >50 335 34.3 36 Race      Black 692 70.8 71.3   Coloured 84 8.6 6.8   White 191 19.5 21.5 Sex      Male 228 23.4 25.5   Female 748 76.6 74.5 Hospital      Universitas 446 45.6 45.9   Pelonomi 342 35.0 37.2   Bongani 175 17.9 16.9   Multiple 15 1.5 Unknown a Of the 461 non-PCHWs in the sample, 206 (44.6%) were administrators and 256 (55.4%) were support staff.   b Of the 513 PCHWs in the sample, 386 (75.2%) were nurses, 53 (10.3%) were doctors and 74 (14.4%) were allied health professionals.  2.3.2 General Perceptions Regarding HIV and AIDS   The results of questions in this domain are summarized in Table 4. Only 35.6% of HWs believed the majority (most or all) of their colleagues would want to know their HIV status.16  If informed that they were HIV positive, 42% believed that most or all of their colleagues would change their sexual behavior, 32% believed most or all would change their work practices and 72.5% believed the majority (most or all) would want to get treatment as soon as possible. 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?16 Although it may be interesting to conduct further analyses on this question based on sociodemographic variables, given the nature of the question and the sociodemographic data collected this would not be meaningful. Specifically, as the question asked participants about what they think about their colleagues? desires regarding their HIV status, and the demographic data pertained to the participants themselves, any associations detected would be difficult to interpret. As such, no further analysis was conducted on this question. 	 ?	 ? 33	 ?Participants felt that if a co-worker found out they were HIV positive, they are most likely to disclose this to their spouse/life partner (49.1%), private doctor (48.2%) and family members (39.7%). Few would share this information with their occupational health nurse (18.1%), supervisor (16.0%), friends (16.0%), close co-workers (14.4%), doctor in the area they work (12.6%) or union representative (3.5%).  Fourteen percent would not share the information with anyone.  Table 4: Results of general perceptions regarding HIV/AIDS  Item and Response Frequency (n) Percentage (%) Proportion of co-workers perceived to want to know their HIV status     Very Few 337 34.5      Some 252 25.8      Most  211 21.6      All 137 14.0 Total 937 95.8 Would change their sexual behaviour if HIV positive     Very Few 248 25.4      Some 275 28.1      Most  253 25.9      All 158 16.2 Total 934 95.5 Would change their work practices if HIV positive      Very Few 333 34.0      Some 286 29.2      Most  195 19.9      All 118 12.1 Total 932 95.3 Would want to get treatment as soon as possible if HIV positive      Very Few 93 9.5      Some 135 13.8      Most  223 22.8      All 486 49.7 Total 937 95.8 If HIV positive, a co-worker would most likely share this with:      Spouse/life partner 480 49.1      Private doctor 471 48.2      Family members 388 39.7 	 ? 34	 ?Item and Response Frequency (n) Percentage (%)      Occupational Health Nurse 177 18.1      Supervisor 156 16.0      Friends 156 16.0      Close co-workers 140 14.4      Doctor in charge of the area they work 123 12.6     Union representative 34 3.5     They would not share this information with anyone 137 14.0  2.3.3 Consequences of having HIV in the Workplace:   A total of 952 (97.3%) responses were received to this question. The majority of participants felt if a HWs boss found out they had HIV, they would be supportive (n=590; 60.3%) and that their boss would encourage them to get treatment (n=547; 55.9%). Many (403; 41.2%) also felt that the employer would encourage the HW to ?follow all procedures to protect themselves from exposure to infectious diseases while at work?.  An additional 225 (23%) stated that HWs relationship with their employer would stay the same.  Few worried that an HIV positive HW would be fired the first time they broke a rule (n=26; 2.7%), that they would not be promoted when they should be (n=58; 5.9%), and that they would be fired if they started missing work (n=34; 3.5%). 2.3.4 Perceptions of HIV Stigma in the Workplace:   A total of 950 (97.1%) participants responded to this question. More than one-third (n=377; 38.5%) indicated they believed that there was HIV stigma in the workplace. Bivariate analysis revealed a significant association between perception of HIV stigma in the workplace and a number of sociodemographic characteristics (Table 5), namely occupation, age and race. Specifically, a higher proportion of PCHWs compared to non-PCHWs perceived HIV stigma in the workplace (48.2% vs. 27.4%, p<0.0001). Similarly, a higher proportion of coloured and white participants perceived HIV stigma in the workplace as compared to blacks (51.2% and 	 ? 35	 ?42.0% vs. 35.5%; p<0.01). A greater proportion of participants under the age of 50 perceived there to be HIV stigma in the workplace compared to their older colleagues (41.2-42.4% vs. 31.0%; p<0.015). No significant differences in perceptions of HIV stigma in the workplace were observed between males and females (p<0.645) or based on the hospital the participants worked in (p<0.182).   Table 5: Association between sociodemographic characteristics and perceived HIV stigma in the workplace  Characteristics Perceive HIV stigma in the workplace [n (%)] Do not perceive HIV stigma in the workplace [n (%)] p-value Occupation   < 0.0001   Non PCHW 122 (27.4) 324 (72.6)     PCHW 241 (48.2) 259 (51.8)   Age   < 0.015   <30 55 (41.7) 77 (58.3)     30-39 105 (42.5) 142 (57.5)     40-49 103 (41.2) 147 (58.8)     >50 99 (31.0) 220 (69.0)   Race   < 0.01   Black 240 (35.5) 437 (64.5)     Coloured 42 (51.2) 40 (48.8)     White 76 (42.0) 112.1 (58)   Sex   < 0.645   Male 82 (36.9) 140 (63.1)     Female 281 (38.7) 446 (61.3)   Hospital   < 0.182   Universitas 168 (39.3) 259 (60.7)     Pelonomi 123 (36.6) 213 (63.4)     Bongani 63 (36.4) 110 (63.6)     Multiple 5 (35.7) 9 (64.3)   Based on the bivariate analysis, the variables age, race and occupation and all two-way interaction terms were entered into a multiple logistic regression model. As a result, a total of 935 participants (95.6%) with no missing values were included in the analysis. After completing 	 ? 36	 ?the model-fitting process described in the methods, all interactions were removed, as they did not contribute significantly to the overall model based on the LRT. The final regression model is presented in Table 6.17 When controlling for additional covariates, a significant relationship remained between the dependent variable (perception of HIV stigma in the workplace) and occupation (p<0.0001), but not race (p<0.063) or age (p<0.090). The adjusted odds ratio (OR) for a HWs occupational status demonstrates that when controlling for age and race, the odds of perceiving HIV stigma in the workplace among PCHWs is 2.38 times that of non PCHWs (95% CI 1.80-3.15).   Table 6: Multivariate logistic regression demonstrating adjusted association between sociodemographic characteristics and perceptions of HIV stigma in the workplace  Variable Adjusted Odds Ratio 95% Confidence Interval p-value Occupation       Non PCHW 1.00      PCHW 2.38 1.80 ? 3.15 0.000 Age          <30 1.00         30-39 1.03 0.66 ? 1.60 0.897      40-49 0.93 0.59 ? 1.44 0.733      >50 0.68 0.44 ? 1.04 0.077 Race          White 1.00         Black 0.90 0.64 ? 1.28 0.527      Coloured 1.60 0.93 ? 2.76 0.090  2.3.5 Knowledge of the OHS:   Of the 972 (99.3%) participants responded to this question, 723 (73.9%) knew where the OHS unit is located, whereas 158 (16.2%) did not, and 95 (9.7%) were unsure.  Furthermore, 606 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?17 Tolerance values were all well above 0.9 and no strong correlations were found among the independent variables included in the model, indicating multicollinearity was not of concern	 ?	 ? 37	 ?(62%) knew the operating times for the OHS unit, but 365 (37.3%) did not know or were unsure. Finally, 503 (51.4%) knew that HIV treatment is available at the OHS unit, however, 469 (48%) did not know or were unsure. Similarly, 543 (55.5%) knew that TB treatment was available at the OHS, however 206 (21.1%) did not and 226 (23.1%) were unsure. 2.3.6 Perception of the OHS:   Of the 926 (94.7%) participants who responded to the question about whether HWs use the HIV/AIDS programme in the OHS unit effectively, 350 (35.8%) responded ?always?, 226 (23.1%) responded ?most of the time?, 232 (23.7%) responded ?not often? and 118 (12.1%) responded ?not at all?. In addition, of the 962 (98.4%) of participants who responded to questions about OHS staff, 609 (62.8%) agreed that OH practitioners are well trained to offer HCT, only 50 (5.1%) disagreed, and 310 (31.7%) were unsure of how well they are trained.  Similarly, 603 (61.7%) agreed that OH practitioners encourage people to use the OHS unit for HIV/AIDS services, only 56 (5.7%) disagreed, and 303 (31.0%) were unsure. 2.3.7 Potential Reasons for not using the HCT Service:   When asked about specific reasons for not using the HCT services, 692 (70.1%) of participants indicated one of the reasons provided contributed to why the service was not being used. Specifically, 180 (18.4%) of participants indicated they were not given enough time during the work day to go to the OHS unit, 85 (8.7%) expressed concern that they would have to pay for the services, 93 (9.5%) saw no point if ARVs were not provided at the OHS, and 100 (10.2%) indicated the options provided were not applicable. By far the most frequently cited reason for not using the HCT service was due to a fear that confidentiality would not be maintained, with 380 (38.9%) of participants indicating this response.  	 ? 38	 ?2.3.8 Perceptions of Confidentiality in the OHS:    A total of 944 (96.5%) participants responded to this question. Nearly one third of participants (n=290; 29.7%) expressed concern that confidentiality is not maintained in the OHS. Bivariate analysis demonstrated a significant association between perceptions regarding confidentiality in the OHS and HWs? occupation, age, and race, but not sex or hospital (Table 7). Specifically, a higher proportion of PCHW thought that confidentiality is not maintained in the OHS compared to PCHW (37.5% vs 21.1%; p<0.0001). Similarly, a higher proportion of black and coloured workers shared this view, as compared to white workers (32.3% and 37.0% vs. 17.5%; p<0.0001). Lastly, a higher proportion of workers in the 40-49 age category, thought that confidentiality is not maintained in the OHS, compared with workers in the other age brackets (37.5% vs. 25.9-29.2%; p<0.017).     Table 7: Association between sociodemographic characteristics and perceptions regarding confidentiality in the OHS  Characteristics Think confidentiality is maintained in the OHS [n (%)] Do not think confidentiality is maintained in the OHS [n (%)] p-value Occupation   < 0.0001   Non PCHW 348 (78.9) 93 (21.1)    PCHW 312 (62.5) 187 (37.5)  Age   < 0.017   <30 92 (70.8) 38 (29.2)    30-39 183 (74.1) 64 (25.9)    40-49 155 (62.5) 93 (37.5)    >50 232 (73.2) 85 (26.8)  Race   < 0.0001   Black 457 (67.7) 218 (32.3)    Coloured 51 (63.0) 30 (37.0)    White 146 (82.5) 31 (17.5)  Sex   < 0.144   Male 164 (74.2) 57 (25.8)    Female 498 (69.1) 223 (30.9)  Hospital   <0.207   Universitas 299 (71.5) 119 (28.5)  	 ? 39	 ?Characteristics Think confidentiality is maintained in the OHS [n (%)] Do not think confidentiality is maintained in the OHS [n (%)] p-value   Pelonomi 241 (71.1) 98 (28.9)    Bongani 112 (64.7) 61 (35.3)    Multiple 12 (85.7) 2 (14.3)   The variables age, race and occupation and all two-way interaction terms were subsequently entered into a multiple logistic regression model, in which 928 participants (94.9%) with no missing values were included in the analysis. The aforementioned model-fitting process resulted in the removal of all interaction terms, based on non-significant LRTs. The final regression model is presented in Table 8.18 When controlling for additional covariates, a significant relationship remained between the dependent variable (perception of confidentiality in the OHS) and the covariates for occupation (p<0.0001) and race (p<0.0001), but not age (p<0.053). The adjusted odds of perceiving confidentiality is not maintained in the OHS among PCHW was 2.39 times (95% CI 1.77-3.23) that of non-PCHW. The adjusted odds of perceiving confidentiality is not maintained in the OHS is higher among black (OR 2.71, 95%CI 1.76-4.18) and coloured (OR 3.01, 95%CI 1.63-5.56) HWs, compared to white HWs.   Table 8: Multivariate logistic regression of the adjusted association between sociodemographic characteristics and perception that confidentiality is not maintained in the OHS  Variable Adjusted Odds Ratio 95% Confidence Interval p-value Occupation      Non PCHW 1.00     PCHW 2.30 1.77 ? 3.23 0.000 Age         <30 1.00        30-39 0.80 0.49 ? 1.30 0.897      40-49 1.30 0.81 ? 2.09 0.733      >50 0.84 0.52 ? 1.34 0.077 Race    	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?18 Tolerance values were all well above 0.9 and no strong correlations were found among the independent variables included in the model, indicating multicollinearity was of no concern	 ?	 ? 40	 ?Variable Adjusted Odds Ratio 95% Confidence Interval p-value      White 1.00        Black 2.71 1.76 ? 4.18 0.000      Coloured 3.01 1.63 ? 5.56 0.000  2.4 Summary of Findings  These results demonstrate that knowledge of the OHS and perceptions of the effectiveness of the OHS staff did not appear to be lacking; nor were there concerns about how an employer would treat a HW after finding out their status. However, some clear barriers to uptake of HCT within the OHS were identified. First, the majority of HWs (64.4%) believed that their coworkers would not want to know their HIV status.  If a HW did know they were HIV positive, nearly half (48.2%) of the participants believed the HW would disclose their status to their private doctor, as compared to only 18.1% that would disclose to their OH nurse. Within the workplace, far less than 20% of participants believed their colleagues would disclose to a supervisor (16.4%), close co-workers (14.4%) or doctor in the area they work (12.6%).  14% of participants believed their colleagues would not disclose to anyone. Furthermore, more than one-third (38.5%) of HWs believed that there was HIV stigma in the workplace.  When controlling for age and race, PCHWs had 2.4 times the odds of perceiving HIV stigma in the workplace when compared to non-PCHWs.   Of the potential reasons for not using the OHS service specifically, a fear that confidentiality would not be maintained was the most commonly cited factor, with 38.9% of participants indicating this response and with nearly one-third of participants expressing concern that confidentiality is not maintained in the OHS. When controlling for age and race, PCHWs had 2.3 times the odds of perceiving that confidentiality is not maintained in the OHS, as 	 ? 41	 ?compared to non-PCHW.  In addition, black and coloured HWs had higher odds (OR 2.71 and 3.01 respectively) of perceiving confidentiality is not maintained as compared to white HWs.   	 ? 42	 ?Chapter 3: Focus Groups Interviews (FGIs)  3.1 Aim To corroborate and identify additional barriers to uptake of HCT service and to understand how these identified barriers are contextualized within this population, in order to solicit feedback on how to overcome such barriers.  3.2 Method  3.2.1 Sampling and Study Sites  Focus groups interviews were conducted on a purposive sample of HWs in Free State Province, South Africa in August 2012.  Participants were selected to ensure HWs from an array of occupations were included. In addition, to ensure representation of HWs from the three hospitals of interest in this study, 5 FGI?s were conducted at Pelonomi Regional Hospital (n=2), Universitas Academic Hospital (n=1), Bongani Regional Hospital (n=1) and at the Department of Health, herein referred to as Bophelo house (n=1). The study participants were recruited at each site by third party recruitment. This third party actively recruited HSC representatives and other HWs through email and word of mouth, based on the following inclusion criteria: 1) the participants had to be employed as a HW at one of the interview sites 2) the participants had to be able to speak and read English. Third party recruitment was appropriate as the author was a researcher from Canada, who may not have context on how proficient participants may have been in English.  	 ? 43	 ?3.2.2 Focus Groups  This study used an inductive qualitative approach, as this approach allows for modification of concepts and relationships between concepts throughout the process of doing research, with the goal of most accurately representing the reality of the situation (Thomas, 2006).  To explore HWs views on why occupationally based HCT services may be underutilized, the author initially organized four FGIs with HWs at the three hospital sites where the survey had been conducted. During the course of recruitment, however, HWs from the Department of Health at Bophelo house approached the author to ask if they could be included in the study. These HWs were subsequently assured that this was possible, but only if the opportunity to participate was advertised to all of their colleagues. Third party recruitment followed and thus, a fifth focus group with HWs from Bophelo house was added.  The five semi-structured FGIs lasted 1 hour each and consisted of 5-9 participants per group. The interviews explored HWs reasons for using services other than the OHS for HCT and solicited feedback on what could be done to improve uptake of the service. All FGIs were audio taped and subsequently transcribed verbatim.  Although notes were taken during the FGIs, all quotes presented are taken directly from the transcriptions. All quotes are cited based on a randomly assigned transcript number (see Table 7).  3.2.3 Introductions and Informed Consent  Ethics approval for this study was obtained from the University of British Columbia Behavioral Research Ethics Board. All interviews began with an introduction to the author (as the interviewer), the purpose of the focus group, the study aims and a reading and clarification of the informed consent form.  The interviewer highlighted all perceived risks and benefits. After learning that some participants may have felt pressured into attending the FGI (or given short 	 ? 44	 ?notice), after the first two FGIs participants were assured that there would be no repercussions if they chose not to participate and that they were able to leave if they wanted to. The interviewer acknowledged her accent and tendency to speak fast, and encouraged participants to ask for clarification if anything was unclear during the interview or if they were unable to understand the interviewer at any time.   Next, the interviewer clarified that the purpose of the focus group was to understand participants? opinions on topics related to HIV and AIDS services offered at the OHS.  Participants were asked to respect each other?s opinions and were asked if they were comfortable disagreeing with each other. Furthermore, they were encouraged to give each other respect if they did disagree.  Participants were further reassured that there were ?no right or wrong answers?.  They were made aware that the focus group was going to be recorded.  They were asked to speak one at time in order to preserve the quality of the recordings and to ensure their anonymity in this process, they were asked not to refer to each other except through numbers they had been assigned.  Participants were reassured that if they were uncomfortable at any time they could leave the interview or ask to speak to the interviewer during or after the interview.  Finally, although the ability to speak English was part of the inclusion criteria, participants were given the option to speak in their native language if they were more comfortable doing so. The group decided that if this was done then another participant would translate what was said in a manner that the group (and interviewer) could understand.  3.2.4 Interview Guide  Five semi-structured interview questions guided the discussion: 1) Where do health workers go most often to get tested for HIV? 2) Why do they go there? 	 ? 45	 ?3) Why do they choose to not go to the OHS? 4) Can we list solutions to make the service better? (or, as a follow-up) How can we improve the OHS service? 5) Which is the best solution?  A cursory transcript was typed immediately following the FGIs, so that preliminary analysis of the data from the first three interviews could be done prior to the final 2 FGs. The first three FGIs identified several recurring concepts, and saturation of the data was noticed in some areas. It was evident after the first two focus groups that that after question 2, the response would often be ?stigma? or ?confidentiality?.  This may have been a consequence of having included these concepts in the informed consent. Thus, a follow-up question asking the participants to define stigma, or to specify why they perceived confidentiality was not maintained, was included in lieu of Question 3. Similarly, after the first two FGIs, it was clear that question 4 and 5 were difficult for participants to answer. Thus, in order to identify facilitators to OHS use, the question was changed to, ?If you were someone with HIV or you knew a colleague with HIV, how would you want to be treated in the workplace?? or ?What should we do for colleagues who are HIV positive?? The final 2 FGIs concentrated on areas such as recommendations to improve the service, where it was identified that saturation had not been reached.    3.2.5 Focus Group Interview Data Analysis The data was reviewed in an iterative manner. The inductive approach used for data analysis was conducted using David Thomas?s (2003), A general inductive approach for qualitative data analysis. First, the raw data files of the audio recordings were transcribed verbatim. Identifiers were added to denote the participants and the interviewer speaking within the transcripts.  Second, transcripts were closely read to understand the ?themes? and details 	 ? 46	 ?emerging from the raw text. Third, within themes, sub-categories were organized in a ?code book? to capture the key analytic data. To do this, the interviewer worked with her co-supervisor (JAB) to identify categories through a process of extensive manual coding of the first 2 transcripts. This was initially conducted independently and in duplicate by JAB and the interviewer, followed by a discussion to establish consensus of the final categories from these transcripts to be used as categories against which to analyze the remaining transcripts.  Once these categories were assigned, the interviewer completed manual coding for the remaining transcripts through a constant comparative approach. From this process, 26 sub-categories were identified.  Thus, the initial ?code book? was refined throughout this analytic process.  To reduce overlap and redundancy among sub-categories, categories were revised and refined by collapsing sub-categories into major categories, which led to the 6 themes19 described in the results: locations for testing, privacy, confidentiality, gossip, stigma and facilitators.  Categories and Subcategories Presented It should be noted that categories presented directly reflect the aims of the study and are by no means inclusive of all responses of barriers identified.  For instance, ?disclosure to partners? was of concern to many participants, however, analysis of these sub-categories, while important to the overall issue of why a person may choose to not test, was not deemed to be relevant to testing at the OHS specifically.  Overlap between and within Categories  As responses showed overlap between many categories, the decision was made that while some quotes may overlap between two categories, they will only be presented in one category.  To determine the ?best fit? for where to place an overlapping quote, the decision was made to remain as true to the literal meaning of the statement, rather than analyzing the quote based on an 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?19 The term ?theme? refers to the combination of a category with its sub-categories.	 ?	 ? 47	 ?ascribed value.  For instance, a participant attributes the lack of willingness to access HCT in hospital to what she perceives to be workplace culture. She states,  They will infect us20. We treat them as if they are not part of us anymore21 so the fear of going to clinics i[n] the hospital where you are knowing that you know some of the people there22. Is that what are they going to talk [about] when I am from there23?   This quote illustrates how one statement can have many dimensions, as is common in qualitative analysis. For instance, ?Is that what they are going to talk [about] when I am from there?? could be both a fear of gossip and a fear of perceived stigma. In this case and as all data cannot be presented in the results of this thesis, the quote would be placed in ?gossip? (while supporting the detection of fear of perceived stigma as well). Thus, while overlap remains within statements or between categories, there has been an attempt to minimize or highlight areas of overlap (depending on when they occur), while trying to preserve the context in which the statement was made.  3.3 Results  The overall participant group consisted of 38 HWs, interviewed over 5 FGs (Table 9). 2 participants were not included in FG analysis. The first asked to leave at the beginning of the interview (FGI 5) and thus did not provide informed consent.  While the second did provide informed consent and remained for the entirety of the FGI, none of her responses were included in the analysis, as she declared at the end of the FGI that she was uncomfortable being in the group from the beginning (FGI 3).  Various HWs were represented in the FGs, including professional nurses; assistant nurses; a nursing student; household aids; HIV and AIDS coordinators, managers and administrators; a social worker; a physiotherapist; and secretaries, 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?20 Coded as: Instrumental stigma	 ?21 Coded as: Discrimination 	 ?22 Coded as: Privacy	 ?23 Coded as: Internalized Stigma	 ?	 ? 48	 ?security officers, admin clerks, drivers and technical assistants. No physicians participated in the FGIs.   Table 9: FGs with number of participants in the group  Transcript (T#) Number of FG participants 1 8 2 5 3 9 4 7 5 9  From analysis of the transcripts, data was collapsed into 6 themes: 1) locations for testing; 2) privacy; 3) confidentiality; 4) gossip; 5) stigma; and 6) facilitators. The first five themes correspond to barriers to uptake of HCT services, whereas the final theme, ?facilitators?, corresponds to recommendations to improve uptake of HCT at the OHS.  3.3.1 Theme 1: Locations for Testing  The first question asked in all FGIs was ?where do employees [HWs, you] go to get tested for HIV?? All FGs identified private doctors as the primary location for HWs to get tested for HIV.  Other options included the poli-clinic24, non-governmental organizations (NGOs) such as Population Services International (PSI)25, outreach programs26, free clinics27, other wards within the hospital and the sick bay (a synonym for the OHS).  Participant felt that a HW was unlikely 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?24 The poli-clinic is a specialized ward within the hospital where HIV testing and treatment is provided. A participant described the difference between the poli-clinic and the OHS as:  ??poli-c they are those who are for ARVs. That is the identified clinic so when you are there everybody knows you are there because of treatment. Its unlike in sick bay [OHS]. Sick bay? you just enter the consultation room of the cubicle with one professional nurse who will attend you?? (T2)	 ?25 The participants who identified this service described it as ?an agency that is responsible for testing, usually its community based and encourages the community to test and know their status.? (T3) 	 ?26 Conducted by the Department of Health and at department stores  	 ?27 Participants in this instance described a generic ?free clinic?.  	 ?	 ? 49	 ?to choose any option other than a private doctor for HCT, except in cases were finances were limited:  People of higher status would not go queue in a clinic, they would rather prefer to see their private doctor?whereas other people don?t have access to private doctor or money to go see a private doctors so they have no choice but to cue in the clinic to get their results. (T2)   When asked if a HWs occupation reflected which HWs had access to private doctors (due to associated differences in income), one participant explained: ?even the cleaner in the government can be a member of the medical aid?The access to medical aid is much more so I will go to my private doctor.? (T2) Thus, it would appear that HW across a range of cadres preferred to visit a private doctor for HIV services.     3.3.2 Theme 2: Privacy  Privacy is defined as ?selective control of access to the self or to one?s group? (Altman, 1976).  Privacy is not a static term but a dynamic process of setting interpersonal boundaries where openness to and closeness from others shifts depending on circumstances (Altman, 1976). As a regulatory mechanism, the desire for privacy tries to achieve an optimal level of social interaction (Altman, 1976).  However, imbalance in a persons desire for privacy takes on two forms: overstimulation, which results from high amount of social inputs that leads to a feeling of intrusion or invasion of personal space, and under stimulation, or too little privacy, which results in isolation from social interaction (Altman, 1976). Regulating privacy requires verbal, nonverbal and environmental mechanisms (Altman, 1976).  When participants explained their reasons for preferring a private doctor to the OHS for HIV testing, multiple participants cited ?privacy? as the primary factor. In this instance, privacy reflected an individual?s desire to access health services without others knowledge that they have done so.  Thus, privacy was not a reflection of the service itself (which speaks to issues of 	 ? 50	 ?confidentiality), but rather the personal preference of the individual, as exemplified in the following quote:  So some want to make it a private matter. Even for myself, I would want to make it a private matter. So they attend private doctors for medications also. Its rare where you find them coming straight to occupational health unless the person is really sick. (T4)  HIV Testing within the Workplace  Participants also noted the location of the OHS as a service within the hospital was a key concern. Undergoing HIV testing in a setting that was visible to co-workers was seen to undermining privacy, as captured in statement ?So they don?t feel like they are ready to talk to people in the workplace?they would rather go to the private doctor where no one knows she went to the doctor.? (T5) This visibility was associated with the fear that co-workers would begin to ask questions about a HWs HIV status: ?Like, if you see me go into the clinic and testing for HIV, people start talking and asking why are you going there?? (T2) This sentiment extends further, as participants also expressed a reluctance to have a co-worker conduct their HIV testing and subsequently have knowledge of their status:  ??you don?t want to go somewhere where there is someone that you know that knows your results.  So you need something that is private with your doctor only.  It?s not that there is no privacy there. No. You just feel because you are working here, you cannot test blood here.? (T3)   Accessing OHS for Testing  Some participants suggested a the reason for wanting to maintain privacy is because in the community, anyone observed entering a free clinic is presumed to be HIV positive:  There is a group of people there in the free clinic, where people, because of the literacy level also, they see you sitting there forming in a queue. It?s not a question of has he tested positive, it?s a question of ?wow, he?s also positive??so those are some of the reasons we we [sic] want to preserve privacy. (T3)   	 ? 51	 ?Similar sentiments were expressed about HIV services within the hospital. For instance, it is assumed that all patients who attend the poli-clinic are HIV positive, as this is a place where only HIV positive individuals go to be monitored and to receive medication. The perception that the OHS is not private appeared to be dependent on how the service was perceived by others when an individual went to access the service. Specifically, participants explained that HWs may assume an individual accessing the OHS is HIV positive if they met with an OHS employee who was solely responsible for providing HIV services only during a given time period: ?If [persons name] and [persons name] are doing HIV and TB for the whole month and you go there, they will know that I am having TB or HIV.? (T5) Thus, the provision of HCT service by one OHS employee was perceived to contribute to a lack of privacy. Furthermore, although at one site OHS nurses alternated providing services, this was not felt to be sufficient to ensure privacy, as within a given employee?s allocated time for providing HIV services, the privacy of HWs visiting the service may be compromised:  P1: I think it?s their allocation and its change time and again. Like in the ward, I am the one doing the medication, then next month its this one doing injections. It?s the allocation, but it changes. P2: But if she is doing it the whole month?  Members of the Focus Group: Yeah, yes P1: But will they know she is doing it for the whole month? P2: People are observant. [All focus group members laugh and agree] (T5)  This conversation outlines how privacy may be perceived to not be regulated within the OHS, due to verbal (talking about why a person is accessing the service or shift allocations of nurses), nonverbal (deducing who is working at the OHS due to their extended time delivering one service) and environmental (the way in which the OHS organizes HIV service allocation) mechanisms, which operate to create a sense that there is a lack of privacy at the OHS.  	 ? 52	 ?3.3.3 Theme 3: Gossip   Gossip is defined as ?evaluative talk about a member of the discussants? social environment who is not present? (Wert and Salovey, 2004).  Gossip is one of the oldest forms of communication and approximately 60% of adults converse about individuals who are not present (Wert and Salovey, 2004; Levin and Arluke, 1985).  Gossipers make a comparison between the excluded individual and some social reference point, including social norms or personal perspectives (Wert and Salovey, 2004).  This evaluative talk can be negative, however, operational definitions of gossip are difficult to state on this basis, as ?negative? depends on the context which is outlined by the mutual understanding of those who are gossiping (Wert and Salovey, 2004).   Another barrier to the uptake of HCT services was gossip related to an individual?s HIV status. Participants noted that a reason why HWs may not want to get tested for HIV at the OHS was due to a fear that their colleagues would gossip about them. Defining and coding the theme ?gossip? was difficult, because as a category ?gossip? has dimensions that overlap with the themes of ?confidentiality? and ?stigma? and may offer an explanation for HWs desire for privacy. However, gossip has been included as a separate theme, as participants felt that above and beyond the concepts of stigma and confidentiality, that the behavior of gossiping was also a significant barrier to a HW accessing the OHS for HIV services in the workplace:    That?s the main thing that everywhere you go, people will be talking behind your back. ?Look at him, he is HIV positive? or ?she?s HIV positive? or something. So, I think the main thing according to me is people gossiping about you. (T5)   Workplace Gossip about HIV Status  Gossip about a person?s weight loss was a reoccurring example of how and when HWs engage in gossip that was assumptive about an individuals HIV status: ?when people will see that 	 ? 53	 ?you are having weight, and you lose weight, they will say you have AIDS.? (T4) Participants noted that when HWs learned a co-worker has accessed HIV services, gossip about their co-workers HIV status often began a cycle of whether they were labeled: ?Cause this will say he?s negative, this will say he?s positive, then it depends what spreads much more better than the other.  Hence, people ought to say I?d rather go to a private doctor because of 1,2,3.? (T3) Furthermore, when HWs gossip about a co-workers status, it appeared that the default was to assume that the co-worker was HIV positive, even in the presence of evidence to the contrary: P1: Even if you have a little needle prick, anybody, everyone should now want to know, what is the results? ? P3: Even if the name is not there. It has happened a number of times. We have followed the protocol, they take prophylaxis, and they rest and they remain negative for the rest of their lives. But that still remains that ?he was pricked by an HIV positive needle? and ?he is definitely positive? because we do not know the outcome.  (T3)   Participants noted the problem with gossip is that it makes the workplace unsafe as a place to disclose one?s HIV status: ?Instead of saying lets support him, we are waiting for that person to fall to confirm [a positive status], and you can see that some people are talk about you, looking at you, at the distance they look at you, then they hide cause they see you looking at them. That is what makes people afraid?(T4)?  Gossip by Professional HWs   Gossip by professional HWs is one area where the theme ?gossip? overlaps with the theme ?confidentiality?. As professional HWs are responsible for ensuring that confidentiality is maintained, gossip by nurses had implications on perceptions of confidentiality not being maintained by professional HWs. Some participants noted that a fear of professional HWs gossiping about their HIV status (and thus breaching their confidentiality) was a factor in preventing testing, stating:  	 ? 54	 ?As nurses?if we talk, people talk and if I told number 7 and number 1 start looking at me funny, I?m going back to number 7 because she told number 1, so that?s why people don?t disclose. Because you tell one person and then its spread like wildfire and that?s not what people want.  You don?t want to be diagnosed and then its going to spread throughout the whole hospital. (T4)   In addition, gossip by HWs appeared to extend beyond the workplace: I think confidentiality starts with us, because some of the professional nurses, they talk about the other person in the taxis. ?You are [sic] so and so. Number 2?she is HIV positive, do you know her?? ?Yeah, that lady is working in...? ?Yeah.?  Something like that. Most people are talking about the people in the taxis28 (T1).      Furthermore, participants expressed the belief that others role model the behavior of professionals, demonstrating the impact of gossip perpetuated by these HWs:  Most of the lay persons who are in our professional settings, they do things because they see them being done by professionals. Like for instance, I come across a patient that is HIV positive, then I end up telling somebody else. Then she goes and tells you that I told her, but she is not a nurse or whatever. Then you go to the next person and by the end of the day 5 people know from a professional person. (T4)  There was also evidence to suggest that in some instances, gossip perpetuated by professional HWs may result in a breach in confidentiality:  I think we are also to be blamed, certain people come to you to disclose to you and something?s get blown out of proportion.  We know confidentiality needs to be maintained but we like to gossip. My supervisor says this person is HIV and I go to my friend. It?s human nature and that?s what is killing us. (T4)  These comments illustrate that beyond speculative gossip within the workplace, gossip that breaches confidentiality undermines trust in professional HWs themselves. 3.3.4 Theme 4: Confidentiality  Confidentiality differs from privacy, in that privacy is the relationship of an individual with others and their environment that establishes boundaries (Altman, 1976), whereas confidentiality is the maintenance of those boundaries by the system or professional charged 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?28 Taxis are primarily used by black HWs as they commute to and from their homes.  Thus, this example may also suggest a racial dimension to gossip.	 ?	 ? 55	 ?with protecting a patient?s right to privacy (Kling, 2010)29.   A participant noted that confidentiality differs from privacy: ?people do go to the private clinics, neh, because of confidentiality issues, ne, privacy, neh.  Sometimes confidentiality doesn?t apply to secrecy and privacy.? (T4) In the FGI?s it was clear that confidentiality was a concern for most participants, as it was brought up in all FGIs and captured in multiple comments including: ?sometimes for confidentiality we are not sure?, ?the thing is they must have confidentiality? and ?they are not even sure that confidentiality is there, even the professionals?, to quote a few.  Trust  There appeared to be a perception among HWs that there is a lack of confidentiality in the OHS and thus a lack of trust in the service, as demonstrated in the statements ?Confidentiality, it plays the most important part because it means I can trust that person? (T1) and ?I think trust, trust in place is the biggest thing in the sick bay [OHS].  When I talk to the professionals?I have to trust in that particular person. I think trust plays an important role.? (T1) This lack of trust appeared to be related to a sense of involuntary disclosure of personal information, as noted by an OHS employee ?I think the employees don?t trust what we do here. They think we divulge their information.? (T2) It appears that the number of HWs employed at the OHS may also impact perceptions of confidentiality and trustworthiness of the service, as demonstrated by the quote: ?In the private sector its one on one; only the doctor and me as the client. But at occupational health I think also because we are not alone, they are another three people?its?not the right way of doing things?as long as we can keep the confidentiality? (T4) In turn, some participants noted that this lack of trust and perceived lack of confidentiality caused some HWs to avoid the OHS when they needed to get tested for HIV: ?others, we test them in the ward because we maintain confidentiality? (T4).   	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?29 For a complete definition of confidentiality, see Chapter 1 	 ? 56	 ?Electronic Breaches in Confidentiality  There seemed to be ambiguity about how files within the OH units were kept. As electronic files were available to many employees within the hospital system, there was a perception that electronic files were not safe.  This fear that electronic files were not confidential may also explain why some HWs preferred to the use of a private doctor for HIV testing and treatment: They decide to be tested in private centers and take it, maybe for treatment they also go private cause it?s a highly confidential matter, because?like for instance, when you take blood in the hospital, they are being computerized. So anybody who is computer literate can open and check my results. (T4)    Focus group participants believed that most HWs have access to the computer system.  Within the hospital, nurses and physicians were thought to have the widest access, supported by the statement ?professional nurse, they have a code to the Meditec so they can see if I am HIV positive.? (T5)  Thus, given that OHS employees are professional nurses and doctors, the conduct of these individuals with electronic access to information about all HWs HIV status had implications for how trustworthy the service appeared. In addition, it seemed that the suspicion around electronic records was not just based on a perception that the system was unsafe, but also due to known breaches in confidentiality: ?one time we had problems with the results that were exposed on the computer? (T4), ?Because when one persons was admitted in the hospital the results will show on the Meditec system.  It was lucky because she was negative? (T5). One participant believed that curiosity was sufficient to make a co-worker want to access another co-workers records: ??sometimes people are too curious sometimes, and people can just go get your PM30 number and get access in the hospital and they can just check on the Meditec.? (T5) 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?30 Employee ID number	 ?	 ? 57	 ?Although changes to the coding for electronic files were discussed during FGIs, no consensus was reached on the type of coding that might improve the current system. 3.3.5 Theme 6: Stigma   Stigma was the most expansive theme related to ?barriers?, supported both by the most comments and the most participants speaking to it. In one FGI, there was unanimous agreement with the statement, ?Stigmatization is the main problem for the health worker.? (T5) Throughout the FGIs, the term ?stigma? was used in a variety of ways, often equating stigma with discrimination, such as: a) Enacted stigma: ?It?s like a derogatory label attached to a person?followed by a discrimination? and b) Symbolic stigma: ?people get discriminated?people just always assume the worst, so you get stigmatized, you get ostracized, because now you are HIV positive?. However, even though discrimination, as a form of ?external stigma? was detected, a number of other typologies pertaining to ?internal stigmas? were detected: c) Perceived stigma: ?People are going to discriminate you so you are not going to get the same treatment, the same love, the same attention like before?; and d) Self-stigma: ?Yeah and somebody, neh, if he feels, if he knows his or her status, he is going to be stressed and he is going to be sick more, that?s why they afraid to go test.? (T5) One participant even expressed an understanding of the multiple dimensions of stigma, stating:  Stigma is not just one thing. It?s community based?its how people would perceive you?its just internal about yourself and how you will view yourself?Stigma may also be like a feeling, to say you don?t know what I am thinking of but you might fear that I might say, ?Oh, she is from the clinic, she is already positive? (T3)  Thus, if a participant used the word ?stigma? they were asked to define the term, so that the interviewer could contextualize their understanding of the term and identify the type of stigma detected by their statement.  	 ? 58	 ?External Stigmas (i) Enacted Stigma  Enacted stigma was detected in comments made by some participants who felt that discrimination by employers due to knowledge of a HWs HIV status would result in negative consequences: ??also at the workplace now when I have made my status been known then I will not be receiving the housing allowance or subsidies anymore, I will not be given a subsidy car, I?ll not be promoted? (T2).  However, it should be noted that there was much debate on whether this fear was justified: ?In your work situation they never treat you differently. I have never seen it.? (T3) (ii) Instrumental Stigma  Instrumental stigma, which is defined as the fear of infection from causal transmission (Herek, 2002), was detected in the following statement:  Even if the sister in charge of whomever, when I [saw] some people having [food], [I] said can you please borrow me your spoon or your cup, [they said] no don?t borrow that one because she is HIV positive. (T1)  This type of stigma may be discouraged through sharing knowledge of HIV transmission between HWs, as demonstrated in the following anecdote: The other day I had a lady who talked, she came to me to talk about another lady who is positive, it?s a known fact. The lady shared it and she told me she?s not going to the bathroom because maybe that lady used the toilet [2 participants laugh in response]. And then I was looking at her and I said but you know there is no way, even if you kiss, French kiss her you will not get it. (T1)  However, as FGIs progressed it became evident that even in the presence of knowledge regarding HIV transmission, this type of stigma persisted: P3: Even the things you are using, the plates of whatever. I know about HIV and how does it get to be infecting other people, it won?t be through the glass.  R: What you are trying to say is people won?t share  Many voices: Plates, cutlery 	 ? 59	 ? P2: They are worried that she was using the glass. I am not going to use it.    P3: Plates also.  P1: Its very difficult to disclose, your life will change tomorrow. (T3)  Participants suggested that professional HWs may share responsibility for this form of stigma.  As one nurse explained:  One day I have the new student nurse. We teach them the wrong things, because we want short cuts. The nurse will say when you get a get a bed pan or you touch a patient wear gloves or whatever, but when you come, you say just go don?t wear gloves or whatever. But when it comes to a patient ?Hey! Never do that without gloves for that one? and then she says why and you [participant whispers] ?She?s HIV positive.31? (T4) 	 ?(iii) Symbolic stigma    Symbolic stigma is a moralistic shaming and blaming (Scrambler, 2009). As one participant stated in reference to why disclosing status is difficult, ?People think you have been living in a manner that is immortal or not what society thinks is right so that?s why.? This fear of being blamed for contracting HIV was further elaborated in the statement:  When I am HIV positive and I want to go to my family or my mother to confess this thing, it becomes quite difficult because we have different cultures. Some of our people tend to judge us, like you have been behaving like this, so this might be the reason. (T3)  Participants from a number of FGs commented that HWs fear that if someone?s positive status were disclosed they would be perceived to be promiscuous:  I think its mainly like being labeled in a negative way, because look she is HIV positive and as we know HIV, we all think of it in a bad way, all of us. We don?t think we can get it from a needle prick of anything. Anybody just associated it with promiscuity and something so everybody is afraid of that. They worry about how people look at them, how people see them and their behavior.  So it?s all about negative labeling when you think of stigma. (T5)   This negative labeling represents one of the main barriers to a person?s willingness to be tested or disclose their HIV status:  HIV and AIDS is not like any other disease. It?s associated with behavioral, like you are not behaving right, you were just going around, sleeping around, so there is a lot of 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?31 There is overlap in this example with discrimination or directed stigma.	 ?	 ? 60	 ?stigma attached to HIV. That?s why I am saying lots of people don?t know their status. (T2)  A participant described that some even believe that HIV is not only associated with promiscuity, but with the perception that HIV positive individuals contracted the disease after resorting to sex-work: When AIDS was for the first time being diagnosed it is if people are being promiscuous. They are just sleeping with every man or every woman, that is how people saw them. Now?they will say that ?I am a whore?. So that is the stigma that goes with it. A certain stigma that these people who are poor because they cannot work. In order to get money to survive they go to that. So it goes to that. (T4)  An OH nurse noted that the fear of being viewed as promiscuous was so pervasive, that even in the absence of a reason to suspect that an OHS employee would blame a HW if they decided to test, their perceived stigma would prevent them from accessing HCT: ?how you contracted that one defeat all the other things we can say. They think that now we are going to discriminate and say you came because of this.? (T2) This is supported by the statement that regardless of the mode of transmission, an HIV positive HW will be perceived to have contracted the disease through a sexual route:  People do not know you had a needle prick so you got the disease through that. In peoples minds, you are promiscuous, you are having a string of partners, they don?t even think it can happen that you can get it accidentally - at an accident scene you were having a cut, it mixed with blood and then you contracted it. And even partners, they don?t think that way. They will just say, if you are a man, you will have a string of ladies and now you have brought it to me and vice a versa (T4)  However, as one nurse demonstrates, the attitudes of those who test may also be promoting this form of stigma: ?This is very difficult area for us as HWs to reach.  To say don?t do it32.  I want to believe that 80% of our community knows about HIV but they just choose a mischievous route.? (T3) 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?32 The nurse explains that it is difficult to promote ?one partner? as a prevention strategy.  Don?t do it here refers to ?don?t have multiple partners.? However, the same nurse also calls implies that if someone does have HIV its because they ?chose a mischievous route?.  	 ?	 ? 61	 ?Some participants also agreed the perception that HIV is a consequence of promiscuity may be particularly difficult to address among black HWs, due to lack of open dialogue about sex: I think you must realize that here in South Africa, especially in the black culture, sex is not a topic that is openly discussed. It?s an adult thing. Its things that are being done you know in a place where it is only two people and the like and you don?t even tell the kids that when you come on earth it was between two people and all those things [giggles] you come with stories to tell children where they are coming from. You don?t just tell that because of two people mate and then you were made.  So sex is not an issue that is discussed easily and openly especially amongst the blacks. (T4)  Internal Stigmas  (i) Perceived and Felt Stigma  Perceived or felt stigma is defined as a stigmatized persons internal sense of shame and fear of persecution. This form of stigma is one where an individual asks questions such as ?how will others view me??33 There was a persistent fear detected in all 5 FGIs that knowing ones status was linked with being treated differently, captured in statements like ?To be stigmatized you are going to be treated differently? (T1) and ?I am positive in my family?but others, definitely they are going to treat me differently.34? (T3) The issue with perceived stigma is that in many ways it is fear of the unknown reaction of others.  One participant identified this, stating, ?We are still fighting fear of the unknown, how people are going to see me, how people are saying about me? (T3) The preoccupation that others know your status and will treat you differently, by virtue of this knowledge and assuming that different has a negative connotation, was illustrated in the statement:  Stigma, if one can elaborate on the stigma, its carrying an emblem?cause they want you to say ?I?m HIV positive?.  When he smiles at me, he knows that it?s alright to come 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?33	 ?As opposed to self-stigma, where the individual asks ?how will I view myself??	 ?34 No participant openly disclosed their HIV status.  This statement reflects the participant stating that it is easier for an HIV positive person to feel safe at home ? that they can be ?positive in their family? but that others, presumably friends, coworkers, strangers, would treat them differently.  	 ?	 ? 62	 ?because he does not know that I am HIV positive. The stigma is what I carry that?s haunting my mental status?Having done the consent to say ?I feel I want to do this? but is then saying ?why did I test?? I am having this now. And that emblem that I am carrying in my mind, that heavy load in my mental status...If someone out of a gesture cracked a joke with you, ?hey, he knows about this??That is stigma. (T3)  This statement reflects the effect of perceived stigma on the mental health of an individual, and how the presence of such stigma may impact the decision to test and subsequently disclose one?s HIV status.   Finally, felt stigma may occur due to the perception that one may discriminate against the individual if their status becomes known. This is where ?gossip? may intersect with this form of stigma, as those who gossip may worry that they may be targets of gossip themselves:  You worrying yourself what are people going to think of me? And you don?t really worry about your health?maybe you have stigmatized some other people [agreement from FG participants], you have looked at them, laughed at them, now you are having this and you thinking, ?Oh, I laughed at this person now what?s going to happen to me?? I think that?s one of the major things. (T5)  	 ?(ii) Self-stigma / Denial  Self-stigma is characterized by low self-efficacy and low self-esteem (citation). It can be identified when an individual asks, ?why try?? based on ones? view of themselves.  The interviewer defined instances of self-stigma based on a HWs choice not to test out of a fear of depression or denial, as captured in statements:  The other reason is like something that you don?t know doesn?t kill so if I don?t know my status I don?t get sick?So people believe that if I don?t know my status then I will not suffer from depression, I will not suffer from stress, so the better I don?t know good for me?Ignorance is bliss. (T2)   But I think people do know where to get help and when to get it but they are always hoping it won?t happen. (T4)  (iii) Fear of Dying 	 ? 63	 ?While a fear of dying is not an internalized stigma, when internalized, the fear that HIV is a ?death sentence? may prevent a person from wanting to know their status and thus is an internalized stigma of an externalized perception based on seeing others die.  Many participants identified a ?fear of dying? due to HIV:  P1: I think that people in our communities see other people dying of this disease and we know there is not a cure for it. I think they are afraid?[they think that] ?so if I am going to reveal to someone I?m going to die, people think ?that that is [that]?  P1: life stops  P2: What?s going to happen to them at the end is death and death is a very scary thing for people.  (T2)   This statement demonstrates that there are two dimensions to a fear of death. First, is the a persons own fear of death which may be a self-stigmatizing attitude, as a preoccupation with the fear of death may lead to self-stigmatizing attitudes: And you know I see my brother, my sister, my aunts children all die and I know they died of this disease so I think people are very afraid. We are all afraid of death and I think that?s also contributing. That might be a reason.35 (T2)  There is this thing of if you have HIV then there is a death in front of you and sudden death and you say I?m positive you think I?m going to die. I don?t think there are people with high blood pressure, diabetes?who will before me. I just think I am going to die. (T3)     The second, ?fear of death? may be classified as a perceived stigma, where the stigma is of being perceived as a sick or dying person: ?In our community once you have been labeled HIV positive?it was regarded as a death sentence. Once somebody says you are HIV positive, people just see a sick person who is going to die very soon.? (T5)   Stigma in Professional HWs  	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?35 This comment was made by a participant in response to a question regarding why HWs may choose not to test for HIV.  	 ?	 ? 64	 ?After saturation was reached during the first two focus groups over stigma and confidentiality being the largest barriers to uptake of service, the interviewer asked for whom testing may be more of an issue (trying to discern sociodemographic factors that may prevent one group from testing over another). When participants were asked if groups (e.g. males vs. females, etc.) differ in their attitude towards HIV testing, professional HW?s36 were specifically identified as having a different experience of HIV testing than others. Professional HWs were seen to exhibit similar self-stigmatizing attitudes as other HWs ??even the professionals, they will deny that they have been HIV positive, even if you can see the signs and symptoms until now they are very ill (T4).?  However, professional HWs appear to differ in their experience of internalized stigma and the behaviors they exhibit as a consequence.  (i) Self-Diagnosis and Self-Medication  Participants felt that a professional HWs ability to self-diagnose and self-medicate distinguished them from other cadres of HWs:   ?for a doctor they can diagnose themselves, and maybe get treatment themselves.  But in other professionals you have to seek help.  But if you are nurse working in an HIV/AIDS clinic you can diagnose yourself but the challenge will come from getting the medication?(T4)   While this statement suggests that there is a challenge to getting medication, other participants disagreed demonstrating that access to medication makes it easier to self-medicate  We do self-medicate our self as nurses. We start in the medicine cupboard and if something is not there, I?m going to ask it from another ward, if its something I think will work for me, then you know we do ask around. (T4)  Participants also felt that nurses would rather self-diagnose: ?I think normally with the nursing profession?we diagnose ourselves if we see some signs, maybe I have HIV.  Should I go to test or should I not [sic] to test. (T4)? As a consequence of self-diagnosis this cadre of HWs may 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?36 Defined by participants as doctors and nurses specifically	 ?	 ? 65	 ?avoid seeking medical treatment at the expense of their health:  ?I will talk about nurses, because I am a nurse. We start off by taking self-medications.  We do not go to doctors, and we start getting advices from outside.  Then, by the time you go for treatment or testing it?s late.? (T4) One participant went so far as to say that some would rather others not know to the point of suicide, stating,  I have a feeling that sometimes even the accidents we are having is caused by these because if you once have it, you are going to crash into another car so you can die so that somebody does not know that you have died of AIDS, and they will just say it was an accident. (T4)    A possible reason why professional health workers might be more likely to self-diagnose and self-medicate may be their belief that they ?should know better?, which is an additional example of internalized stigma unique to this population:  I think if you are a health workers, there is this attitude that you are supposed to know better.  But then I got this, so you start judging yourself from that perspective. They think I should be knowing how to protect myself. I didn?t protect myself, then what are people going to say about you? (T5) (ii) Occupationally Acquired HIV  Professional HWs face additional challenges in the context of occupationally acquired HIV:  And even though when you get the needle prick in the ward and the patient is HIV and AIDS, you say I already have HIV because that patient could have been HIV and when you get sick you say I already have HIV without testing yourself. (T1)    The preceding statement demonstrates that this form of transmission is also associated with self-diagnosis.  This may be exacerbated by a culture of gossip that may either be a reason to self-diagnose, or may add a layer of symbolic stigma37 (i.e. the idea ?its your fault, you should have been more careful?) to already existent internal stigma:  P2: If you got pricked you have to get tested. 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?37 Moralistic shaming or blaming, see Chapter 1	 ?	 ? 66	 ?P3: They need to know?[interrupted] P4: The incident of needle pricks in the institution, here you would go to occupational health. The clerk opens up the file that is a protocol itself.  You need to go through all that. So they know the incident, ?oh, you had a needle prick?. And sometimes its known if that patient is HIV positive the one that you pricked your hand with, then that is the worst. You can see how many hands, how many people this person has to go through.  P2: And the report also in our meetings.  The report comes to the health and safety.  R: Does it have the name of the person? P3: No, no no, no, there is no name but somebody knows.  Various group members: We know, we know.  (T3)  Apparent Lack of Stigma in Laundry Workers  In opposition to professional HWs, laundry workers were identified as a potential cohort that may experience stigma at lower levels then other cadres of HWs.  While describing outreach efforts within health settings, one OH employee explained that during HIV testing outreach, they ??barely get 50% of employees, except at the laundries [where they receive more than 50%]? (T2).  This participant further noted that laundry workers appeared to be the group of HWs most comfortable with testing and disclosure: Our employees?at one laundry, whenever they have gone out to a clinic, when they come with their results the first thing before I even know about it they?ve made that known to their employer. Then when they come in the clinic and we talk and then they tell me that ?Now I?ve even disclosed this to the manager?. Without asking what made you feel so brave to disclose that to the employer, then the person will say ?but I?m going to need days off, I?m going to be very sick so my employer needs to know? then that one makes you to wonder those people are just managers, supervisors but this person has just decided to disclose that. So all those that have done that in that laundry, you?ll never believe when I say that this person is the one that has said it! They are helping, they are working, they don?t feel any pity for themselves. (T2)   Support by management, and the view that disclosure would not change their relationship with others may be reasons why this cohort may be described as ?normalized? to HIV within the workplace: Maybe it?s the attitude of the manager makes this people feel that now the person is mothering them so much that they can now say anything to her and it will remain there, even if they are known to be HIV positive?When I asked what if she discloses this to the other supervisors or employees, how will you feel? [she replies] ?No I won?t feel 	 ? 67	 ?anything, I want treatment, I?ve tested, I feel very proud.  We must also go and test.? So when I say, ?so you don?t mind if people know??[she replies] ?No, I don?t mind.? (T2)  3.3.6 Theme 6: Facilitators  Participants identified a number of facilitators that were perceived to be important to addressing the previously identified barriers.  Most of these facilitators fell into one of two categories: educational programming and changing attitudes. The first addressed educational programming on a wide range of topics, from perceived knowledge gaps about HIV to training of managers and professional HWs. The second facilitator addressed changing HWs attitudes to HIV in the workplace. While these are presented as distinct sub-categories, much overlap between the two exists, as both knowledge and attitudes must be addressed in order to change HWs behaviour. Subsequently, changing the behaviors of specific HW cadres (professional HWs, managers and support groups) was discussed.   Education    Numerous topics for educational programming were cited as potential facilitators to addressing previously identified barriers: ?I think we have to educate people on the importance of knowing their status, to deal with the disease, to accept the condition and to talk about it to the people, specifically they trust so that they can get support in return.? (T2)  Participants in one FG reached a unanimous consensus that education was the most important facilitator to improving the OHS service, explaining:  ?education because?before we can even give them a service of testing or there would be outreach, before there can be a support system in place, you need to know what?s going on. You need to be educated. You can?t expect people to support you if you yourself don?t know what it is you are going through or what lies ahead. So I think, with anything, education. (T2)   	 ? 68	 ?However, participants within this group were clear that ??education of all because one size doesn?t fit all? (T2), meaning that education had to be catered to the needs of specifics cadres of HWs within the hospital (e.g. married couples, professionals, etc.). Participants noted that for some, general education about HIV is important: ?Yes counseling is there, but you don?t understand what is HIV, that you might not die today, you might die after a long time. The understanding of this HIV thing is a problem.? (T3) However, they were also clear that education should move beyond how HIV is contracted: ??its not to say lets have an education program or awareness day where we are only going to talk about how you contracted [HIV]?? (T2)  Instead, participants strongly favoured a focus on practical implications of what knowing one?s status means in the context of the workplace. Their belief is that employees with more knowledge of HIV specific to the workplace may increase willingness to access services provided within this setting: ?the patient must be well informed because an informed patient, person is a willing person.? (T5)  (i) Policies and Programs on HIV in the Workplace  Participants emphasized that education on policies that address HIV and AIDS in the workplace is essential:  Firstly, I would like to see everyone knowing about the policy we are talking about health care worker. We must involve her.  Must have information about HIV, AIDS in the workplace.  We hear about it outside and what.  How that information relates to me in the workplace and knowing it what is the next step when my colleague, myself get infected, what are the steps to take, where to go, what does the policy says. (T5)  Specifically, participants noted the importance of including the Employment Equity Act and SA?s constitution within this training, as these documents establish the code of conduct and expectations of HWs with regard to HIV in the workplace:  I think we have to trust on the education and the Act, the Employment Equity Act, and the Constitution of the republic of South Africa, because the employment equity act has the 	 ? 69	 ?code of conduct for HIV and AIDS, and some of the people are not aware of it which is why we have to give knowledge to the people because recently most of the nurses are being disciplined by the counsel for negligence for things they have been taught but didn?t practice correctly.  Disciplined. Education knowledge is very important. Then we must give it. (T5)  Knowledge of workplace policies is also important, so that HWs can learn how such policies protect their health and support them in coping with their illness: If people say, I am a health care worker, what does my employer says, what does my policy says about me being HIV positive and experiencing all the signs and symptoms?  I am a health care worker working at the admission. I have to face all other clients being there, my condition deteriorating. What, how are the policy protecting me? I have cough, I am now becoming thin and everything. And I am there in the front, I am the front mirror for every person coming in.  How am I coping with such things as a healthcare worker? We need reality now to face me.  [So that it] is?easy to tell my fellow workers look here, I am HIV positive? (T5)  This knowledge was also seen as a way to highlight existing support pathways for HWs, which in turn may have improved uptake of OHS services. Specifically, the Employee Assistance Program (EAP) may provide the dual advantage of protecting HWs and marketing the value of HIV testing at the OHS as a value-added service to address their needs: ?If I have got information, policy saying that immediately when I?ve got this, I can go to EAP, EAP can refer me to occupational, occupational can say, okay you can be removed as a change of job, maybe move me to somewhere??(T5)  (ii) Improving Knowledge and Perceptions of the OHS Along with highlighting national policies and programs, participants noted the importance of education specifically on the services provided by the OH unit, including and beyond HCT: ?We still need to start from the scratch, educate people that occupational health is there from 1, 2, 3, 4, 5. And if you give, for HIV and AIDS [testing], after testing them, this is what [else] you can get.? (T2)  	 ? 70	 ?In this vein, participants in one FG were united that the OHS is limited if it does not provide services beyond HCT, and that it should become a holistic service: One stop for everything.  It?s the issue that they are dealing with it now. Presently, that comes from the head of AIDS, that now you must stop, ?that must be HIV clinic, that must be medical clinic.?  If a person, like HIV positive patient has got medical conditions, why can?t the medical clinic offer everything there?  So when I attend your medical clinic, I know I am going to get my treatment also from you. Not going to her, as an individual clinic offering only HIV. (T5)   An additional benefit to holistic service provision is that concerns regarding privacy of the OHS may be mitigated, as all services would be provided by all OHS employees and HWs would not be identified as accessing HIV services solely based on the person they were seeing: Now I?m thinking, if they are doing HIV and TB, occupational health and safety, its supposed to be comprehensive, you do everything? [OHS employee responds: ?everything?], so [employee 1?s name] does everything, [employee 2?s name] does everything? (T5)  Finally, building trust in the service requires clear communication about how the OHS maintains confidentiality: ??it is for us to give them the information that there is confidentiality in the occupational health and safety unit?? (T1). The power of such education can be seen in the following exchange between two participants: P1: I just want to make a follow up with number two [everyone laughs] Yeah, right now at [hospital name] when it comes to blood test it is highly confidential.  On the form we write ?confidential? with no name, only the id number?and then we send to virology.  And then me, myself, I am the one who has to go there and actually get the results myself and bring it back to the file of the client.  P2: The other thing is I think not all the people are aware of that, because it is for the first time now I hear from number 7 that as a personnel you can go to occupational health and your things will be treated confidentially.  We never knew that.  It?s for the first time I hear that there is a system in place for something like that.  We were always under the impression that your staff will be on the computer and everybody [laughter from FG] will be able to see that.  Maybe that is why the officials are not using the occupational health clinic to go and test there. (T4)  	 ? 71	 ?The above statement also highlights the importance of providing knowledge about health records within OHS to all HWs: ??in occupational health and safety, the results are not captured in the Meditec?? (T5) but ??the other people are not informed of where the files are kept?so we should give that information to people that are not a part of the group38? (T1)?.  (iii) In-Service Training    In-service training was the most commonly cited modality to deliver HIV education to HWs on a range of topics previously described: ?All in all we need in service training for HIV and AIDS, confidentiality, and the awareness and ignorance, everything, I think if we can get the best in service training then maybe something can be really be done or come out of it?. (T1)  Participants felt that the benefit of this form of training would be most beneficial if it was provided frequently: ?Education, but something like in-service training, keep on talking once or maybe twice a month. Not once yearly, like we do it. (T3)? and what is needed is ? ?continuous in service training.? (T5)  However, in service training may be difficult for some co-workers to execute due to personal relationships between trainers and trainees: I don?t know that my friend has tested positive or not to HIV. I speak freely of the subject even part of educating in there, because we nurses, you know, like to educate all the time. When we speak we are on a teaching manner, not like we will be discriminating or we will be pin pointing you know. Yeah, but immediately if you the nurse learned your friend has tested positive you are no more like allowed to speak about the subject anymore because now we, you are no more sure how should she will take this now. (T2)  Furthermore, a participant suggested that for those HWs for whom in-service training is intimidating, asynchronous modalities for delivering information may be more suitable: ?but you know for our lower educated people maybe its easy to educate and educate but people are sitting there and they are not learning something you know, its always the nurse who knows better or the social worker knows more. You know, yeah, like practical things like posters, pamplets that you can put up in your workplace, that you can hand out for people, you know where they can take it home, sit and read it. Practical things, not just educate ? yeah you know we are getting the people and they are here in the tents 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?38 The group here refers to the group of employees that are a part of the OHS service.	 ?	 ? 72	 ?and they are going to get tested but it must be a continuous thing every day neh.  I walk past a poster and I remember you now, not just now and then?(T2)  Attitudes   Changing the ?mindset? or ?attitudes? of individuals was considered by some to be the most important aspect of encouraging HIV testing. Participants viewed changing HWs attitudes toward HIV as a key factor in overcoming HIV stigma, but acknowledged that this was difficult to achieve: ?So I think if people could be, if we could target the attitude of people, maybe information in terms of attitude, if people could be free enough and accept the whole thing and find a way to remove the stigma, that?s the main challenge.  But how do we do that?? (T5) To address this, one participant suggests providing education on the psychological impacts of HIV and AIDS, to humanize the disease experience: I think you must also learn more about the psychological impact of HIV on individual so you know what the other person or colleague is going through and then be there to support her or him so that even that we can do more on the psychological impact of HIV on our colleagues. (T1) 	 ?(i) Treating Everyone ?the same?  As much of the discussion on stigma related to concerns about being treated differently, comments on ?treating everyone the same? formed the core recommendations on how to address HIV stigma in the workplace. There was a sentiment that the culture of workplace gossip had to change: ?That?s what?s killing our people?Like number 1 said, we gossip.  If we can [change] the mindset of the whole thing, it will really help.? (T4) One potential cultural shift that was suggested to directly address gossip was the rhetoric of ?treat everyone as having HIV and AIDS?: ?I was going to say as HWs, if we can remember that we are all positive until tested otherwise, maybe to also help the whispering.? (T3) The belief is that equal treatment of all people, regardless of their disease status, would normalize HIV within the workplace. This 	 ? 73	 ?sentiment is also perceived to be legitimized by the views of political leaders: ??our Minister of Health said ?treat everyone as having HIV and AIDS? (T4)   Targeting Specific HWs (i) Professionalism and Confidentiality among Professional HWs  Although there are clear benefits to using the OHS, it was noted that these benefits only exist if HWs trust the service, which is dependent on the level of professionalism demonstrated by OHS employees:  There is no better place then being at the occupational health. Number 1: it is monitored by the government rules, number 2 there is no better place and time to go for occupational health because you can go at your own time. Or what we need to do is make sure everybody, nurses and everybody behave professionally and we, what call it, when we go on whatever we do we display that professionalism then the trust will automatically come. (T2)  The notion of professionalism was intimately tied to the concept of trust. As previously demonstrated, trust and confidentiality are linked, which is supported by participants suggestion that specific and continuous training on confidentiality as an essential aspect of educational programming for professional HWs: I think over and above as healthcare workers, we have got to be educated and trained again regarding confidentiality.  I remember when I started nursing the very, very first lecture that we got was regarding confidentiality. So I don?t know now of late, what is normally done but I think that is no longer done. I think that if we can be reminded again that we have to keep all the information that is entrusted to us, we should keep it secret. It?s not for us to discuss during our tea times and our lunches, and for that to happen it must have some punitive actions. If it is just about saying this and this, and then nothing is done, then they are going to continue. (T1)  The above statement also highlights the need for clear consequences to gossip and breaches in confidentiality by professional HWs. However, participants highlighted that the presence of such 	 ? 74	 ?consequences should be coupled with education about professional duties of the HW and the nature of and reasons for disciplinary procedures in these areas: It is important they learn or remind them about the duties of a nurse, because now a days many nurse they go to disciplinary committees because of those39 mistakes.  It is very important that we remind each other about those, because when we have done wrong, they don?t say have you forgotten or what, they say ?you are negligent? [agreement from the group]. It is very important that you give information to people when they are here, so you have to give information to others when they are not here, so we can be more careful in the future. (T1)  The preceding statement also demonstrates that such educational programming can have the added benefit of making professional HWs role models in the workplace. Through this role modeling, professional HWs can demonstrate the value of maintaining confidentiality and behaving in a professional manner, in turn changing the attitudes of those around them:  One of the basics in nursing is confidentiality.  If you could stick to confidentiality and we as professionals could be confidential to our patients and health workers status, then it will go to the next person - the lay person, who is not informed about HIV and AIDS. And they would see the difference between a professional and a lay person and it will change the mindset of the lay person. Most of the lay persons who are in our professional settings, they do things because they see them being done by professionals?If professionals could act professionally, and go back to the basics of confidentiality - that is a main thing. (T4)  (ii) Managers? Roles in Promoting Testing and Disclosure  Similar to professional HWs, by virtue of their position in the workplace, managers were perceived to influence culture related to HCT, highlighting the importance of these HWs behaving as role models: ?You know now, people don?t want to be the people leading the cue?its rare to see a CO [corporate officer] there40, they can just come there and open and run away.  They always have meetings, so HIV is for us, the professionals and the people down there, not for higher people.? (T3) and ?If I see the CO is very comfortable with OHS department, then I too will go and test.? Managers within FGs were cognizant of this responsibility, stating: ?One 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?39 In reference to breaches in confidentiality	 ?40 Testing for HIV in outreach	 ?	 ? 75	 ?can not change a persons attitude, they can only influence it. As managers the way we carry ourselves?influence other staff.? (T5) Thus, managers were identified as a second group that requires specific training on a variety of topics, including confidentiality and workplace policies on HIV. With respect to the HIV status of HWs, the issue of confidentiality becomes particularly important, given that managers in supervisory roles have access to employee health records:  ?it does not just end your diagnosis being secret with your doctor, but it can also affect your workplace because now everything now needs to be captured because the risk assessors they will then say we want a detailed report of your doctor, meaning now your doctor will also have to disclose?we are teaching continuously to supervisors and saying that now whatever diagnosis you are seeing in the medical certificate please don?t tell the other supervisor or the other employee about this person. (T2)  Along with maintaining confidentiality, participants also expressed that managers required additional training in creating a supporting work environment that is conducive to disclosure and discussion of HIV among HWs: Currently, like for example, the government has put this thing called, like, ?work conducive environment for HIV in the workplace? so people can disclose. And if you are going [to disclose], a conducive environment would mean [if] the people are around any other person including management, are we having a management that is sensitive to people who are positive?  So I think we also go for tailor making our education. That we can also train management on what is the role of management, what?s creating a conducive environment in the workplace, so that allow people to disclose. Because if management are not playing any role then we are going nowhere. (T2)  Furthermore, it was perceived that managers may lack knowledge of HIV and AIDS employment practices, as evidenced in this statement ?The management also are not well versed with the issues of HIV and AIDS especially related to employment practices.? (T2) One strategy identified to educate these HWs was mentorship by more experienced colleagues:  I think the main thing there is a problem with manager not being informed.  I think if the hospital would implement something like mentorship because it is a fact that there are other new managers and older managers that does not have the information but those in higher level if they can just mentor the one in the lower levels, operational managers, just 	 ? 76	 ?assist them like have a quarterly meeting on how to help them with their managerial skills and all that, I think it will also help. (T2)  (iii) Support Groups  Finally, participants also suggested forming HIV support groups for HWs within the hospital, as is common practice in community settings:  I also think it?s necessary to have support groups in the [institution], in the employees. Usually they have the support group in the community. Maybe for HIV?its necessary to have one for employees. They can formulate one. (T1)   These support groups were perceived to be beneficial, both in increasing support for HWs related to HIV, and in shifting attitudes within the hospital: ?I think with a support group, when they see you going to that support group, shifting of mind looking at this thing differently.? (T4) Within hospitals, it was suggested that support groups may go beyond the needs of individual HWs, to serve the additional function of educating others on issues related to HIV and AIDS: ?And I think the support group must go to the wards?so they can teach the staff about HIV and AIDS. Not the HIV patients only. All of us, we must go there.? However, it was suggested to ensure that these groups are well attended, and to serve this additional educational role, they should include all HWs regardless of their HIV status:  But I think if we start support groups, it doesn?t have to be for people with HIV and AIDS, neh, the moment we start a support group for HIV and AIDS, everyone will be out. But if I am negative and I go there for information because I have somebody at home with that, then if it?s a support group with different things?(T4)   As one HW attests, the value in including all HWs in these groups is: ?Truly there is none that is [n]either affected or infected.? (T1)  3.4 Summary of Results   The results of the FGIs have been summarized in tabular format in Table 10.     	 ? 77	 ? Table 10: Summary of FGI results    Theme Sub-categories Summary of theme Most common locations for testing ? None ? HWs prefer to get tested for HIV with their private doctor Privacy ? HIV testing within workplace ? Accessing OHS for testing   ? ?Privacy? cited as primary factor for preferring private doctor over OHS for HIV testing ? Less privacy at OHS as co-workers know when a HW accesses services and/or are involved in HIV testing and subsequently learn the HW?s status ? Provision of HCT by a single OHS employee contributes to lack of privacy, as obvious that HWs visiting this individual are accessing HCT Gossip ? Workplace gossip about HIV status  ? Gossip by professional HWs  ? Fear of gossip cited as a reason not to test  ? Gossip pervasive; assumptions made about HWs HIV status based on changes in weight, occupational exposure, etc.  ? Gossip by professional HWs occurs within and beyond the workplace, represent negative role modeling, and associated with breaches in confidentiality that undermine trust in the system Confidentiality ? Trust ? Electronic breaches in confidentiality ? Lack of confidentiality equated with a lack of trust in the OHS and involuntary disclosure ? The number of employees impact perceptions of confidentiality of the OHS ? HWs perceive electronic employee records are not secure based on known breaches in confidentiality, felt to be accessible by non OHS co-workers in the hospital, especially professional HWs ? These factors influence if/where HWs go to test Stigma ? External Stigmas o Enacted Stigma o Instrumental Stigma o Deviance and Symbolic Stigma ? Internal Stigmas o Perceived/ Felt stigma o Self stigma/ Denial ? Stigma in Professional HWs o Self-diagnosis and self-medication o Occupationally acquired HIV ? Apparent lack of stigma in laundry workers ? Stigmatization perceived to be the main barrier to uptake of HCT; may have consequences for mental health of HWs and desire to test or disclose ? multiple types detected including: o enacted (e.g. overt discrimination by employer);  o instrumental (fear of infection through objects); o Symbolic (HWs are viewed as ?promiscuous? if accessing HIV services of the reason) o perceived/felt (e.g. fear of being gossiped about, treated differently, or as a sick/dying person); o self-stigma (e.g. denial and fear of ?being sick?) ? Professional HWs experience additional internalized and symbolic stigma (e.g. ?I should know better? or ?its your fault?); leading to specific behaviours (e.g. self-diagnosis and self-treatment)  ? Non-professional HWs (e.g. laundry workers) may represent a cadre in which HIV is ?normalized? 	 ? 78	 ?Theme Sub-categories Summary of theme Facilitators ? Attitudes o Treating everyone the same ? Education o Policies and programs on HIV in the workplace o Improving knowledge and perceptions of the OHS o Modality of training ? Targeting specific HWs o Professionalism and confidentiality among professional HWs o Managers? roles in promoting testing and disclosure o Support Groups ? Educate HWs on psychological impact of HIV/AIDS ? Address gossip through attitude ?treat everyone the same? ? Tailor education about HIV/AIDS to different HW cadres; provide training on national and workplace policies along with general education about HIV  ? Highlight existing benefits/pathways of support within existing programs for HWs (e.g. EAP, OHS) to increase uptake of services ? Improve knowledge of OHS as a holistic service and perception of confidentiality in the unit ? Provide a variety of modalities to educate HWs, including in-service training and asynchronous information delivery strategies (e.g. pamphlets) ? Provide training for professional HWs on professionalism and maintaining confidentiality and consequences of breaching confidentiaity ? Train managers to create conducive work environments  ? Encourage professional HWs and managers to be role models for other HWs in the workplace, provide support groups for HIV positive and other HWs to discuss HIV/AIDS   	 ? 79	 ? Chapter 4: Discussion    It has been the aim of this thesis to identify barriers and facilitators to HCT uptake within hospital-based OH units. Both the quantitative and qualitative findings of this study confirm that HWs face numerous challenges in accessing HCT services in the workplace, including a lack of desire to test, concerns regarding privacy and confidentiality and HIV stigma in the workplace.  This chapter will discuss these barriers and how they may operate to decrease uptake of HCT, in order to provide recommendations to strengthen existing services. This will be achieved by integrating the results from Chapters 2 and 3, to allow the findings to be combined and corroborated, in order to generate a deeper understanding of how these barriers impact the uptake of HCT. 4.1 Accessing HCT Services  The ?desire to test? is a prerequisite for accessing HCT. However, the results of the current study demonstrate that more than 60% of HWs sampled believe their co-workers would not want to know their HIV status. This question may be capturing two dimensions of the ?desire to test?.  First, a perception that others may not want to know their status may be a reflection of the culture of HIV within the workplace itself. For instance, if co-workers are not commonly observed to access HCT, or if colleagues in the workplace do not openly discuss knowledge of co-workers? HIV status, then it may be inferred that HWs do not want to know their status. Second, the question may be a self-reflection of the respondent?s own belief that they do not want to know their status and thus a projected belief that others share this opinion.  In either instance, it would appear that what is ?normal? in this population is to not want to know one?s status. As HIV testing is the means by which to find this out, these results suggest that HIV 	 ? 80	 ?testing is not yet ?normalized? in the study sites. Goffman (1963) articulates that normality only exists in reference to the abnormal. Facilitating uptake of HIV testing requires that norms shift, such that an individual wanting to know their HIV status becomes ?normalized?. Thus, understanding why a person would not want to know their status becomes critical to this discussion of barriers and facilitators to uptake of HCT services in the OHS. 4.2 Preference for Private Physician           While a majority of participants believe HWs do not want to know their HIV status, it appears that of those who do prefer to test, many prefer to test with their private physician. The vast majority of FG participants demonstrated a preference for private physician over the OHS, if they were going to get tested at all. Similar findings have been demonstrated in corporate occupational health programs, where employees with Medicaid expressed preferences to undertake HIV testing at ?off-site? clinics that were less ?visible? to colleagues and friends (Arimoto et al., 2013). This is also supported by the survey results, which indicate almost 50% of participants felt HWs are most likely disclose their HIV status to their private doctor, while by comparison only 18% felt co-workers would disclose to their OHS nurse. FG participants identified a number of reasons for preferring to test with private physicians. First, there was a reluctance to test in the workplace due to the fact that it was the workplace itself. Second, as OHS nurses are also considered by many to be colleagues, there was a reluctance to be ?tested by co-workers?.  This is supported by survey results that indicted the majority of HWs would be unlikely to disclose their HIV status to a co-worker. Thus, the dual nature of the OHS staff being both caregiver and co-worker may make using the OHS for testing uncomfortable for some and may present a preference for privacy. This perspective is in keeping with existing literature on barriers to uptake of occupationally based HCT services in the 	 ? 81	 ?corporate setting. A recent study conducted at a large multinational company in RSA found that employees with an extensive social network in the workplace who were reluctant to share their HIV status with co-workers had significantly lower probability of accessing HCT at on-site OH clinics (Arimoto et al., 2013). This was attributed to the perception that there is a higher probability of co-workers finding out the individual had taken an HIV test (Arimoto et al., 2013). These concerns over privacy and confidentiality of the OHS (which are not perceived to exist with private doctors) are also demonstrated in the current study and constitute the primary reason why testing with a private physician is preferred over HCT at the OHS among participants.  However, preference alone is not grounds on which to dismiss the use of HCT within OHS services. The OHS offers additional benefits, such as convenient, accessible and free HIV services. In addition, improving access to HIV services for HWs may support TB care, which is necessary for HWs, some of whom are at an increased risk of acquiring this infection through occupational exposure. As a consequence of a HW disclosing their HIV status, they may be able to change their work assignments to reduce their risk of acquiring such infectious diseases (and in turn, decrease the risk of transmission of TB to co-workers and patients). Despite these potential benefits, concerns regarding a lack of privacy and confidentiality may be sufficient for HWs to access to care through private doctors, or even worse, to not access care at all. In turn, this may have a potentially detrimental effect on the overall health of the HW. 4.3 Privacy and Confidentiality  While the literature describing OHS for HIV testing suggests these services are akin to private physician care (Uebel et al. 2007; WHO-ILO 2010), participants in this study did not perceive the OHS in this manner. Instead, the HCT service is viewed as comparable to a HIV clinic. For the individual, waiting to be seen at the OHS is akin to ?queuing? at an HIV clinic, 	 ? 82	 ?both of which are perceived to lack privacy. HIV clinics are identifiable by the fact that they provide only one service (HIV testing and treatment). Similarly, when OHS staff provided HCT in a manner by which the OH employee is identified as providing only HIV services for a given time period, then the HCT service within OH unit becomes identifiable. As an FG participant explained, an individual waiting for the OH nurse may fear that others will see them and know that they are accessing this service. Existing literature suggests that such privacy concerns represent a barrier to uptake of HCT in workplace settings, independent of concerns regarding the confidentiality of the service (Arimoto et al., 2013). It has been argued that the normalization of HIV testing in the workplace may lessen the perceived need for anonymity in testing (De Cock, Mbori-Ngacha & Marum, 2002) and in turn reduce concerns regarding privacy. However, as previously noted, in the sites sampled in this study, such normalization has not yet occurred. Thus, in the present climate accessing the OHS for HIV services may be perceived by a HWs to be an ?involuntarily disclosure? of their HIV status. Such involuntary disclosure has been linked to the lack of private spaces that characterize health facilities (Obermyer et. al, 2011), which allow HWs to gain knowledge about the services accessed by their colleagues and may subsequently lead to gossip and assumptions about their HIV status. This ?involuntary disclosure? is an issue for HWs, as it is clear from the FGI and survey results that very few HWs would want to disclose their status to their supervisor, closes co-workers, or friends. When coupled with a lack of privacy, and the fact that the service is situated within a workplace where disclosure of status is not yet normalized, the OHS may potentially be even less desirable than a clinic in the community as a location for testing, especially among individuals who have extensive social networks in the workplace (Arimoto et al., 2013). 	 ? 83	 ?The fear of involuntary disclosure becomes heightened when the OHS is perceived to lack confidentiality, as the potential for overt disclosure of a HWs HIV status is introduced. The study results indicate that fears about confidentiality are a major concern for HWs, with nearly one-third of participants believing that confidentiality is not maintained in the OHS. In specific reference to HCT services at the OHS, a fear that confidentiality is not maintained was the most commonly cited factor for not using the service, with 38.9% of participants indicating this response.41  This supports existing literature that lack of confidentiality is one of the largest barriers to uptake of HCT in hospital settings (Uebel et al. 2007; Namakhoma et al., 2010), and within the general public (Corbett et al., 2006; Dieleman et al., 2007). In this study, breaches in confidentiality were identified through involuntary disclosure of HWs status by professional HWs, and through a perception that unauthorized access of HWs electronic records by co-workers occurred outside the OHS.  These findings support the body of literature that examines breaches in confidentiality within patient care, and demonstrate that while breaches in confidentiality pertaining to HIV status have also been observed in patient-physician relationships in the community (Surlis & Hyde 2001; Lindau et al. 2006), within the hospital it becomes even easier to identify when confidentiality is not maintained. Within the OHS itself, participants endorsed the perception that having more than one OHS employee manage patient care may increase the likelihood that confidentiality will not be maintained. This finding supports the results of the study by Uebel et al. (2004), which indicates that the presence of one trusted OHS staff appeared to overcome issues of confidentiality in the service. The current study demonstrates that while this may be the case in single-provider OH 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?41 It may be worth noting that there was no question in the survey that addressed the concept of ?privacy?.  However, as confidentiality and privacy are related, it may be possible those participants viewed the two interchangeably and that this statistic is capturing both.  However, as both concepts involve a fear of involuntary disclosure, which concept is the ?main problem? is inconsequential.  	 ? 84	 ?units, when there are multiple employees, perceptions of confidentiality and subsequently HWs trust in the service may be undermined. FGIs identified that this may occur for a number of reasons. First, having multiple employees at the OHS inevitably results in turn over of staff that provide HCT at different points in time, such that HWs must discuss their HIV status and receive counseling from multiple caregivers. Thus, in contrast to single-practitioner services (e.g. a private doctor) where a one-on-one relationship is sustained over the course of the individual?s illness, HWs must establish a trusting relationship with multiple caregivers at the OHS. Second, HWs may associate the fact that there are multiple employees within the OHS with increased opportunities for gossip. Paradoxically, while the provision of HCT by a single OHS employee is associated with improved perceptions of confidentiality, as noted earlier it is also associated with less privacy. Another dimension identified in the self-administered survey is that HWs perceptions of confidentiality in the OHS may vary by race. Specifically, black and coloured HWs surveyed have 2.7 and 3.0 times the odds, respectively, of perceiving confidentiality is not maintained in the OHS, as compared to white HWs in the sample. It is not entirely clear why this difference was observed and there is limited literature specific to the HW population to explain this effect. One explanation may be that black and coloured HWs in the sample more freely admitted their concerns regarding confidentiality as compared to their white colleagues. An alternate explanation, based on the FG responses, could be that differences in the exposure to breaches in confidentiality differ among HWs of different racial groups. As one OHS employee noted, breaches in confidentiality have been observed within taxis, suggesting that for some HWs, the culture of gossip extend beyond the workplace. As HWs are likely to travel home with co-workers living in their communities, exposure to these breaches in confidentiality may differ 	 ? 85	 ?between HWs living in different communities. However, this is only a theoretical explanation, and further investigation is needed to better understand the observed findings.  Electronic health records also appear to be a source of concern for FG participants with respect to confidentiality of the OHS. This stems from their misperception that electronic health records are easily accessible to any employee with ?basic computer skills?. Coupled with the allegation that curiosity about a co-workers HIV status may entice a HW to confirm their suspicions, the perceived ease of accessibility of health records may make the OHS appear unsafe for testing. While this perception may be valid for patient records, there was ambiguity about how employee health records are housed within the OHS. Regardless, the literature supports the need for securing health records in order to ensure confidentiality is maintained. (Magnusson, 1994)  4.4 Stigma  Both the survey and FGI findings demonstrate that HIV stigma is a pervasive issue. This is in keeping with numerous studies that argue stigma is the largest barrier to normalizing HIV and HCT uptake (Corbett et al., 2006; Dieleman et al., 2007; Uebel et al., 2007; Namakhoma et al., 2010). Nearly forty percent of HWs surveyed in this study believed that there is HIV stigma in the workplace. However, while this question detects that stigma exists in the workplace42, it does not quantify the levels or types of stigma within this population. This limitation is partially addressed by the FGIs.  In keeping with the Siyam?kela report (2003), FG findings are broadly grouped within two sub-categories of ?stigma?, each with multiple dimensions: internal (e.g. 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?42 It should be noted that stigma, as a concept, is difficult to define, even in the English language (see Chapter 1). Furthermore, there is no direct translation of the word ?stigma? in any South African language. In the SeSotho survey, the translation of ?stigma? in this question was directly translated to ?ridicule?.  Thus, while the survey detected that stigma was present, it was not able to distinguish the type or amount of stigma present. I acknowledge that stigmatizing beliefs are conceptually heterogeneous (Goffman, 1963) and that this was not captured in the measures used in the current study, which treated stigma as a unidimensional construct.   	 ? 86	 ?perception of self, social withdrawal and fear of disclosure) and external (fear of dying, avoiding, rejecting, labeling and gossip). These dimensions operate within the types of stigma identified, namely: enacted stigma, instrumental stigma, symbolic stigma, perceived or felt stigma and self-stigma (Herek, 2002). While FGI provided some examples of stigma that may be active within this community, it is by no means an exhaustive list, and other stigmas may exist within this population.  However, the presence of these particular stigmas support the Mbonu et al. (2009) study that identified that HIV stigma in the African context is largely embedded in beliefs about contamination and morality.   Only one typology of stigma can be corroborated between survey and FGI findings. Although enacted stigma was detected in FGs through comments regarding HWs fears of not being promoted or given housing and vehicle subsidies, this overt discrimination by employers was not felt to be an issue among all FG participants. The quantitative findings support the latter perspective by demonstrating the majority of HWs surveyed believe that employers would be supportive of an HIV positive employee and encourage them to get treatment, with only a small proportion expressing concern that an HIV positive HW would not be promoted in a timely manner or would lose their job after an employer learns of their status. A further analysis of the typologies of stigma captured in this study is beyond the scope of this discussion, as there may be little value to this analysis in the absence of a measure to quantify the amount of each type of stigma expressed in this population. Instead, this discussion will center on the way in which HIV stigmatization occurs in the study sites.  It is important here to remember that stigma is a highly contextualized process based on the norms and values of the community within which this process occurs. 	 ? 87	 ?Within this study, the ?fear of stigma? itself was pervasive, captured by a fear of being ?treated differently?. Stigma is expressed in this population through stigmatizing attitudes and behaviors (Maman et al., 2009). FG participants cite examples of stigmatizing attitudes directed towards HWs losing weight or those sustaining needle stick injuries. Stigmatizing behaviors, especially gossip, were present in all three-hospital sites and reinforced stigmatizing attitudes. Additionally, two factors that contribute to stigmatizing attitudes and behaviors were identified: the fear of transmission from causal contact and a fear of death. Both of these fears have been previously identified as factors contributing to HIV stigma and discrimination (Maman et al. 2009). These contributing factors make stigma permissible as a form of self-protection by HWs who perceive a co-worker with (or assumed to have) HIV as a threat (Maughan-Brown, 2006; Herek & Capitanio, 1999).  As stigma is a context-specific concept and occurs ?within specific context of culture and power? (Parker & Aggleton, 2003), outlining how HIV stigma may operate within the hospitals studied to prevent HWs from accessing the HCT service should also be explored. As HIV stigmatization may be viewed as a dynamic social process, Link and Phelan?s conceptualization of stigma (2001; Link et al. 2004) will be used to explore the mechanisms by which stigma operates through the hospital system in order to understand the underlying process of stigmatization in this population. As articulated by the authors, this conceptualization is meant to be useful, not definitive or comprehensive (Link & Phelan, 2013: 579). It seeks to explore how and why stigma has a pervasive and persistent effect on people?s lives and how it is connected to the health of a population (Link & Phelan, 2001). It is particularly suited for the data generated in this study, as unlike other conceptualizations of stigma (Scambler, 2009; Aggleton & Parker, 2002), it recognizes and accounts for the overlap in meaning among concepts like stigma, 	 ? 88	 ?labeling, and discrimination, which became difficult to tease apart in the FG data of the current study.  Link and Phelan propose that stigma contains the following six elements (Link et al., 2004; Link & Phelan, 2013: 578-9), all of which are necessary for this phenomenon to exist: (1) Labeling: differences among individuals are distinguished and labeled (e.g. HWs identified as HIV positive or negative); (2) Stereotyping: these differences are associated with negative attributes or stereotypes, either in the minds of others or the labeled person him or herself (e.g. HIV positive individuals are ?sick?, ?dying?, ?contagious through causal contact?, ?promiscuous? or ?careless?); (3) Separating: a separation of ?us? (HIV negative HWs) vs. ?them? (HIV positive HWs) occurs, (e.g. HWs assumed to be HIV positive are seen to be ?different?); (4) Emotional reactions: stigmatizers experience a range of emotions (e.g. fear of contamination) that can be detected by, and subsequently shape behaviour toward, the stigmatized, who also experience various emotions (e.g. fear of becoming sick or dying, shame over ?not knowing better?); (5) Status loss and discrimination: the stigmatized is devalued, rejected and excluded (e.g. through gossip or social isolation); and (6) The dependence of stigma on power: stigma can only be perpetuated when a potential stigmatizing group has the social, cultural, economic and political influence to ensure that their views are adopted by others.  Furthermore, stigma in the context of this study may be seen to operate at three levels (Link & Phelan, 2001): 1) ?structural discrimination? is the mechanism by which social contexts enforce stigma without person-to-person interaction, such as being seen by an OHS employee directly linked with HIV services only; 2) ?direct discrimination? is the mechanism that operates at the person-to-person level, including enacted or experienced stigma related to gossip or social isolation resulting from instrumental stigma; and 3) ?social psychological processes? that operate 	 ? 89	 ?through individual stigmatized person, such as when a HW labels themselves as someone who ?should have known better? and subsequently fears being viewed differently. In the present context, the process of stigmatization among HWs in occupational settings (Figure 1) may begin when a lack of privacy, breach in confidentiality or observed exposure to HIV (e.g. through a needle stick injury) leads to accurate or inaccurate labeling of a HW as someone with HIV. This results from an assumption or ?involuntary disclosure? of the HWs status. The negative attributes associated with HIV that are subsequently applied to the HW stem from emotional responses such as fear of contamination, moral judgments about the HWs sexual behaviour, etc. In turn, these attributes are perpetuated through discriminatory behaviours such as gossip. As a result, the HW may be devalued and isolated. In this way, the HW may face the instrumental, enacted, symbolic and internalized stigmas captured in this study.   Figure 1: Proposed process of stigmatization among HWs and relationship to other barriers 	 ? 90	 ?4.5 The Role of Professional HWs  As a process, stigma can only occur if an individual or groups of individuals have the power to perpetuate it  (Link & Phelan, 2001; 2013; Link et al. 2004). Hospitals are hierarchical communities, where power dynamics play a role in creating and perpetuating stigma (Scambler, 2000).  The results of this study suggest that a common link among the barriers to uptake of HCT identified (including confidentiality and stigma) is the professional HW (Figure 2). Thus, further analysis of the relationship between the professional HW and the workplace may provide valuable insights into these barriers and how they may be overcome.   Figure 2: Relationship between barriers to uptake of HCT and professional HWs 	 ? 91	 ?In the context of this discussion, is important to define exactly what constitutes a professional HW43. Fox (1989) characterizes a HW as a professional by having an ?organized body of knowledge, technical skills? and ?mode of education? acquired through education and training that socializes them into the profession. In addition, professionals have ?autonomy at work and collective self-control?.  Fox (1989) contends that this ?autonomy? allows the professional to regulate their own individual and corporate behaviors, and differentiates professionals from other occupations. Thus, unlike other occupations within the health workforce, professionals may have ?more to lose? from being labeled as HIV positive in the workplace (with all the resulting perceptions they know accompany such labels). Studies suggest professional HWs fear that being labeled as HIV positive will mean that they have failed as role models. As a result, HIV positive professional HWs struggle with testing, disclosure, and promotion of HCT (Turan et al., 2008; Namakhoma et al., 2010).   The self-administered survey demonstrates significant associations between HWs occupation and perceptions of confidentiality. Specifically, when controlling for age and race, PCHWs had 2.3 times the odds of perceiving that confidentiality is not maintained in the OHS, as compared to non-PCHW. This finding may be explained through the FG results, which demonstrate that by virtue of their professional role and knowledge of how HWs health records are managed, this cadre of HWs are most likely to have knowledge of both actual and potential breaches in confidentiality (Namakhoma et al. 2010). In addition, professional HWs may be 	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?43In the analysis of the survey, the term ?patient care health worker? was used to differentiate doctors, nurses, and allied health professionals (AHP) from other HW cadres, while the term ?professional HW? was used by participants during the FGIs. Although these terms are not the same, as it may be argued that the same cadres of HWs fall into both categories, comparisons between data generated on both have been made in this discussion. As a caveat, while Fox (1989) would contend that AHPs may have ?less autonomy? than a doctor or nurse (and thus may not always qualify as a ?professional?), I felt it was more appropriate to categorized them as professional HWs rather than with other HWs cadres (e.g. administrative staff, household aids, laundry workers, etc.) and am supported by (HCJ van Rensburg, 2012: 410). As AHPs adhere to similar ?values? as nurses and doctors, I feel that the inclusion of this cohort within the category of ?professional HW? is justified.   	 ?	 ? 92	 ?implicated in these breaches themselves (Magnusson, 1994). This is demonstrated in FG participants? comments that confidentiality is undermined by the culture of workplace gossip, which was seen (at least in part) to be perpetuated by professional HWs, despite evidence that such gossip undermines trust among HWs and in the system as a whole (Friedson, 2001; Magnusson, 1994). In addition to the direct impact of breaches in confidentiality, it was also observed by a FG participant that when professional HWs engage in gossip, it legitimizes this behaviour for other HWs, who model the behaviours displayed by their professional colleagues. Implicit in this belief is the assertion that professional HWs have the power to enact such influence in the workplace, which is supported by a participant?s comments that when professional HWs have doubts about confidentiality of the service, this also impacts the perceptions held by HWs of other occupational cadres.  Further analysis of survey data also demonstrates that stigma is particularly a problem for professional HWs. When controlling for age and race, PCHWs had 2.4 times the odds of perceiving HIV stigma in the workplace when compared to non-PCHWs. This finding may be attributable to the fact that professional HWs are in a unique position to understand and bear witness to all forms of stigma that exist in the hospital, based on their ?insider? (emic) and ?outsider? (etic) perspectives on the HIV experience (Greeff et al., 2008; Weiss et al., 1992). Above and beyond the stigmatizing behaviours discussed previously, the results of the FGIs suggest professional HWs face additional challenges when it comes to HIV stigma in the workplace. First, professional HWs may exhibit stigmatizing behaviours towards PLWHA through the use of universal precautions only with HIV positive patients, which may perpetuate the ?fear of casual contact? (Nyblade et al., 2009) and reinforce instrumental stigma when these behaviours are observed by others (e.g. students or other HWs). Second, as professional HWs are 	 ? 93	 ?implicated as perpetrators of gossip in the workplace, they would be privy to examples of symbolic stigma, where HWs perceived to be HIV positive would be subject to moral judgments based on assumptions of promiscuity (Nyblade et al., 2009). Finally, and possibly most importantly, both by virtue of participating in or bearing witness to these various stigmatizing behaviours and in contending with the attitude that they ?should know better? (Namakhoma et al., 2010), professional HWs exposed to HIV may experience internalized stigmas that manifest in self-stigmatizing behaviours including self-diagnosis, self-medication, denial and in extreme cases even suicide (Uebel et. al., 2007; Namakhoma et al., 2010).  Thus, when professional HWs are not comfortable with HIV in the workplace, they may perpetuate stigma through their stigmatizing attitudes and behaviours, which sets a negative example for other HWs to emulate. In their role as witness to stigmatization, the silence of a professional HW may indicate what constitutes acceptable behaviors. In either role, the motivation for a professional HW may be to avoid becoming victims themselves. The fear of devaluation and subsequent loss of status may lead a professional to exhibit behaviors such as isolation and the avoidance of HIV services out a fear to not want to disclose (Mbonu et al., 2009). Through (negative) role modeling, these behaviours may be normalized for their non-professional colleagues. Thus, the power of a professional and their desire for self-preservation may perpetuate a cycle of stigma within the workplace.  4.6 Recommendations Arising from Study Results   A number of recommendations to address the aforementioned barriers arise from the results of this study. Given the multiple dimensions of each barrier, and the complex interactions between them, any recommendations aimed at improving uptake of HCT must recognize that ?one size does not fit all? (T2). Instead, careful consideration of the individual, local and social 	 ? 94	 ?contexts in which HCT is provided will help identify appropriate recommendations to address identified barriers (Obermeyer et al., 2011).  4.6.1 Privacy and Confidentiality  Educating HWs about national policies and guidelines around HIV in the workplace may provide clear guidance on rights and responsibilities of HWs. In June 2012, RSA?s Department of Labour approved a Code of Good Practices on HIV and AIDS and the World of Work as an addition to its Employment Equity Act (1998), to remain current with ILO guidelines. While this document addresses many of the barriers that participants in this study identified, it lacks specific guidance on how to operationalize its recommendations. However, the assurance by the government that barriers such as confidentiality, privacy and stigma are important to address is in and of itself an important fact that HWs should be aware of.  With respect to privacy, the Code of Good Practices on HIV and AIDS and the World of Work (2012) states that ?legal privacy is a requirement by law?, but does not go further to elaborate on how exactly privacy is to be protected. One way that the perceived lack of privacy observed in this study may be addressed is to promote wider knowledge that the OHS is a holistic service. Since the OHS provides comprehensive medical exams, TB screening and other medical services, it is a ?holistic service?. To ensure that this is clear, each nurse at the OHS should perform many services within a given time slot, rather than specializing in one service at one time (Arimoto et al., 2013; Greeff et al., 2008).  The Code of Good Practices on HIV and AIDS and the World of Work (2012) also states that while testing services can be provided in the workplace, the results should be confidential. As employee health records are housed separately from general public records at the OH units in the three study sites, it appears that employee health records are not at risk of the same breaches 	 ? 95	 ?in confidentiality as patient care records. What FG participants highlighted is that there is a misperception that employee health records are housed within the larger hospital system. However, beyond where employee health records are housed, the idea of access is also necessary to address. While a ?shared confidentiality? exists in healthcare settings, this is mostly for the purposes of providing healthcare to a patient. FG participants noted that some people perceive that diagnoses are shared with employers for the purpose of sick leaves or with human resource officials for the purpose of data gathering, which, of course, would be counter to all international standards for occupational health services. Addressing this misperceived lack of confidentiality requires a) ensuring that implementation of OH HIV services do indeed adhere to the strict privacy and confidentiality rules, liable to sanctions or penalties (including access of employee files only by OHS employees and those involved in providing healthcare to the HW in question and b) communicating how the OHS maintains its health records to all HWs in order to increase trust in the service.  To help with the latter suggestion, and in keeping with FG participants? desire for in-service training and print materials on policies and procedures specific to their role, education about how HWs health records are kept confidential should be part of a comprehensive ?confidentiality awareness campaign?, that may: a) assure HWs of confidentiality during all processes with HCT; b) provide HWs with policies and guidelines concerning HIV in the workplace (Labor Equity Act, among others); c) ensure that HWs are aware of their rights to confidential care, as situated within these policies; and d) provide an explanation of how confidentiality should be and is maintained. Such a campaign may have two advantages: 1) it may require that confidentiality within the workplace be audited and system-level breaches rectified, and 2) it educates the workforce on expectations around confidentiality so that 	 ? 96	 ?individual breaches in confidentiality can be addressed as they arise (e.g. through appropriate disciplinary action).   Finally, the Code of Good Practices on HIV and AIDS and the World of Work (2012) states that treatment and support for HIV be provided to employees, whether on-site or not. Thus, to improve the value of the HCT service, hospitals should enable professional HWs in the OHS to initiate ART, so that the service extends beyond testing to on-site treatment as well (Fairall et al., 2012; Callaghan et al., 2010; Wood et al., 2009).  The survey results may support the functionality of including treatment as part of a holistic service, as nearly three-quarters of HWs surveyed indicated if a co-worker found out they were HIV positive they would want to get treatment as soon as possible. Thus, if the OHS provided treatment, they would be able to offer this additional feature as soon as a test confirmed status as positive, potentially increasing the value of this service for HWs accessing HCT. The literature suggests an additional benefit of providing ART is that access to treatment may reduce the fear of HIV as a ?death sentence? (Zuch & Lurie, 2012), thereby reducing the stigma associated with a fear of dying. Once the mechanism for offering HIV testing and treatment is strengthened, it should be advertised within the auspices of a confidential holistic OHS (WHO/ILO Guidelines, 2010).  4.6.2 Stigma    Addressing HIV stigma in the workplace as a barrier to uptake of HCT services is a complex but necessary process. As the results of this study and prior literature suggest, HIV stigma is difficult to define and manifests in different ways (Nyblade, 2006; Genberg et al. 2008 in Ugarte et al. 2013). Thus, prior to the institution of interventions to reduce HIV stigma in healthcare facilities, it is necessary to measure and quantify the complex nature of HIV/AIDS stigma within these settings (Genberg et al., 2008). While the current study has detected HIV 	 ? 97	 ?stigma and acknowledges that various typologies of stigma exist within the HW population at these hospitals, this is not sufficient. The prevalence and magnitude of the various components of HIV stigma and factors that contribute to its existence within this population must also be measured using appropriate instruments (Ugarte et al., 2008; Holzemer et al., 2009). Where such efforts to measure the stigma experience involve partnerships with HWs who are PLWHA, the process of generating the measurement may also prove beneficial (Stackpool-Moore et al., 2013). Validated instruments, such as the People Living with HIV Stigma Index, have gained momentum as tools to measure and detect changing trends in stigma and discrimination within specific populations (People Living with Stigma Index website found at: http://www.stigmaindex.org/). As it is a community-based initiative that directly involves members of the population of interest, the process of completing the index links participants together and initiates communication to ?find a solution, not for now, but maybe later? (Stackpool-Moore et al., 2013). The analytic results generated from the index can be used to monitor and evaluate stigma-reduction efforts while simultaneously highlighting neglected areas within a population over time. However, it should be noted that this measure has yet been applied in the health care setting.    The measurement of HIV stigma should also be coupled with targeted interventions to address the attitudes and behaviours that perpetuate its existence. A number of recent studies have demonstrated the benefits of such efforts in decreasing the attitudes and behaviours associated with HIV stigma in health care settings, namely the avoidance of PLWHA (e.g. through a fear of causal contact) and prejudicial attitudes (e.g. the associations of HIV infection with immoral behaviour) (Nyblade et al., 2009; Li et al., 2013). Despite the fact that these interventions target stigma within the HW-patient relationship, rather than between HWs 	 ? 98	 ?themselves, these studies provide much insight into the components of stigma reduction efforts that can make them effective. Specifically, at the individual level it is important to increase awareness among HWs about stigma and its detrimental effects, correct inappropriate fears related to HIV (e.g. mechanisms of transmission and shifting focus to HIV as a chronic illness), and allow HWs to reflect on ?shame and blame? value judgments (e.g. that PLWHA have engaged in immoral behaviour) (Nyblade et al., 2009). This corroborates suggestions made during FGs to educate HWs about the psychosocial effects of HIV infection and the negative consequences of gossip (i.e education to ?treat everyone the same?).  However, such individual-level education and skill-building activities alone may be insufficient to initiate behaviour change among HWs (Li et al., 2013). Thus, these efforts should be coupled with: 1) environmental level changes, including removal of structural barriers such as inadequate access to universal precautions; and 2) policy level changes, including the creation and communication of national and workplace based policies that promote safety of HWs, whether they are HIV positive or otherwise (Nyblade et al., 2009). These recommendations also support FG participant?s suggestions to educate HWs on the Employment Equity Act, policies on HIV in the workplace, and existing support structures such as the OHS. Finally, stigma reduction efforts should also increase institutional support and address problematic social norms that exist among HWs (Li et al., 2008; Li et al., 2009). When interventions are constructed to target each of these levels, significant effects on HIV stigma may be observed. In a recent cluster-randomized trial of HWs at 40 county-level hospitals in China, Li et al. (2013) demonstrated that a stigma reduction intervention that integrated behavioural and structural components resulted in a significant decrease in prejudicial attitudes and avoidance of PLWHA and an increase in institutional support. As the behavioral change included training a critical mass of popular 	 ? 99	 ?opinion leaders to disseminate stigma reduction messages, this study demonstrates that role modeling by ?leaders? is an effective strategy to promoting changes in attitudes and behaviors among HWs. 4.6.3 Role Modeling   As mentioned, stigma can only exist where there is power, thus, two groups are especially important in role modeling positive behaviors: managers and professional HWs.  In this study almost a quarter of HWs surveyed believed that there would be no change to the employer-employee relationship, and that only 41% believe that an employer would actively encourage an HIV positive employee to protect themselves against infectious diseases while at work. This may suggest that at present, while managers and supervisors are not actively discriminating against HIV positive HWs, they are not necessarily improving the working environment for HW who are PLWHA. This is supported by the recommendations raised by FG participants, who identified managers as a cohort that lack adequate knowledge of HIV in the workplace and require further training on maintaining confidentiality, workplace policies on HIV/AIDS and fostering supportive working environments for all employees, in keeping with the Code of Good Practices on HIV and AIDS and the World of Work (2012). Furthermore, as managers are seen as leaders and ?higher people? in the workplace, their actions towards HWs who are HIV positive, along with their promotion and use of the OHS, may improve perceptions of the service. Increasing awareness among managers and supervisors regarding this responsibility is imperative, as they are perceived by FG participants to be role models for other HWs, regardless of their actions.   Role modeling in professional HWs differs from the supportive role modeling offered by managers, in that it may be less overt, but is perhaps the most necessary for mobilizing action 	 ? 100	 ?among HWs. Professional HWs are also perceived to be role models, whether they want to be or not.  This requires that they embody the values and principles of their given professions in order to uphold the oaths they have taken.  First, by respecting the privacy of patients and colleagues, they can demonstrate how to create a supportive environment for those who access services.  Second, by upholding their professional responsibility to maintain confidentiality of patient and co-worker information, they will elevate the standard of behaviour expected of all HWs. Finally, by treating all patients equally through the use of universal precautions in all cases, they demonstrate ?equality? in service delivery (Nyblade et al., 2009). In order to support professional HWs to uphold this level of professionalism, as a FG participant reminds us, this means professional HWs must understand what it means to be ?professional?, through continued education that reintroduces and reinforces topics like confidentiality, privacy, and the role of the HW in alleviating stigma. In the absence of codes of conduct, and without guidance on expectations and consequences, professionals may unintentionally continue to promote behaviors that go against their ?calling?.     4.6.4 Support Groups    While the preceding recommendations have centered on ?top down? process that try to make the workplace more supportive of the HIV positive individual, actions that empower PLWHA and those around them are noted to be among the most efficacious interventions to date (Liamputtong et al., 2013).  This is perhaps the reason that FG participants seemed to want support groups within the workplace to specifically encourage dialogue about HIV in the workplace. The participants in this study differed from those in the literature that had formed support groups from within PLWHA (Lyttleton, 2004; Liamputtong et al., 2013). Participants were clear that they did not want support groups for HIV positive individuals, but rather wanted 	 ? 101	 ?support groups that involve all HWs focused on the topics related to HIV. This stems from the desire to move toward a unity over the realization that ?there is none that is [n]either infected [n]or affected?.  Thus, support groups provide the opportunity for HWs to support themselves and find role models among their peers44 (Liamputtong et al. 2013: 386-393).  Regardless of the ways in which support groups are formed, studies support that HIV and AIDS support groups accommodate ?social normalization of HIV infection by creating the social platform and moral support for public disclosure of HIV status? (Lyttleton et al., 2004; Lyttleton et al., 2007).  	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?	 ?44	 ?Enthusiasm for support groups was evident when I was asked separately in 2 FGIs when we would be meeting again.  As a participant clarified to the person asking, ?this is a focus group, not a support group?.  	 ?	 ? 102	 ?Chapter 5: Conclusion  5.1 Summary This study identifies that concern about lack of privacy, fear that confidentiality will not be maintained and HIV stigma are the primary barriers to uptake of occupationally-based HCT by HWs in the study sites. This study demonstrates that even though political will exists, in the absence of clear strategies to implement services that have addressed barriers to uptake using evidence-based strategies, services targeting HIV in HWs may be unsuccessful. While this study discusses and recognizes the complexities associated with the concepts of privacy, confidentiality and stigma, it also offers some recommendations to overcome these barriers. To overcome privacy and confidentiality concerns, this study recommends that policies affirming employee health records be accessible only to those OH professionals that require access to these records be communicated to all HWs, alongside careful record-keeping of who accesses the patient?s record, when and for what purpose. This requires educating HWs on OHS policies and guidelines that govern HIV in the workplace in order for HWs to know ?right from wrong? as it pertains to confidentiality and so that they can monitor, and, if necessary, advocate for structural changes to promote HIV supportive environments.  Finally, OH practitioners must be absolutely transparent about OHS record keeping. To address HIV stigma, stigma reduction interventions that combine structural and behavioural components are endorsed. Finally, role modeling of positive behaviours by managers and professional HWs, along with encouraging HWs to form support groups, are necessary steps to challenge negative attitudes toward HIV and AIDS in the workplace.  	 ? 103	 ?5.2 Limitations and Strengths of the Study In order for the reader to appropriately contextualize the results discussed above, there are important limitations to this study that should be acknowledged. First, the population sampled in this study consisted of HWs at three large regional hospitals, all from the Free State province. Although these hospitals employ a significant proportion of the HWs in this region, their size and urban location differentiates them from the smaller or rurally based community and primary care centers in the remainder of the province. Thus, generalization of the findings of this study to other geographic regions and demographic groups may be limited.  Furthermore, bias in response may occur as the self-administered survey relied on self-reported measures administered during business hours directly at participants? place of work. This may have discouraged some individuals from participating, or caused those who did complete the survey to do so in a hurried fashion due to time constraints. In addition, a quota sampling strategy was used, rather than a random sample, which may have introduced bias in the selection of participants. Also, the cross sectional-nature of the study design precludes predictive or causal interpretations of the findings.  Additionally, the design of some survey questions limits the analysis and interpretation that can be drawn from the responses provided. Specifically, the association between sociodemographic characteristics and questions that asked participants about their co-worker?s perspectives and behaviours (e.g. questions about whether co-workers would want to know and to whom they would disclosure their HIV status) are somewhat difficult to interpret, as subject of the sociodemographic data is the participant, whereas the subject of the aforementioned questions are participants? co-workers. Similarly, for questions using the ?list of options? response type (e.g. related to reasons for not using the OHS for HCT, where ?fear that 	 ? 104	 ?confidentiality will not be maintained? was the most frequent answer), as participants were not directly asked about their perspective on each option, analyses extended beyond the frequency with which each response was selected (e.g. associations to the sociodemographic characteristics) were not felt to be appropriate.   Fourth, FGIs were limited in sample size, although the main themes reoccurred in each focus group and saturation was reached in some areas.  In addition, sampling bias may have been introduced by the sampling method and the interview time schedule (during the work day, with short notice given to some participants).  The generalizability of the findings are limited by the unique features of tertiary-care hospital employees in Free State province, SA, however the ?proposed process of stigmatization? conceptualization of stigma may be valuable to similar settings. Additionally, like all qualitative studies that use focus groups, this study has the potential to be biased by the researcher?s focus. Finally, as both the survey and FGIs generated data based on participant?s perspectives, these methods are limited in that there may be a gap between stated and actual behavior, as well as problems with meaning and omission among participants.  The mixed-methods approached used in this study addressed some of the stated limitations by helping contextualize quantitative finds through FGIs. As the questionnaire asked participants their views on coworkers, it captured the culture and perceptions of HIV related to the workplace and thus was appropriate in creating an explanatory model. These perceptions were further explored in FGIs where participants were particularly forthcoming. In engaging health and safety representatives as part of the FGIs and allowing employees of the OHS be a part of the groups as well (if they wanted to be), the participatory nature of the focus groups themselves highlight necessary first steps in fostering dialogue about HIV in the workplace. 	 ? 105	 ?Furthermore, as there are gaps in the literature related to provider-specific stigmatization, as opposed to generalized stigmatization, this study provides context around the potential effects of stigmatization in hospital settings and its effects on HWs specifically (as opposed to patients). 5.3 Directions for Future Research 	 ? With due consideration of the methodological limitations described above, the findings of this study have important implications for future research, as scaling-up of workplace-based HCT programs in RSA continues.  First, research on confidentiality requires explicit examples of what constitute breaches in confidentiality and how to strengthen confidentiality in health systems.  Second, measures of stigma should be included as a necessary first step in the monitoring and evaluation of the process of introducing HCT into hospital-based HCT programs. Third, stigma interventions that specifically target HWs in health care settings should be implemented, monitored, and evaluated.  Fourth, a single observation was made within FGs suggesting that laundry workers within this sample may experience fewer stigmas than other cadres. Further research should be conducted to see whether this observation is anomalous, or represents a demonstrable phenomenon. If found to have merit, further investigation specifically on this cohort may reveal factors that contribute to the normalization of HIV within this healthcare setting.    	 ? 106	 ?Bibliography  	 ?Abdool Kareem, S.S. (2005). HIV/AIDS in South Africa. Cambridge: Cambridge University Press.  Abrahams, N., & Jewkes, R. (2012). Managing and resisting stigma: a qualitative study among people living with HIV in South Africa. Journal of the International AIDS Society, 15, 17330.  Aggleton, P., & Parker, R. G. (2002). World AIDS Campaign 2002-2003: A conceptual framework and basis for action: HIV/AIDS Stigma and Discrimination. Switzerland: UNAIDS.  Altman, I. (1976). Privacy: A Conceptual Analysis. Environment and behavior, 8(1), 7-29.  Arimoto, Y., Ito, S., Kudo, Y., Tsukada, K. (2013). Stigma, social relationship and HIV testing in the workplace: evidence from South Africa. (No. 2012-06). Center for Economic Institutions, Institute of Economic Research, Hitotsubashi University.  Bester, C. L., & Engelbrecht, M. C. (2009). Job satisfaction and dissatisfaction of professional nurses in primary health care facilities in the free state province of South Africa. Africa Journal of Nursing and Midwifery, 11(1), 104-117.  Blaauw, D., Ditlopo, P., Maseko, F., Chirwa, M., Mwisongo, A., Bidwell, P., & Normand, C. (2013). Comparing the job satisfaction and intention to leave of different categories of health workers in Tanzania, Malawi, and South Africa. Global health action, 6, 19287.  Buregyeya, E., Nuwaha, F., Wanyenze, R. K., Mitchell, E. M., Criel, B., Verver, S., ... & Colebunders, R. (2012). Utilization of HIV and Tuberculosis Services by Health Care Workers in Uganda: Implications for Occupational Health Policies and Implementation. PloS one, 7(10), e46069.  Bryman, A. (2012). Social research methods. Oxford university press.  Cameron, E. (2007). Normalizing Testing Normalizing AIDS. Theoria, 54(112), 99-108.  Cameron, E. (2005). Witness to AIDS. London: IB Tauris.  Campbell, M., Fitzpatrick, R., Haines, A., Kinmonth, A. L., Sandercock, P., Spiegelhalter, D., & Tyrer, P. (2000). Framework for design and evaluation of complex interventions to improve health. BMJ: British Medical Journal, 321(7262), 694.  Chen, L., Evans, T., Anand, S., Boufford, J. I., Brown, H., Chowdhury, M., Cueto, M., Dare, L., Dussault, G., Elzinga, G., Fee, E., Habte, D., Hanvoravongchai, P., Jacobs, M., Kurowski, C., Michael, S., Pablos-Mendez, A., Sewankambo, N., Solimano, G., Stilwell, B., de Waal, A., & 	 ? 107	 ?Wibulpolprasert, S. (2004). Human resources for health: overcoming the crisis. The Lancet, 364(9449), 1984-1990.  Colvin, C. J., Fairall, L., Lewin, S., Georgeu, D., Zwarenstein, M., Bachmann, M., ... & Bateman, E. D. (2010). Expanding access to ART in South Africa: the role of nurse initiated treatment. SAMJ: South African Medical Journal, 100(4), 210-212.  Colvin, C. J. (2011a). HIV/AIDS, chronic diseases and globalisation. Globalization and health, 7, 31.  Colvin, C. J. (2011b). Think locally, act globally: developing a critical public health in the global South. Critical Public Health, 21(3), 253-256.  Connelly, D., Veriava, Y., Roberts, S., Tsotetsi, J., Jordan, A., DeSilva, E., ... & DeSilva, M. B. (2007). Prevalence of HIV infection and median CD4 counts among health care workers in South Africa. South African Medical Journal, 97(2), 115-120.  Corbett, E. L., Dauya, E., Matambo, R., Cheung, Y. B., Makamure, B., Bassett, M. T., ... & Hayes, R. J. (2006). Uptake of workplace HIV counselling and testing: a cluster-randomised trial in Zimbabwe. PLoS medicine, 3(7), e238.  Coovadia, H.M. (2000). Access to voluntary counselling and testing for HIV in developing countries. Annals of the New York Academy of Sciences, 918, 57?63.   Craig, P., Dieppe, P., Macintyre, S., Michie, S., Nazareth, I., & Petticrew, M. (2008). Developing and evaluating complex interventions: the new Medical Research Council guidance. BMJ: British Medical Journal, 337, a1655.   Day, J., Miyamura, K., Grant, A., Leeuw, A., Munsamy, J., Baggaley, R., and Churchyard, G. (2003). Attitudes to HIV voluntary counselling and testing among mineworkers in South Africa: Will availability of antiretroviral therapy encourage testing?, AIDS CARE, 15(5), 665?672.   Deacon, H., Uys, L., & Mohlahlane, R. (2009). HIV and stigma in South Africa. In HIV/AIDS in South Africa 25 Years On (pp. 105-120). Springer New York.  Deacon, H.J., and Stephney, I. (2007) HIV/AIDS, stigma and children: a literature review. Cape Town: Human Sciences Research Council.   Deacon, H.J. (2006). Towards a sustainable theory of health-related stigma: lessons from the HIV/AIDS literature. Journal of Community and Applied Social Psychology, 16(6), 418?425.   Deacon, H.J., Stephney, I., and Prosalendis, S. (2005) HIV/AIDS stigma: a theoretical and methodological review. Cape Town: Human Sciences Research Council.   De Cock, K., Mbori-Ngacha ,D., and Marum, E. (2002). Shadow on the continent: public health and HIV/AIDS in Africa in the 21st century. The Lancet, 360, 67?71.  	 ? 108	 ? Dieleman, M., Biemba, G., Mphuka, S., Sichinga-Sichali, K., Sissolak, D., van der Kwaak, A., & van der Wilt, G. J. (2007). ?We are also dying like any other people, we are also people?: perceptions of the impact of HIV/AIDS on health workers in two districts in Zambia. Health Policy and Planning, 22(3), 139-148.  Doherty, T., Chopra, M., Nkonki, L., Jackson, D., and Greiner, T. (2006). Effect of the HIV epidemic on infant feeding in South Africa: ?When they see me coming with the tins they laugh at me?. Bulletin of the World Health Organization, 84, 90?96.   Engelbrecht, M. C., Bester, C. L., Van den Berg, H., & Van Rensburg, H. (2008). A study of predictors and levels of burnout: the case of professional nurses in primary health care facilities in Free State. South African Journal of Economics, 76(s1), S15-S27.  Field, Andy. (2009) Discovering statistics using SPSS, 3rd edition, Sage Publications, page 273-4.  Freidson, E. (2001). Professionalism, the third logic: on the practice of knowledge. University of Chicago Press.  Fox, R. C. (1989). The sociology of medicine: a participant observer's view. Englewood Cliffs, Nova Jersey Nova Jersey: Prentice Hall.  George, G., Quinlan, T., Reardon, C., & Aguilera, J. (2012). Where are we short and who are we short of? A review of the human resources for health in South Africa. Health SA Gesondheid, 17(1).   Goffman, E. (1963). Stigma: notes on the management of spoiled identity. New Jersey: Prentice  Hall.  Greeff, M., and Phetlhu, D.R. (2007). The meaning and effect of HIV/AIDS stigma for people living with AIDS and nurses involved in their care in the North West Province, South Africa. Curationis, 30(2), 12?23.  Greeff, M., Phetlhu, D.R., Makoae, L.N., Dlamini, P.S., Holzemer, W.L., Naidoo, J., Kohi, T.W., Uys, L.R., and Chirwa, M.L. (2008a). Disclosure of HIV status: Experiences and perceptions of persons living with HIV/AIDS and nurses involved in their care in five African countries. Qualitative Health Research, 18(3), 311?324.   Greeff, M., Uys, L.R., Holzemer, W.L., Makoae, L.N., Dlamini, P.S., Kohi, T.W., Chirwa, M.L., Naidoo, J., and Phetlhu, D.R. (2008b) Experiences of HIV/AIDS stigma of PLWA and nurses from five African Countries. African Journal of AIDS Research, 10(1), 78?108.   Harrison, D. (2010). An overview of health and health care in South Africa 1994-2010: priorities, progress and prospects for new gains.  Discussion document on Health in South Africa 	 ? 109	 ?1994-2010.  National Health Leaders' Retreat, Muldersdrift, January 24-26, 2010.  Available at http://www.doh.gov.za/docs/reports/2010/overview1994-2010.pdf  Herek, G.M. (2002). Thinking about AIDS and stigma: a psychologist?s perspective. Journal of Law, Medicine and Ethics, 30, 594?607.   Herek G.M., Capitanio J.P. (1999). AIDS stigma and sexual prejudice. Am Behav Sci, 42, 1130?1147.  HIV, I. (2004). AIDS and work: global estimates, impact and response. The ILO Programme on HIV/AIDS and the World of Work. Geneva: ILO.  Holzemer, W.L., Uys, L., Makoae, L., Stewart, A., Phetlhu, R., Dlamini, P., Greeff, M., Kohi, T.W., Chirwa, M., Cuca, Y., and Naidoo, J. (2007a). A conceptual model of HIV/AIDS stigma from five African countries. Journal of Advanced Nursing, 58(6), 541?551.   Holzemer, W.L., Uys, L.R., Chirwa, M.L., Greeff, M., Makoae, L.N., Kohi, T.W., Dlamini, P.S., Stewart, A.L., Mullan, J., Phetlhu, R.D., Wantland, D., and Durrheim, K. (2007b) Validation of the HIV/AIDS Stigma Instrument ? PLWA (HASI-P). AIDS Care, 19(8), 1002?1012.   Holzemer, W. L., Makoae, L. N., Greeff, M., Dlamini, P. S., Kohi, T. Q., Chirwa, M. L., Naidoo, J.R., Durrheim, K., Cuca, Y., & Uys, L. R. (2009). Measuring HIV stigma for PLHAs and nurses over time in five African countries. SAHARA J (Journal of Social Aspects of HIV/AIDS Research Alliance), 6(2).  HRH South Africa, (2011) Human Resources for Health South Africa - HRH Strategy for the Health Sector: 2012/13 ? 2016/17: available at http://www.doh.gov.za/docs/reports/2013/Human_resources.pdf; Accessed on September 29, 2013  Hutchinson, P.L., and Mahlalela, X. (2006). Utilization of voluntary counseling and testing services in the Eastern Cape, South Africa. AIDS Care, 18, 446?455.   Jennings, R., Mulaudzi, J., Everatt, D., Heywood, M., and Richter, M. (2002). Discrimination and HIV/AIDS. Paper for the Department of Health Strategy & Tactics and the AIDS Law Project. http://alp.org.za.dedi20a.your-server.co.za/images/upload/HIV.AIDSstigma.pdf    Jewkes, R. (2006). Beyond stigma: social responses to HIV in South Africa. Lancet, 368, 430?431.   Joshi, R., Reingold, A. L., Menzies, D., & Pai, M. (2006). Tuberculosis among health-care workers in low-and middle-income countries: a systematic review. PLoS medicine, 3(12), e494.  Kalichman, S.C., and Simbayi, L.C. (2003). HIV testing attitudes, AIDS stigma, and voluntary HIV counselling and testing in a black township in Cape Town, South Africa. Sexually 	 ? 110	 ?Transmitted Infections, 79, 442?447.  Kalichman, S.C., and Simbayi, L. (2004). Traditional beliefs about the cause of AIDS and AIDS related stigma in South Africa. AIDS Care, 16, 572?580.  Kalichman, S.C., Simbayi, L.C., Jooste, S., Toefy, Y., Cain, D., Cherry, C., and Kagee, A. (2005). Development of a brief scale to measure AIDS-related stigma in South Africa. AIDS and Behaviour, 9(2), 135?143.  Kalichman, S.C., Simbayi, L.C., Cain, D., Jooste, S., Skinner, D., and Cherry, C. (2006). Generalizing a model of health behaviour change and AIDS stigma for use with sexually transmitted infection clinic patients in Cape Town, South Africa. AIDS Care, 18, 178?182.  Kekana, H. P. P., Du Rand, E. A., & Van Wyk, N. C. (2007). Job satisfaction of registered nurses in a community hospital in the Limpopo Province in South Africa. Curationis, 30(2), 24-35.  Kling, S. (2010). Confidentiality in medicine. Current Allergy & Clinical Immunology, 23(4), 196-198.   Kober, K., & Van Damme, W. (2004). Scaling up access to antiretroviral treatment in southern Africa: who will do the job?. The Lancet, 364(9428), 103-107.  Kohi, T.W., Makoae, L., Chirwa, M., Holzemer, W.L., Phetlhu, D.R., Uys, L., Naidoo, J., Dlamini, P.S., and Greeff, M. (2006). HIV and AIDS stigma violates human rights in five African countries. Nursing Ethics, 13(4), 404?415.  Larrabee, J. H., Janney, M. A., Ostrow, C. L., Withrow, M. L., Hobbs, G. R., & Burant, C. (2003). Predicting registered nurse job satisfaction and intent to leave. Journal of Nursing Administration, 33(5), 271-283.  Levin, J., & Arluke, A. (1985). An exploratory analysis of sex differences in gossip. Sex Roles, 12(3-4), 281-286.  Li, L., Wu, Z.Y., Liang, L. J., Lin, C., Guan, J., Jia, M., Rou, K., & Yan, Z. (2013). Reducing HIV-Related Stigma in Health Care Settings: A Randomized Controlled Trial in China. American journal of public health, 103(2), 286-292.  Li L, Liang LJ, Wu Z, Lin C, Wen Y. (2009). Individual attitudes and perceived social norms: reports on HIV/ AIDS-related stigma among service providers in China. International Journal of Psychology, 44(6), 443-450.  Li L, Liang LJ, Wu ZY, Lin C, Wu S. (2008). Institutional support for HIV/AIDS care in China: a multilevel analysis. AIDS Care, 20(10), 1190-1196.  	 ? 111	 ?Liamputtong, P. (2013). Stigma, discrimination and living with HIV/AIDS: a cross-cultural perspective. Springer.  Lindau, S. T., Jerome, J., Miller, K., Monk, E., Garcia, P., & Cohen, M. (2006). Mothers on the margins: Implications for eradicating perinatal HIV. Social Science & Medicine, 62, 59?69.   Link, B.G., and Phelan, J.C. (2001). On Stigma and its public health implications. Annual Review of Sociology, 27, 363?385.  Link, B.G., and Phelan, J.C. (2006). Stigma and its public health implications. The Lancet, 367, 528?529.  Link, B. G., & Phelan, J. C. (2013). Labeling and stigma. In Handbook of the sociology of mental health (pp. 525-541). Springer Netherlands.  Lyttleton, C. (2004). Fleeing the fire: Transformation and gendered belonging in Thai HIV/AIDS support groups. Medical Anthropology, 23, 1?40.   Lyttlteon, C., Beesey, A., & Sitthikriengkrai, M. (2007). Expanding community through ARV provision in Thailand. AIDS Care, 19(Suppl 1), S44?S53.   MacPhail, C. L., Pettifor, A., Coates, T., & Rees, H. (2008). ?You must do the test to know your status?: Attitudes to HIV voluntary counseling and testing for adolescents among South African youth and parents. Health Education & Behavior, 35(1), 87-104.  Magnusson, R. S. (1994). Privacy, confidentiality and HIV/AIDS health care. Australian Journal of Public Health, 18(1), 51-58.  Makombe, S. D., Jahn, A., Tweya, H., Chuka, S., Yu, J. K. L., Hochgesang, M., & Harries, A. D. (2007). A national survey of the impact of rapid scale-up of antiretroviral therapy on health-care workers in Malawi: effects on human resources and survival. Bulletin of the World Health Organization, 85(11), 851-857.  Makoae, L., Greeff, M., Phethlu, R., Uys, L.R., Naidoo, J.R., Kohi, T., Dlamini, P., Chirwa, M., and Holzemer, W.L. (2008) Coping strategies for HIV and AIDS: A multinational African Study. Journal of the Association of Nurses in AIDS Care, 19(2), 137?146.  Makoae, L. N., Portillo, C. J., Uys, L. R., Dlamini, P. S., Greeff, M., Chirwa, M., ... & Holzemer, W. L. (2009). The impact of taking or not taking ARVs on HIV stigma as reported by persons living with HIV infection in five African countries. AIDS care, 21(11), 1357-1362.  Malcolm, A., Aggleton, P., Bronfman, M., Galvao, J., Mane, P., & Verrall, J. (1998). HIV-related stigmatization and discrimination: Its forms and contexts. Critical Public Health, 8(4), 347-370.  	 ? 112	 ?Maman, S., Abler, L., Parker, L., Lane, T., Chirowodza, A., Ntogwisangu, J., et al. (2009). A com- parison of HIV stigma and discrimination in five international sites: The influence of care and treatment resources in high prevalence settings. Social Science & Medicine, 68(12), 2271?2278.   Maman, S., Mbwambo, J., Hogan, M., Kilonzo, G., Sweat, M., and Weiss, E. (2001). HIV and partner violence. Implications for HIV voluntary counselling and testing programmes in Dar es Salaam, Tanzania. New York: Horizons/The Population Council.   Mann, J. (1996). The impact of homophobia and other social biases on AIDS. United States of America Public Media Center.  Maughan-Brown, B. G. (2006). Attitudes towards people with HIV/AIDS: stigma and its determinants amongst young adults in Cape Town, South Africa. South African Review of Sociology, 37(2), 165-188.  Mbonu, N. C., van den Borne, B., & De Vries, N. K. (2009). Stigma of people with HIV/AIDS Sub-Saharan Africa: A literature review. Journal of Tropical Medicine, 2009, 1?14.    Menzies, D., Pai, M., & Comstock, G. (2007). Meta-analysis: new tests for the diagnosis of latent tuberculosis infection: areas of uncertainty and recommendations for research. Annals of internal medicine, 146(5), 340-354.  Moffatt, S., White, M., Mackintosh, J., & Howel, D. (2006). Using quantitative and qualitative data in health services research what happens when mixed method findings conflict. BMC Health  Services Research, 6, 28.  Moodley, P. P., & Bachmann, M. O. (2002). Inequity in occupational health services for government hospital workers in South Africa. Occupational Medicine, 52(7), 393-399.  Myeko N.B., Thlapuletsa F.B., Brandsel M.N.L., Ntlola N.L., (2012). Improving utilization of workplace HIV/AIDS programme for healthcare workers at Pelonomi Hospital. Available from: http://www.ghrp.ubc.ca/files/2012/09/Group-1-presentation-for-graduationworking-copy-11.FINAL_.pdf. Accessed on September 23, 2013   Naidoo, J.R., Uys, L.R., Greeff, M., Holzemer, W.L., Makoae, L., Dlamini, P., Phetlhu, R.D., Chirwa, M., and Kohi, T. (2007) Urban and rural differences in HIV/AIDS stigma in five African countries. African Journal of AIDS Research, 6(1), 17?23.   Namakhoma, I., Bongololo, G., Bello, G., Nyirenda, L., Phoya, A., Phiri, S., Phiri, S., Theobald, S., & Obermeyer, C. M. (2010). Negotiating multiple barriers: health workers' access to counselling, testing and treatment in Malawi. AIDS care, 22(S1), 68-76.  Naughton, J. D., & Vanable, P. A. (2013). HIV Stigmatization Among Healthcare Providers: Review of the Evidence and Implications for HIV Care. In Stigma, Discrimination and Living with HIV/AIDS, 97-114. Springer Netherlands. 	 ? 113	 ? Nyblade, L., Stangl, A., Weiss, E., & Ashburn, K. (2009). Combating HIV stigma in health care settings: what works?. Journal of the International AIDS Society, 12(1), 15.  Nyblade, L. (2006). Measuring HIV stigma: existing knowledge and gaps. Psychology, Health and Medicine, 11, 335?345.   Nyblade L., MacQuarrie, K., Fausta, P., et al. (2005). Working report measuring HIV stigma: results of a field test in Tanzania. Washington: USAID.   Obermeyer, C.M., and Osborn, M. (2007). The utilization of testing and counselling for HIV: A review of the social and behavioural evidence. American Journal of Public Health, 97(10), 1762?1774.   Obermeyer, C. M., Baijal, P., & Pegurri, E. (2011). Facilitating HIV disclosure across diverse settings: a review. Journal Information, 101(6), 1011-1023.  Ogden, J., and Nyblade, L. (2005). Common at its Core: HIV-Related stigma across contexts. Washington: International Center for Research on Women.   Parker, R., and Aggleton, P. (2003). ?HIV and AIDS-related stigma and discrimination: a concep- tual framework and implications for action?. Social Science & Medicine 57, 13?24.   Parker, R. G., & Aggleton, P. (2002). HIV/AIDS-related stigma and discrimination: A conceptual framework and an agenda for action. New York: Population Council.  Pillay, R. (2009). Work satisfaction of professional nurses in South Africa: a comparative analysis of the public and private sectors. Human resources for health, 7(1), 15.  POLICY Project, South Africa; Centre for the Study of AIDS, University of Pretoria; United States Agency for International Development; Chief Directorate: HIV, AIDS & TB, Department of Health, 2003  Purtilo R. Ethical Dimensions in the Health Professions. 3rd ed. Philadelphia: WB Saunders, 1999.   Rankin, W. W., Brennan, S., Schell, E., Laviwa, J., & Rankin, S. H. (2005). The stigma of being HIV-positive in Africa. PLOS medicine, 2(8), e247.  Raviola, G., Machoki, M., Mwaikambo, E., and Good, M.J. (2002). HIV, disease plague, demoralization and ?burnout?: resident experience of the medical profession in Nairobi, Kenya. Culture, Medicine & Psychiatry, 26(1), 55?86.  Rohleder, P., Kalichman, S. C., & Schwartz, L. (Eds.). (2009). HIV/AIDS in South Africa 25 years on: Psychosocial perspectives. Springer.  	 ? 114	 ?Scambler, G. (2009). Health?related stigma. Sociology of Health & Illness, 31(3), 441-455.  Shisana, O., Simbayi, L., et al. (2002). Nelson Mandela/HSRC Study of HIV/AIDS: South African National HIV prevalence, behavioral risks and mass media, household survey 2002. Cape Town: Human Sciences Research Council.   Shisana, O., Hall, E., Maluleke, K.R., Stoker, D.J., Schwabe, C., Colvin M., Chauveau, J., Botha, C., Gumede, T., Fomundam, H., Shaikh, N., Rehle, T., Udjo, E., and Gisselquist, D. (2003). The impact of HIV/AIDS on the health sector: national survey of health personnel, ambulatory and hospitalised patients and health facilities 2002. Cape Town: Human Sciences Research Council.   Shisana, O., Rehle, T., Simbayi, L., Parker, W., Zuma, K., Bhana, A., Connolly, C., Jooste, S., Pillay, V., et al. (2005). South African national HIV prevalence, HIV incidence, behaviour and communication survey, 2005. Cape Town: Human Sciences Research Council.   Siegler M. Confidentiality in medicine ? a decrepit concept? New Engl J Med 1982; 307: 1518-21. Reprinted in: Kuhse H, Singer P. Bioethics: An Anthology. Oxford: Blackwell, 2004: 490-492.   Simbayi, L.C., Kalichman, S., Strebel, A., Cloete, A., Henda, N., and Mqeketo, A. (2007). Internalized stigma, discrimination, and depression among men and women liv- ing with HIV/AIDS in Cape Town, South Africa. Social Science & Medicine, 64(9), 1823?1831.   Siyam?kela Project (2003). Siyam?kela: measuring HIV/AIDS related stigma: HIV/AIDS stigma indicators. Cape Town: POLICY project, Centre for the Study of AIDS, USAID and Depart- ment of Health.   Skinner, D., and Mfecane, S. (2005). Stigma, discrimination and the implications for people living with HIV/AIDS in South Africa. Journal of Social Aspects of HIV and AIDS, 1(3), 157?164.   South Africa 2007. Department of Health, HIV and AIDS and STI Strategic Plan for South Africa, 2007?2011. http://www.doh.gov.za/docs/misc/stratplan-f.html . Accessed 3 January 2009.   South Africa Department of Labour (1998) Employment Equity Act.  South Africa Department of Labour (2012) Code of Good Practices on HIV and AIDS and the World of Work.   South African Institute of Race Relations (2006) South Africa Survey Johannesburg: SAIRR, Health and Wellness, Available at: http://www.sairr.org.za/services/publications/south-africa-survey/south-africa-survey-online-20067-2007  South African National AIDS Council. (2012). National Strategic Plan on HIV, STIs and TB, 2012-2016. 	 ? 115	 ? Stackpool-Moore, L., Chirwa, M. L., de Croy, S., Hudson, A., Kamkwamba, D., Kundecha, R. M., Mahaka, P. T., Onyango, S., & Sinyemu, E. (2013). An Antidote to Stigma? The People Living with HIV Stigma Index in Malawi and the United Kingdom (UK). In Stigma, Discrimination and Living with HIV/AIDS (pp. 337-355). Springer Netherlands.  Strachan, B., Zabow, T., & Van der Spuy, Z. M. (2011). More doctors and dentists are needed in South Africa. South African Medical Journal, 101(8), 523-528.  Surlis, S., & Hyde, A. (2001). HIV-positive patients' experiences of stigma during hospitalization. Journal of the Association of Nurses in AIDS Care, 12(6), 68-77.  Tarwireyi, F., & Majoko, F. (2003). Health workers' participation in voluntary counselling and testing in three districts of Mashonaland East Province, Zimbabwe. The Central African journal of medicine, 49(5-6), 58.  Thomas, D. R. (2003). A general inductive approach for qualitative data analysis. School of Population Health, University of Auckland.  Turan JM, Bukusi EA, Cohen CR, Sande J, Miller S. (2008). Effects of HIV/AIDS on maternity care providers in Kenya. J Obstet Gynecol Neonatal Nurs, 37(5), 588-595.   Uebel, K., Friedland, G., Pawinski, R., & Holst, H. (2004). HAART for hospital health care workers-an innovative programme: occupational health: SAMJ forum.  Uebel, K.E., Nash, J., and Avalos, A. (2007). Caring for the caregivers: models of HIV/AIDS care and treatment provision for health care workers in southern Africa. The Journal of Infectious Diseases, 196, S500?504.   Ugarte, W. J., H?gberg, U., Valladares, E. C., & Ess?n, B. (2013). Measuring HIV- and AIDS-related stigma and discrimination in Nicaragua: results from a community-based study. AIDS Education and Prevention, 25(2), 164-178.  UNAIDS. (2001).The impact of voluntary counselling and testing. A global review of the benefits and challenges. Geneva: UNAIDS. Retrieved 1 May 2007 from http://www.uniads.org.   UNAIDS/WHO. (2004). UNAIDS/WHO policy statement on HIV testing. Geneva, Switzerland: UNAIDS.   UNAIDS. (2006). Report on the global AIDS epidemic. Retrieved 15 June 2008 from http://www.unaids.org/en/KnowledgeCentre/HIVData/GlobalReport/2006/   UNAIDS. (2008). 2008 Report on the Global AIDS Epidemic: Joint United Nations Programme on HIV/AIDS.   	 ? 116	 ?UNAIDS/WHO. (2008). Status of the global HIV epidemic: 2008 report on the global AIDS epidemic. Annex 1: HIV and AIDS estimates and data, 2007 and 2001. Available: http://data.unaids.org/pub/GlobalReport/2008/jc1510_2008_global_report_pp211_234_en.pdf. Accessed 15 September 2013.  Uys, L.R., Holzemer, W.L., Chirwa, M.L., Dlamini, P., Greeff, M., Kohi, T.W., Makoae, L.N., Stewart, A.L., Mullan, J., Phetlhu, R.D., Wantland, D., Cuca, Y., and Naidoo, J. (2009a). The development and validation of the HIV/AIDS Stigma Instrument-Nurse (HASI-N). AIDS Care, 21(2): 150?159.  Uys, L., Chirwa, M., Kohi, T., Greeff, M., Naidoo, J., Makoae, L., Diamini, P., Durrheim, K., Cuca, Y., & Holzemer, W. L. (2009b). Evaluation of a health setting-based stigma intervention in five African countries. AIDS patient care and STDs, 23(12), 1059-1066.  Uys, L. R., Minnaar, A., Reid, S., & Naidoo, J. R. (2004). The perceptions of nurses in a district health system in KwaZulu-Natal of their supervision, self-esteem and job satisfaction. Curationis, 27(2), 50-56.  Van Damme, W., Kober, K., & Laga, M. (2006). The real challenges for scaling up ART in sub-Saharan Africa. Aids, 20(5), 653-656.  Van Damme, W., Kober, K., & Kegels, G. (2008). Scaling-up antiretroviral treatment in Southern African countries with human resource shortage: How will health systems adapt?. Social Science & Medicine, 66(10), 2108-2121.  Van Rensburg, H. C. J. (Ed.). (2012). Health and health care in South Africa. Van Schaik Publishers. Pg: 400-409, 666  Van Rooyen, H., Richter, L., Coates, T. J., & Boettiger, M. (2009). Approaches to HIV Counselling and Testing: Strengths and Weaknesses, and Challenges for the Way Forward. In HIV/AIDS in South Africa 25 Years On (pp. 165-182). Springer New York. Visser, M., & Sipsma, H. (2013). The experience of HIV-related stigma in South Africa. In Stigma, Discrimination and Living with HIV/AIDS, 205-227. Springer Netherlands.  Wadee, H., & Khan, F. (2007). Human resources for health. South African health review, 141-149.  Weiss, M. G., Doongaji, D. R., Siddhartha, S., Wypij, D., Pathare, S., Bhatawdekar, M., Bhave A., Sheth A., & Fernandes, R. (1992). The Explanatory Model Interview Catalogue (EMIC). Contribution to cross-cultural research methods from a study of leprosy and mental health. The British Journal of Psychiatry, 160(6), 819-830.  Wolvaardt G. (2010). A review of doctors experiences over the first 10 years of Community Service. Pretoria: Foundation for Professional Development.   	 ? 117	 ?World Health Organization. (2006a). Treat, train, retain: the AIDS and health workforce plan. Report on the consultation on AIDS and human resources for health.  World Health Organization. (2006b). WHO Fact Sheet No 302: The Global Shortage of Health Workers and Impact.  World Health Organization (2008) World Health Statistics. Health Service Coverage. Available at: http://www.who.int/whosis/whostat/2008/en/  WHO (2009) WHO policy on TB infection control in health-care facilities, congregate settings and households. Geneva: World Health Organization, 11?15.   World Health Organization, International Labor Organization, UN AIDS (2010) The joint WHO ILO UNAIDS policy guidelines for improving health workers' access to HIV and TB prevention, treatment, care and support services, available at: http://www.who.int/occupational_health/publications/hiv_tb_guidelines/en/  Wouters, E., Van Damme, W., van Rensburg, D., & Meulemans, H. (2008). Impact of baseline health and community support on antiretroviral treatment outcomes in HIV patients in South Africa. Aids, 22(18), 2545-2548.  Yassi, A., Nophale, L. E., Dybka, L., Bryce, E., Kruger, W., & Spiegel, J. (2009). Building capacity to secure healthier and safer working conditions for healthcare workers: A South African-Canadian collaboration. International journal of occupational and environmental health, 15(4), 360-369.  Zuch, M., & Lurie, M. (2012). ?A virus and nothing else?: the effect of ART on HIV-related stigma in rural South Africa. AIDS and Behavior, 16(3), 564-570.  	 ? 118	 ?Appendix A: Occupational Health Survey 	 ? 119	 ?	 ? 120	 ?  	 ? 121	 ? 	 ? 122	 ? 	 ? 123	 ? 	 ? 124	 ? 	 ? 125	 ? 	 ? 126	 ?	 ? 127	 ?Appendix B: Sample Frame     	 ? 128	 ?Appendix C: Occupation Coding Based on Sampling Frame and PCHW vs. Non-PCHW  Survey Coding Sample Frame Coding PCHW/ Non-PCHW 1 - General Assistant Administrative Staff Non-PCHW 2 - Maintenance worker Support Services Staff Non-PCHW 3 - Porter Support Services Staff Non-PCHW 4 - Laundry worker Support Services Staff Non-PCHW 5 - Food Service Worker Support Services Staff Non-PCHW 6 - Security Support Services Staff Non-PCHW 7 - Clerk Administrative Staff Non-PCHW 8 - Registrar  MD PCHW 9 - Doctor MD PCHW 10 - Specialist MD PCHW 11 - Professional Nurse Nurse PCHW 12 - Assistant/Auxillary Nurse Nurse PCHW 13 - Staff Nurse Nurse PCHW 14 - Therapist AHP PCHW 15 - Technologist/Technician (IT department) Support Services Staff Non-PCHW 16 - Manager/Administrator (all nursing managers) Nurse PCHW Other (write-in responses) - Boiler Operator  - Cleaner  - Messenger  - Artisan Electrician  - Linen Collector  - Household Aid  - Tradesmen Aid   - Garden Service/Gardener  - Senior Messenger  - Technician  - Auxiliary Officer  - Porter  - Senior Operator  - Senior Housekeeper  - Senior Secretary  - Anchor Fitter  - General worker/refuse  - Painter  - Mechanical   - Ward Clerk  - Handyman  - Network Controller   - Senior Admin officer  - Supply Chain Practitioner  - SCM Practitioner  - Admin Officer   Support Services Staff Support Services Staff Support Services Staff Support Services Staff Support Services Staff Support Services Staff Support Services Staff Support Services Staff Support Services Staff Support Services Staff Administrative Staff Support Services Staff Support Services Staff Support Services Staff Administrative Staff Support Services Staff Support Services Staff Support Services Staff Support Services Staff Administrative Staff Support Services Staff Support Services Staff Administrative Staff Administrative Staff Administrative Staff Administrative Staff  Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW 	 ? 129	 ?- Human Resource Officer  - Statistician  - Informational Technology Specialist  - LAN Administrator  - Senior Procurement officer - Volunteer - Operational Manager ANC  - Chief Professional Nurse working as Case Manager  - Radiographer  - Medical Social Worker  - Case Manager Administrative Staff Support Services Staff Support Services Staff Support Services Staff Administrative Staff Support Services Staff Nurse Nurse AHP AHP Nurse Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW Non-PCHW PCHW PCHW PCHW PCHW PCHW       


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items