Table 2.

Summary Table.

Authors, year, and country	Main aims	Method and data collection/analysis	Participants	Results	Stated conclusion	Limitation, critical appraisal	Reviewer’s conclusion
Bhandari and Chan 30 (2016), Nepal	Investigate caste/ethnicity-based inequity in women’s health service utilization, focusing on ANC in Nepal	Secondary cross sectional: Nepal Demographic Health Survey Data 2011, bivariate and multivariate analysis	4018 mothers aged 15 to 49 years who gave birth past 5 years	53% mother had ANC visit 4+ (mean 3.63, median 4.0), Hill Dalits and Terai Dalits were only 4%. Only 6% of disadvantages caste/ethnicity belongs to the wealthiest quintile.	Disadvantaged mothers using less ANC independently based on their caste/ethnicity and their household wealth. Advantaged mothers are also disadvantaged of utilizing ANC depending on wealth.	Secondary data, cross-sectional nature, not focus on quality visit. Appraisal: CASP, 2018 (75%, medium).	In-depth study association double discrimination: caste and wealth.
Kumar 31 (2007), India	Explore the link between SHGs + women’s access to health services	Mixed method: survey, interviews, case studies, and focus group discussions	SHGs women (n = 200), family members, community leaders	84% SC used unlicensed “private doctors,” paid high charges. No change reported in health, health knowledge, health utilization, spending on food, basic needs compared with OBC. Participations’ health impact was reportedly greater for OBC women than SC.	Caste and class powerful in determining women’s access to health. Dependent on gender relations, income, education, and general standards of living. SHGs fail to overcome structural contexts hence failed to produce equitable health services to marginalized.	Small sample, number of participants other than SHGs women is not made clear. Appraisal: McGill, 2018 (85%, medium).	Women are discriminated on gender, caste, and class, information on double discrimination would be helpful.
Polit 35 (2005), India	To show how the relative marginality of Dalits affects the well-being of Dalit women	Mixed method: Survey and ethnographic research	Garhwali (state) women central Himalayas of north India	First village: Dalits = poor, literate, high discrimination, and dependency. Second had 2 areas, occasional clashes, moderate dependency, employed, and educated people. Third village: only Dalits, less land, education, and jobs; no discrimination and dependency.	People’s affects well-being more than trans location. Dalit living exclusively Dalit village not consider themselves marginal and well-being = greater. Dalit in village with a high-caste majority will feel more marginal therefore, well-being likely to be less.	Small sample, limited information on methods, and data collection. Appraisal: McGill, 2018 (80%, medium).	No clear identified result. Dalits not marginal in all villages and level of discrimination, access to health and well-being are very different.
Priya and Sathyamala 36 (2007), India	To explore level of ill health of people from low castes, capacities to respond to adult illness, and support needed	Cross-sectional Mixed methods: survey and interviews	1171 household Uttar Pradesh (UP) + 900 Tamil Nadu (TN) from SC, interviews = 62 in UP + 52 in TN	Two regions had distinctive health vulnerabilities and support systems. Death rates UP (9.4) and TN (11.4) not as expected. UP 19%/94% had treatment and in TN 97% with long-term illness had some treatment. Sources of treatment were loans. Stigma long-term illness not problem.	People who are poor and lower castes are not equally susceptible to HIV. Social cohesion provided security from impact of poor living and working conditions. Traditional forms of social cohesion are under stress and new forms are inadequate. No social norms protect women.	Methods and selection of participant not clearly explained. Appraisal: McGill, 2018 (85%, medium).	Discrimination on long-term illness is not a major issue; however, fear of stigma led to preventable death.
Mohindra et al 33 (2006), India	Examine social patterning of women’s self-reported health status: Kerala; 2 hypotheses: (1) low caste and socioeconomic position is associated with worse health status, (2) associations between socioeconomic position and health vary across castes	Secondary cross-sectional data: household survey implemented by the Centre for Development Studies in 2003. Multilevel multinomial logistic regression model.	4196 non-elderly women of marital age (18-59 years).	Lower caste women, more likely never attended school and are predominantly wage laborers, OBC are slightly more likely to work as wage laborers and forward caste engage in nonwage activities. Odd rations poor perceived health and ADL.	Caste and socioeconomic interrelated; lower caste magnifies health inequity. Being both lower caste and poor can trap people into poor health than either inequality on its own. Implementing interventions that deal with caste and socioeconomic disparities to produce more equitable results than targeting either inequality in isolation.	Cross-sectional study, multilevel multinomial modes in ADL, self-perceived health, regular contact with professionals and attitudes and perceptions. Appraisal: CASP, 2018 (85%, medium).	Information on effects of gender inequality of women’s self-reported health status would be helpful.
George 38 (2015), India	Examine Dalits in significant positions of rural health + improvement provisioning of health services in tribal India	Secondary analysis: National Sample Survey Office (un)employment (2011-2012)	National survey	Dalits are underrepresented in health professionals. Despite only 24% of rural population, other castes shares 40% in health professional. Shortages subcenters, PHCs, and CHCs.	Underrepresentation of Dalits in rural health care delivery due to untouchability. Indian health system is not equipped to address exclusion, which for urgent policy attention.	Secondary analysis of health work force. Appraisal: CASP, 2018 (80%, medium).	Further study on why in rural India significant jobs are likely to be taken by higher castes.
Daniel et al 37 (2012), Nepal	Examine health care access to Dalits through experiences of stakeholders throughout health system	Ethnography, participatory approach: KI interviews and FGD, stakeholder, and institutional analysis	19 FGD and 19 KI totals (n = 209)	Dimension: info access, physical access, financial access, discrimination, and social capital restricting access to health services identified 5 themes: human rights education, health education, advocacy, public inclusion, and dialogue.	Dalits and non-Dalits less access to health services due to lack of resources, absence of monitoring health care, and problems decentralization also main causes of weak Nepali health system.	Less information: district due to language and geography—social relations for field visits. Appraisal: CASP, 2018 (70%, medium).	A table reflecting KI and FGD would be helpful in better understanding of results.
Verma and Acharya 34 (2018), India	Explore health interaction of Dalit health staffs with non-Dalit care seekers and vice versa	Qualitative—in-depth interviews and 4 FGD, systematic random sampling; thematic analysis	20 ANMs, 20 ASHAs + 80 care seekers who delivered babies in last 6 months	5 Themes: Caste, perception and social identity, profession and social identity, maintaining identity, conflict and dilemmas, control and autonomy. Variation across caste of providers and seekers in shaping perception of each other.	Dalit providers lacked skills and health seekers are suspicious of their knowledge. Other staff limited interaction with Dalit care seekers and staff. Women faced gender and caste. Provider and seekers’ caste more weight than profession and need.	Small number, no justification how themes were identified, no written consent. Appraisal: CASP, 2018 (90%, high).	Clarification on how FGDs were conducted and further information on double discrimination would be helpful.
Rao 32 (2015), India	Discuss key conceptual ideas of agency, voice, and interjectionally in relation to the role of marriage and sexuality in reinforcing caste and gender boundaries	Mixed method: Survey, In-depth interview, FGDs, and KI, narrative	Rural couples: 400 surveys, 40 in-depth interviews	Choice of marriage partner—arranged marriage among OBC, contestations among Dalits degree of control. Facing violence, resisting it, narrative by all women. Jobs are not easy for women, enhances dependence on their men.	More understanding needed on Dalit women’s “acceptance” of violence. Inseparability/lack of agency, of action/patience, as strategies to challenge hierarchy and strengthen their bargaining position.	Narrative research. Appraisal: McGill, 2018 (80%, medium).	More clear results, details related to data analysis, dowry-related violence/marriage.

Abbreviations: ANC, antenatal care; CASP, Critical Appraisal Skills Program; SHG, self-help groups; SC, scheduled castes; OBC, other backward classes; HIV, human immunodeficiency virus; ADL, activities in daily living; PHC, primary health center; CHC, community health center; FDG, focus group discussion; KI, key informant; ANM, auxiliary nurse-midwife; ASHA, accredited social health activist.