Accepted Manuscript The limits to agency in humanitarian settings: a qualitative study of adherence to antiretroviral therapy among refugees situated in Kenya and Malaysia Joshua B. Mendelsohn, Tim Rhodes, Paul Spiegel, Marian Schilperoord, John Wagacha Burton, Susheela Balasundaram, Chunting Wong, David A. Ross PII:
S0277-9536(14)00369-4
DOI:
10.1016/j.socscimed.2014.06.010
Reference:
SSM 9519
To appear in:
Social Science & Medicine
Received Date: 15 October 2013 Revised Date:
3 June 2014
Accepted Date: 9 June 2014
Please cite this article as: Mendelsohn, J.B, Rhodes, T., Spiegel, P., Schilperoord, M., Burton, J.W., Balasundaram, S., Wong, C., Ross, D.A, The limits to agency in humanitarian settings: a qualitative study of adherence to antiretroviral therapy among refugees situated in Kenya and Malaysia, Social Science & Medicine (2014), doi: 10.1016/j.socscimed.2014.06.010. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
RI PT
SSM-D-13-02591R2 The limits to agency in humanitarian settings: a qualitative study of adherence to antiretroviral therapy among refugees situated in Kenya and Malaysia AUTHORS, AFFILIATIONS AND ADDRESSES
Joshua B Mendelsohn1‡, Tim Rhodes2, Paul Spiegel3, Marian Schilperoord3, John Wagacha Burton4, Susheela Balasundaram5, Chunting Wong5, David A Ross1
MRC Tropical Epidemiology Group, Department of Infectious Disease Epidemiology, London School of Hygiene
SC
1
and Tropical Medicine, London, UK
Department of Public Health and Policy, London School of Hygiene and Tropical Medicine, London, UK
3
M AN U
2
Public Health and HIV Unit, United Nations High Commissioner for Refugees, Geneva, Switzerland
4
United Nations High Commissioner for Refugees, Nairobi, Kenya
5
United Nations High Commissioner for Refugees, Kuala Lumpur, Malaysia
‡Corresponding author
Joshua B Mendelsohn [email protected]
Paul Spiegel [email protected]
TE D
Tim Rhodes [email protected]
Marian Schilperoord [email protected] John Wagacha Burton [email protected]
Susheela Balasundaram [email protected]
EP
Chunting Wong [email protected]
David A Ross [email protected]
AC C
Total word count = 7,980 (limit 8000 including abstract, text, tables, references)
ACKNOWLEDGEMENTS
We are grateful to the study participants for their candid accounts. We thank UNHCR field offices in Kuala Lumpur, Nairobi, and Kakuma for providing critical logistical support. Dr. Christopher Lee, Dr. Anuradha Radhakrishnan, Chung Han Yang, Jayanthi Arumugam, Wong Kok Mun, Aung Zan Wai (Saw), Dr. Bosco Muhindo, Dr. Jonathan Imaana, Geoffrey Luttah, and Monica Eshikeda also provided essential support. We thank Jerry Manuel, Ngun Sui Sin, Juliana Ooi, Tan Sok Teng, Betty Akot, Abdikarim Sharif Mohamed, Habiba Abdi Adan, and Kevin Ekal Ekutan for conducting the in-depth interviews. Amrita Daftary provided valuable feedback on an earlier draft of the manuscript. We also thank the National Council for Science and Technology (Kenya) and the Economic Planning
ACCEPTED MANUSCRIPT
Unit, Prime Minister's Department (Malaysia) for approving this work. JBM acknowledges the support of the Canadian Institutes of Health Research (Priority Announcement for HIV/AIDS), the Parkes Foundation, the London
AC C
EP
TE D
M AN U
SC
RI PT
School of Hygiene and Tropical Medicine, and UNHCR for funding support that made this study possible.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
The limits to agency in humanitarian settings: a qualitative study of adherence to antiretroviral therapy among refugees situated in Kenya and Malaysia
1
ACCEPTED MANUSCRIPT Abstract HIV-positive refugees confront a variety of challenges in accessing and adhering to antiretroviral therapy (ART) and attaining durable viral suppression. However, there is little understanding of what these challenges are, how refugees navigate them, or how they may differ across humanitarian settings. We sought to document and examine accounts of the threats, barriers and facilitators experienced in relation to HIV treatment and care and
RI PT
to conduct comparisons across settings. We conducted semi-structured interviews among a purposive sample of 14 refugees attending a public, urban HIV clinic in Kuala Lumpur, Malaysia (July-September 2010), and 12 refugees attending a camp-based HIV clinic in Kakuma, Kenya (February-March 2011). We used framework
SC
methods and between-case comparison to analyze and interpret the data, identifying social and environmental factors that influenced adherence. The multiple issues that threatened adherence to antiretroviral therapy or precipitated actual adherence lapses clustered into three themes: “migration”, “insecurity”, and “resilience”. The
M AN U
migration theme included issues related to crossing borders and integrating into treatment systems upon arrival in a host country. Challenges related to crossing borders were reported in both settings, but threats pertaining to integration into, and navigation of, a new health system were exclusive to the Malaysian setting. The insecurity theme included food insecurity, which was most commonly reported in the Kenyan setting; health systems insecurity, reported in both settings; and emotional insecurity, which was most common in the Kenyan setting.
TE D
Resilient processes were reported in both settings. We drew on the concept of “bounded agency” to argue that, despite evidence of personal and community resilience, these processes were sometimes insufficient for overcoming social and environmental barriers to adherence. In general, interventions might aim to bolster
EP
individuals’ range of action with targeted support that bolsters resilient processes. Specific interventions are
AC C
needed to address locally-based food and health system insecurities.
Total word count=7,983 Abstract: 298
Keywords: Kenya; Malaysia; antiretroviral therapy; adherence; migration; forced displacement; refugee; qualitative analysis
2
ACCEPTED MANUSCRIPT INTRODUCTION Adherence to antiretroviral therapy (ART) is a critical, modifiable determinant of HIV treatment success that must be consistently excellent in order to achieve durable viral suppression (Lima et al., 2010; Lima et al., 2008). By definition, refugees have fled across an international border, obtaining an international legal
RI PT
status in the country of asylum (e.g., hosting country) that promises a level of healthcare similar to the standard received by host nationals (United Nations General Assembly, 1951, 1967). For HIV-positive refugees, ART and supportive services are typically by a mix of actors including humanitarian non-governmental organizations and national ART programs. There is now general agreement that refugees should receive ART when clinically
SC
indicated (The Sphere Project, 2004, 2011; UNHCR & Southern African HIV Clinicians Society, 2007).
However, hosting governments may argue that it is their prerogative to supply medications preferentially to
M AN U
nationals, in contravention of public health principles and international law, or that refugees are inherently unstable, and therefore incapable of adhering to treatment, despite evidence to the contrary (Mendelsohn et al., 2014; Mendelsohn et al., 2012b). Given the clear public health and clinical rationales for delivering ART to all in need, a better understanding of refugee experiences may help to identify critical social and environmental barriers to adherence, thereby helping to improve ART programs delivered by humanitarian actors (UNHCR,
TE D
2013; World Health Organization, 2013).
Adherence to medication has previously been conceptualized as a highly “individuated” behavior, situated in responsibility and action with individuals who take medication (Fisher et al., 2006; Weinstein, 1993). Alternative accounts have characterized adherence as a product of social condition and action, while
EP
emphasizing an interplay of factors exogenous to individuals that mediate individual-environment interactions and structure the specific contexts which constrain or enable individual action (Ewart, 1991; Rhodes, 2002).
AC C
These social models of HIV treatment and prevention have stressed an interplay of macro-level factors that operate in the legal, political, and economic domains, meso-level factors embedded in institutions, health systems, and social networks, and micro-level factors including interpersonal relationships and social interactions (Blankenship et al., 2000; Gupta et al., 2008; Rhodes et al., 2005; Wolfe et al., 2010). While recent discussions have specifically called for more focus on social influences in studies of adherence to ART (Castro, 2005; Krusi et al., 2010; Wolfe et al., 2010), few qualitative studies have examined the role of social and environmental factors in shaping adherence (Vervoort et al., 2007). Among the few examples of qualitative work conducted among conflict-affected groups, studies conducted in Uganda identified travel, insecurity in attending clinics, food insecurity, distance to health centers, and inadequate planning for the return phase of the 3
ACCEPTED MANUSCRIPT displacement cycle as barriers to adherence (Garang et al., 2009; Olupot-Olupot et al., 2008; Wilhelm-Solomon, 2009). Additionally, disruption of social networks and severe economic hardship during and after forced displacement, serve to highlight the role of social and environmental constraints on individual health (Rhodes, 2009). The influence of these constraints on outcomes may be viewed as the practical consequences of structural
RI PT
violence, or the “social arrangements that put individuals in harm’s way” (Farmer et al., 2006, p. e449).
Although social and environmental constraints may impede the attainment of health, this is not the end of the story. Resilience, a process by which individuals and communities harness “biological, psychosocial, structural,
SC
and cultural resources to sustain wellbeing” (Panter-Brick & Leckman, 2013, p. 333), may help individuals to overcome highly constraining structures. For refugees, this interplay occurs within a range of settings including
M AN U
camps, urban centers, peri-urban areas, remote locations, and dispersed settings, which are differentially impacted by local socioeconomic realities (Spiegel et al., 2010).
Given that an improved understanding of the experiences of clients in receiving HIV care in humanitarian settings could inform service improvements, we aimed to explore the opportunities and limitations afforded by the contexts in which ART was delivered by drawing on the experiences of refugees situated in two
TE D
contrasting asylum settings: a refugee camp located in a remote region of Kenya and an urban setting situated in the outskirts of urban Kuala Lumpur, Malaysia. Our comparative approach engaged with social, environmental, and operational differences in the provision of HIV treatment and care to refugees in urban and camp settings, and was based on purposive samples drawn from two larger surveys conducted by our research team. In these
EP
larger surveys, we aimed to recruit all refugees on ART for at least one month in each setting, achieving an 86% (73/85) participation rate in Kenya and a 90% (153/170) participation rate in Malaysia. The Malaysian survey
AC C
found that 81% (98/121) of refugees on treatment for at least 25 weeks were virologically suppressed, 74% achieved high levels of adherence (≥95%) to pharmacy prescription refills, and 72% self-reported ≥95% adherence over the previous month. In the Kenyan survey only 58% (34/59) of refugees on treatment for at least 25 weeks, were virologically suppressed, 85% had achieved high levels of adherence (≥95%) to pharmacy prescription refills, and 62% self-reported ≥95% adherence over the previous month (Mendelsohn et al., 2014; Mendelsohn et al., 2012a). In drawing on participant accounts from purposive samples, we sought to consider how social and environmental factors contributed to the contrasting treatment outcomes observed in these settings and to explain the helping and constraining effects of settings upon agency. To this end, we employ the concept of 4
ACCEPTED MANUSCRIPT “bounded agency”, whereby highly structured environments may limit the expression of individual agency (Evans, 2007, p.1), as a means for explaining how social and structural dissimilarities across treatment settings produce discordant treatment outcomes. We argue that relative differences in social and physical environments should be carefully considered in future HIV treatment and care programs situated in humanitarian settings.
RI PT
METHODS Research design and case selection
Refugee clients who were receiving ART were recruited from Sungai Buloh Hospital, an urban setting
SC
located on the outskirts of Kuala Lumpur (Malaysia) and the national reference hospital for infectious diseases; and the Comprehensive HIV Care Clinic in Kakuma, Kenya (CCCK) managed by the International Rescue
M AN U
Committee in the main hospital of Kakuma refugee camp, situated in a remote part of Northwestern Kenya. These settings varied in relation to range of attributes including geography, remoteness, and attributes of the treatment facility (Table 1). [***Table 1***]
TE D
Participants and sampling
Interview candidates were drawn from a sampling frame of refugees who had recently participated in a structured survey interview focused on adherence to ART. Participants were purposively sampled with respect to sex, refugee status, and self-reported adherence over the month prior to the survey interview. We employed a
EP
quota sampling strategy to recruit similar proportions of men and women, and refugees from different ethnic groups. “Refugees” were defined as individuals claiming a statutory designation as defined by the Refugee
AC C
Convention and Protocol (United Nations General Assembly, 1951). Eligible clients were at least 18 years of age, had been on regular ART for at least 30 days, and gave informed consent.
Case study settings
Kuala Lumpur, Malaysia At the start of the study (April 2010), over 91,985 total individuals had been registered by UNHCR as refugees and asylum seekers in Malaysia. Most had fled the protracted conflict in Myanmar fought over grievances related to autonomy and resources in outlying ethnic states (International Crisis Group, 2011). Among HIV-positive refugees registered at the clinic at study start, 54% (171/315) had started ART. The 5
ACCEPTED MANUSCRIPT Malaysian government is not a signatory to the Refugee Convention; however, the Ministry of Health issued a circular in 2006 that permitted refugees to access public health services, including the national ART program. Of refugees who had started ART, 98% were from Myanmar. For host nationals attending the clinic, the national treatment program fully subsidized first-line treatment and virological monitoring; however second-line treatment was only partially subsidized. For refugees, the national program fully subsidized first-line fixed-dose
RI PT
treatments but more expensive first and second-line drugs (e.g. efavirenz; lopinavir/ritonavir) and virological monitoring were supported by UNHCR; these medications were from an alternative private pharmacy. Financial assistance for travel to the clinic and pharmacy was provided to refugees by UNHCR on a case-by-case.
SC
Kakuma, Kenya
At the start of the study in Kakuma, Kenya (February 2011), the population of Kakuma Refugee Camp
M AN U
was 82,409 representing 18% of all refugees situated in Kenya. The camp’s population was made up of individuals from Somalia (54%), Sudan (32%), Ethiopia (8%) and the Democratic Republic of the Congo (5%). Among HIV-positive refugees registered at the clinic at study start, 26% (85/329) had started ART. The Kenyan government is a signatory to the Refugee Convention and Protocol, therefore access to first- and second-line ART was provided by the national program to all who met national clinical eligibility criteria. Members of the
TE D
local host community were primarily Turkana, a nomadic-pastoralist ethnic group although many were also sedentary in this setting. Immune status (CD4) testing, nutritional support and counseling services were routinely provided to all clients. If diagnosed with HIV infection, clients were routinely counseled and started on multivitamins and cotrimoxazole. When indicated by national guidelines (CD4
S0277-9536(14)00369-4
DOI:
10.1016/j.socscimed.2014.06.010
Reference:
SSM 9519
To appear in:
Social Science & Medicine
Received Date: 15 October 2013 Revised Date:
3 June 2014
Accepted Date: 9 June 2014
Please cite this article as: Mendelsohn, J.B, Rhodes, T., Spiegel, P., Schilperoord, M., Burton, J.W., Balasundaram, S., Wong, C., Ross, D.A, The limits to agency in humanitarian settings: a qualitative study of adherence to antiretroviral therapy among refugees situated in Kenya and Malaysia, Social Science & Medicine (2014), doi: 10.1016/j.socscimed.2014.06.010. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
RI PT
SSM-D-13-02591R2 The limits to agency in humanitarian settings: a qualitative study of adherence to antiretroviral therapy among refugees situated in Kenya and Malaysia AUTHORS, AFFILIATIONS AND ADDRESSES
Joshua B Mendelsohn1‡, Tim Rhodes2, Paul Spiegel3, Marian Schilperoord3, John Wagacha Burton4, Susheela Balasundaram5, Chunting Wong5, David A Ross1
MRC Tropical Epidemiology Group, Department of Infectious Disease Epidemiology, London School of Hygiene
SC
1
and Tropical Medicine, London, UK
Department of Public Health and Policy, London School of Hygiene and Tropical Medicine, London, UK
3
M AN U
2
Public Health and HIV Unit, United Nations High Commissioner for Refugees, Geneva, Switzerland
4
United Nations High Commissioner for Refugees, Nairobi, Kenya
5
United Nations High Commissioner for Refugees, Kuala Lumpur, Malaysia
‡Corresponding author
Joshua B Mendelsohn [email protected]
Paul Spiegel [email protected]
TE D
Tim Rhodes [email protected]
Marian Schilperoord [email protected] John Wagacha Burton [email protected]
Susheela Balasundaram [email protected]
EP
Chunting Wong [email protected]
David A Ross [email protected]
AC C
Total word count = 7,980 (limit 8000 including abstract, text, tables, references)
ACKNOWLEDGEMENTS
We are grateful to the study participants for their candid accounts. We thank UNHCR field offices in Kuala Lumpur, Nairobi, and Kakuma for providing critical logistical support. Dr. Christopher Lee, Dr. Anuradha Radhakrishnan, Chung Han Yang, Jayanthi Arumugam, Wong Kok Mun, Aung Zan Wai (Saw), Dr. Bosco Muhindo, Dr. Jonathan Imaana, Geoffrey Luttah, and Monica Eshikeda also provided essential support. We thank Jerry Manuel, Ngun Sui Sin, Juliana Ooi, Tan Sok Teng, Betty Akot, Abdikarim Sharif Mohamed, Habiba Abdi Adan, and Kevin Ekal Ekutan for conducting the in-depth interviews. Amrita Daftary provided valuable feedback on an earlier draft of the manuscript. We also thank the National Council for Science and Technology (Kenya) and the Economic Planning
ACCEPTED MANUSCRIPT
Unit, Prime Minister's Department (Malaysia) for approving this work. JBM acknowledges the support of the Canadian Institutes of Health Research (Priority Announcement for HIV/AIDS), the Parkes Foundation, the London
AC C
EP
TE D
M AN U
SC
RI PT
School of Hygiene and Tropical Medicine, and UNHCR for funding support that made this study possible.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
The limits to agency in humanitarian settings: a qualitative study of adherence to antiretroviral therapy among refugees situated in Kenya and Malaysia
1
ACCEPTED MANUSCRIPT Abstract HIV-positive refugees confront a variety of challenges in accessing and adhering to antiretroviral therapy (ART) and attaining durable viral suppression. However, there is little understanding of what these challenges are, how refugees navigate them, or how they may differ across humanitarian settings. We sought to document and examine accounts of the threats, barriers and facilitators experienced in relation to HIV treatment and care and
RI PT
to conduct comparisons across settings. We conducted semi-structured interviews among a purposive sample of 14 refugees attending a public, urban HIV clinic in Kuala Lumpur, Malaysia (July-September 2010), and 12 refugees attending a camp-based HIV clinic in Kakuma, Kenya (February-March 2011). We used framework
SC
methods and between-case comparison to analyze and interpret the data, identifying social and environmental factors that influenced adherence. The multiple issues that threatened adherence to antiretroviral therapy or precipitated actual adherence lapses clustered into three themes: “migration”, “insecurity”, and “resilience”. The
M AN U
migration theme included issues related to crossing borders and integrating into treatment systems upon arrival in a host country. Challenges related to crossing borders were reported in both settings, but threats pertaining to integration into, and navigation of, a new health system were exclusive to the Malaysian setting. The insecurity theme included food insecurity, which was most commonly reported in the Kenyan setting; health systems insecurity, reported in both settings; and emotional insecurity, which was most common in the Kenyan setting.
TE D
Resilient processes were reported in both settings. We drew on the concept of “bounded agency” to argue that, despite evidence of personal and community resilience, these processes were sometimes insufficient for overcoming social and environmental barriers to adherence. In general, interventions might aim to bolster
EP
individuals’ range of action with targeted support that bolsters resilient processes. Specific interventions are
AC C
needed to address locally-based food and health system insecurities.
Total word count=7,983 Abstract: 298
Keywords: Kenya; Malaysia; antiretroviral therapy; adherence; migration; forced displacement; refugee; qualitative analysis
2
ACCEPTED MANUSCRIPT INTRODUCTION Adherence to antiretroviral therapy (ART) is a critical, modifiable determinant of HIV treatment success that must be consistently excellent in order to achieve durable viral suppression (Lima et al., 2010; Lima et al., 2008). By definition, refugees have fled across an international border, obtaining an international legal
RI PT
status in the country of asylum (e.g., hosting country) that promises a level of healthcare similar to the standard received by host nationals (United Nations General Assembly, 1951, 1967). For HIV-positive refugees, ART and supportive services are typically by a mix of actors including humanitarian non-governmental organizations and national ART programs. There is now general agreement that refugees should receive ART when clinically
SC
indicated (The Sphere Project, 2004, 2011; UNHCR & Southern African HIV Clinicians Society, 2007).
However, hosting governments may argue that it is their prerogative to supply medications preferentially to
M AN U
nationals, in contravention of public health principles and international law, or that refugees are inherently unstable, and therefore incapable of adhering to treatment, despite evidence to the contrary (Mendelsohn et al., 2014; Mendelsohn et al., 2012b). Given the clear public health and clinical rationales for delivering ART to all in need, a better understanding of refugee experiences may help to identify critical social and environmental barriers to adherence, thereby helping to improve ART programs delivered by humanitarian actors (UNHCR,
TE D
2013; World Health Organization, 2013).
Adherence to medication has previously been conceptualized as a highly “individuated” behavior, situated in responsibility and action with individuals who take medication (Fisher et al., 2006; Weinstein, 1993). Alternative accounts have characterized adherence as a product of social condition and action, while
EP
emphasizing an interplay of factors exogenous to individuals that mediate individual-environment interactions and structure the specific contexts which constrain or enable individual action (Ewart, 1991; Rhodes, 2002).
AC C
These social models of HIV treatment and prevention have stressed an interplay of macro-level factors that operate in the legal, political, and economic domains, meso-level factors embedded in institutions, health systems, and social networks, and micro-level factors including interpersonal relationships and social interactions (Blankenship et al., 2000; Gupta et al., 2008; Rhodes et al., 2005; Wolfe et al., 2010). While recent discussions have specifically called for more focus on social influences in studies of adherence to ART (Castro, 2005; Krusi et al., 2010; Wolfe et al., 2010), few qualitative studies have examined the role of social and environmental factors in shaping adherence (Vervoort et al., 2007). Among the few examples of qualitative work conducted among conflict-affected groups, studies conducted in Uganda identified travel, insecurity in attending clinics, food insecurity, distance to health centers, and inadequate planning for the return phase of the 3
ACCEPTED MANUSCRIPT displacement cycle as barriers to adherence (Garang et al., 2009; Olupot-Olupot et al., 2008; Wilhelm-Solomon, 2009). Additionally, disruption of social networks and severe economic hardship during and after forced displacement, serve to highlight the role of social and environmental constraints on individual health (Rhodes, 2009). The influence of these constraints on outcomes may be viewed as the practical consequences of structural
RI PT
violence, or the “social arrangements that put individuals in harm’s way” (Farmer et al., 2006, p. e449).
Although social and environmental constraints may impede the attainment of health, this is not the end of the story. Resilience, a process by which individuals and communities harness “biological, psychosocial, structural,
SC
and cultural resources to sustain wellbeing” (Panter-Brick & Leckman, 2013, p. 333), may help individuals to overcome highly constraining structures. For refugees, this interplay occurs within a range of settings including
M AN U
camps, urban centers, peri-urban areas, remote locations, and dispersed settings, which are differentially impacted by local socioeconomic realities (Spiegel et al., 2010).
Given that an improved understanding of the experiences of clients in receiving HIV care in humanitarian settings could inform service improvements, we aimed to explore the opportunities and limitations afforded by the contexts in which ART was delivered by drawing on the experiences of refugees situated in two
TE D
contrasting asylum settings: a refugee camp located in a remote region of Kenya and an urban setting situated in the outskirts of urban Kuala Lumpur, Malaysia. Our comparative approach engaged with social, environmental, and operational differences in the provision of HIV treatment and care to refugees in urban and camp settings, and was based on purposive samples drawn from two larger surveys conducted by our research team. In these
EP
larger surveys, we aimed to recruit all refugees on ART for at least one month in each setting, achieving an 86% (73/85) participation rate in Kenya and a 90% (153/170) participation rate in Malaysia. The Malaysian survey
AC C
found that 81% (98/121) of refugees on treatment for at least 25 weeks were virologically suppressed, 74% achieved high levels of adherence (≥95%) to pharmacy prescription refills, and 72% self-reported ≥95% adherence over the previous month. In the Kenyan survey only 58% (34/59) of refugees on treatment for at least 25 weeks, were virologically suppressed, 85% had achieved high levels of adherence (≥95%) to pharmacy prescription refills, and 62% self-reported ≥95% adherence over the previous month (Mendelsohn et al., 2014; Mendelsohn et al., 2012a). In drawing on participant accounts from purposive samples, we sought to consider how social and environmental factors contributed to the contrasting treatment outcomes observed in these settings and to explain the helping and constraining effects of settings upon agency. To this end, we employ the concept of 4
ACCEPTED MANUSCRIPT “bounded agency”, whereby highly structured environments may limit the expression of individual agency (Evans, 2007, p.1), as a means for explaining how social and structural dissimilarities across treatment settings produce discordant treatment outcomes. We argue that relative differences in social and physical environments should be carefully considered in future HIV treatment and care programs situated in humanitarian settings.
RI PT
METHODS Research design and case selection
Refugee clients who were receiving ART were recruited from Sungai Buloh Hospital, an urban setting
SC
located on the outskirts of Kuala Lumpur (Malaysia) and the national reference hospital for infectious diseases; and the Comprehensive HIV Care Clinic in Kakuma, Kenya (CCCK) managed by the International Rescue
M AN U
Committee in the main hospital of Kakuma refugee camp, situated in a remote part of Northwestern Kenya. These settings varied in relation to range of attributes including geography, remoteness, and attributes of the treatment facility (Table 1). [***Table 1***]
TE D
Participants and sampling
Interview candidates were drawn from a sampling frame of refugees who had recently participated in a structured survey interview focused on adherence to ART. Participants were purposively sampled with respect to sex, refugee status, and self-reported adherence over the month prior to the survey interview. We employed a
EP
quota sampling strategy to recruit similar proportions of men and women, and refugees from different ethnic groups. “Refugees” were defined as individuals claiming a statutory designation as defined by the Refugee
AC C
Convention and Protocol (United Nations General Assembly, 1951). Eligible clients were at least 18 years of age, had been on regular ART for at least 30 days, and gave informed consent.
Case study settings
Kuala Lumpur, Malaysia At the start of the study (April 2010), over 91,985 total individuals had been registered by UNHCR as refugees and asylum seekers in Malaysia. Most had fled the protracted conflict in Myanmar fought over grievances related to autonomy and resources in outlying ethnic states (International Crisis Group, 2011). Among HIV-positive refugees registered at the clinic at study start, 54% (171/315) had started ART. The 5
ACCEPTED MANUSCRIPT Malaysian government is not a signatory to the Refugee Convention; however, the Ministry of Health issued a circular in 2006 that permitted refugees to access public health services, including the national ART program. Of refugees who had started ART, 98% were from Myanmar. For host nationals attending the clinic, the national treatment program fully subsidized first-line treatment and virological monitoring; however second-line treatment was only partially subsidized. For refugees, the national program fully subsidized first-line fixed-dose
RI PT
treatments but more expensive first and second-line drugs (e.g. efavirenz; lopinavir/ritonavir) and virological monitoring were supported by UNHCR; these medications were from an alternative private pharmacy. Financial assistance for travel to the clinic and pharmacy was provided to refugees by UNHCR on a case-by-case.
SC
Kakuma, Kenya
At the start of the study in Kakuma, Kenya (February 2011), the population of Kakuma Refugee Camp
M AN U
was 82,409 representing 18% of all refugees situated in Kenya. The camp’s population was made up of individuals from Somalia (54%), Sudan (32%), Ethiopia (8%) and the Democratic Republic of the Congo (5%). Among HIV-positive refugees registered at the clinic at study start, 26% (85/329) had started ART. The Kenyan government is a signatory to the Refugee Convention and Protocol, therefore access to first- and second-line ART was provided by the national program to all who met national clinical eligibility criteria. Members of the
TE D
local host community were primarily Turkana, a nomadic-pastoralist ethnic group although many were also sedentary in this setting. Immune status (CD4) testing, nutritional support and counseling services were routinely provided to all clients. If diagnosed with HIV infection, clients were routinely counseled and started on multivitamins and cotrimoxazole. When indicated by national guidelines (CD4
