A Mixed Methods Policy Analysis and Epidemiology Study of Caste and Psychological Morbidity in Nepal
Total Page:16
File Type:pdf, Size:1020Kb
Brandon A. Kohrt Vulnerable social groups in postconflict settings: a mixed methods policy analysis and epidemiology study of caste and psychological morbidity in Nepal Brandon A. Kohrt Designing and implementing psychosocial interven- Keywords: policyanalysis, mixed methods, tion programmes in post con£ict settings requires a caste, Nepal breadth of knowledge of the context, circumstances, and needs of vulnerable social groups. However, Introduction One of the major challenges facing psycho- mixed methods research focusing on which groups social interventions in settings of political are vulnerable, and their speci¢c psychosocial needs, violence is addressing the needs of vulner- is rarely conducted. This study uses historical able populations. Con£ict and other forms policy discourse analysis to identify the origins of of political violence do not a¡ect all persons contemporary social categories related to vulner- equally. Therefore, psychosocial program- ability in Nepal, speci¢cally caste. The policy mes typically target vulnerable groups.This analysis is employed to interpret cross-sectional is exempli¢ed by the focus on groups such epidemiological¢ndingsfrom asampleof 316 adults. as children and women. Interventionists Analyses test the relationship between caste and often also consider ethnic minorities as psychological morbidity, assessed with the General vulnerable populations. For example, action Health Questionnaire (GHQ-12), including poten- sheets 2.1, 4.1, 5.3, 7.1, and 10.1 of the Inter- tial mediators. Low caste (Dalit/Nepali) groups Agency Standing Committee (IASC) Guidelines were found to have a 53,3 percent prevalence of reference addressing vulnerability associ- psychological morbidity compared with 28,2 percent ated with ethnicity (IASC, 2007). The goals of the analyses presented here are to develop prevalence among other caste and ethnic groups recommendations to address the mental (odds ratio 2,91, 95% con¢dence interval 1,71^ health and psychosocial needs of margina- 4,96). Income and stressful life events partially med- lised groups a¡ected by the Maoist People’s iate the relationship between caste and psychological War in Nepal. This study draws upon a morbidity.These ¢ndings are interpreted in relation mixed methods approach, employing policy to themes from the policy analysis including restric- analysis and epidemiology. This study is the tions in social interactions, access to resources, social ¢rst to combine policy analysis and epide- control and punishment, social mobility and gender miology to examine caste in Nepal. relations. The study concludes with recommen- There are three challenges to working dations for addressing the needs of vulnerable social with vulnerable social groups in postcon£ict groups in post con£ict settings. settings: 239 Copyright © War Trauma Foundation. Unauthorized reproduction of this article is prohibited. Vulnerable social groups in post-conflict settings: a mixed methods policy analysis Intervention 2009, Volume 7, Number 3, Page 239 - 264 1. How are social groups categorised and by Among aid organizations, it has been whom? When designing interventions in common to prioritise psychosocial post con£ict settings, the classi¢cation services for child soldiers. However, in of individuals into di¡erent ‘vulnerable’ some psychosocial functioning arenas, groupsbasedonreligion,ethnicity,region child soldiers may not have greater and other a⁄liations is rarely straight- de¢cits than children never associated forward. Social categories are highly with armed groups (Kohrt, Jordans, political and often in £ux (Anderson, Tol, Speckman, Maharjan, Worthman 1991). The categorisation used by people & Komproe, 2008). in power may be very di¡erent from 3. What are the underlying processes that make the social grouping described by those a certain socialgroup more vulnerableto psycho- marginalised from political power social problems? Another challenge in (Kleinman,1995). For example, in Nepal, working with vulnerable groups is deter- while government bureaucrats may refer mining the underlying risk factors. to a certain community as Hindu, the Group membership is rarely the causal group may self-identify as Buddhist or factor for mental health andpsychosocial animist, or vice versa (Whelpton, 2005). problems. More typically, it is the di¡er- Moreover, group categorisation is related ential exposure to risk factors associated more to social constructs than genetic with group membership. Being part of a di¡erences (Braun, Fausto-Sterling, speci¢c group may predispose members Fullwiley, Hammonds, Nelson, Quivers, to certain experiences and forms of Reverby & Shields, 2007; Brown and deprivation and/or exclusion in politics, Armelagos, 2001; Hacking,1999), education, ¢nancial transactions, legal 2. What relationship do social group categories proceedings, and access to health- have with psychosocial vulnerability? Vulner- care (Ahmed, Mohammed &Williams, able social groups often bear a greater 2007). This has important implications burden of psychosocial problems for intervention because the goal of (Desjarlais, Eisenberg, Good & Klein- psychosocial programming would not man, 1995). However, ethnic minorities necessarily be to change group member- do not universally have poorer psycho- ship but rather to promote economic, social wellbeing or mental health com- educational, and healthcare equity pared with dominant social groups. For across groups. example, in the United States, African Americans have a lower prevalence of Because of these challenges, it is argued in depression than European Americans this paper that interventionists need to (Williams, Gonzalez, Neighbors, Nesse, address these three issues when considering Abelson, Sweetman & Jackson, 2007). psychosocial care for vulnerable groups in Mexican Americans with schizophrenia post con£ict settings. First, the social cat- have a lower relapse rate than European egorisation process of groups needs to be Americans (Karno, Jenkins, de la Selva, explored rather than taking standard classi- Santana, Telles, Lopez & Mintz, 1987; ¢cations at face value. This is because Lopez, Hipke, Polo, Jenkins, Karno & group divisions are more likely to re£ect Vaughn, 2004). This does not only apply labels of those in power. Qualitative re- to categorisation of ethnic groups. search approaches are well suited for this 240 Copyright © War Trauma Foundation. Unauthorized reproduction of this article is prohibited. Brandon A. Kohrt endeavour. Second, epidemiological studies provided by research participants.Typically, are necessary to demonstrate an associ- researchers work with the displaced, the ation of membership in a vulnerable group traumatised and the marginalised ^ all with poorer mental health and psychosocial groups disenfranchised from positions of wellbeing. Programmes should not rely upon power. Therefore, the research does not assumptions that psychosocial problems are reveal the processes and actions of those in inherent in marginalised groups. Third, the positions of power whose behaviours and underlying factors that explains the associ- policies perpetuate the marginalisation of ation of group membership and psychosocial others. However, policy development does problems needs to be assessed. Identi¢cation re£ect these processes. In addition, policies of the underlying factors will lead to prio- contribute to the marginalisation of certain ritisation of the most e¡ective targets for populations, which then makes them vulner- intervention. able to exploitation by armed groups. To address these three challenges, this study Finally, policy analysis is important to con- is an exploration of caste, a category central sider for psychosocial practitioners because to religious identity, life trajectories, and policies themselves can be powerful sites of socio/economic/political relations in South intervention. Interventionists increasingly Asia. Scholars, politicians, interventionists, focus on promoting psychosocial programs and social scientists have viewed caste as an that contribute to sustainable changes. axis of vulnerability and marginalisation Investing in policy change can in£uence (Bob, 2007; Kisan, 2005; Kohrt, Kunz, long term social processes that will reduce Baldwin, Koirala, Sharma & Napal, 2009). the vulnerability of marginalised groups. Using policy analysis, major themes related Policy analysis of vulnerable groups to marginalisation, which can then be tested There is a tremendous array of qualitative using epidemiological models, are indenti- approaches from which to select methods. ¢ed. An approach that has received less attention is policy analysis, a form of critical discourse Epidemiological analyses of vulnerable groups analysis (Wodak & Meyer, 2001; Yanow, There are a range of analytical approaches 2000). Discourse analysis draws upon pro- to assess the relationship of vulnerable duced materials such as literature, religious groups with psychological morbidity and texts, artwork, and historical artefacts. This other mental health and psychosocial out- elucidates historical processes for which liv- comes. Figure 1 illustrates three di¡erent ing key informants are not available. This is types of associations. Standard crude associ- particularly useful to identify the historical ations (A) simply test whether there is an origins of current social problems. Policy association between group membership discourse analysis examines texts related to and psychological morbidity. This is done governance and public practices