Table 1.

Characteristics of the included studies

Author	Study location/community	Objectives	Study design/type	Sample size, participants and methods	Results
Banerjee et al., 1986[47]	District of Nadia West Bengal Santal	Rate and pattern of mental disorders	Cross-sectional study	205 families in urbanized tribal community by the method of door-to-door survey	Depression was common. Very low prevalence rate of Neurotic illness, epilepsy and mental retaliation. Married individuals were affected more than the unmarried ones. Males had a slightly higher rate to mental morbidity. The population showed a general tendency of greater vulnerability to mental illness with advancing age
Nandi et al., 1992[41]	West Bengal	Assess the change if any in the extent and pattern of mental morbidity in the urbanized group in comparison to the rural group of the same tribe	Cross-sectional study	Urban Santals (771)	Urbanization had little effect on the total mental morbidity. But stress-dependent disorders were more common in the urban tribe
	Santal			Rural Santals (653) Household survey
Ganguly et al., 1995[16]	Western Rajasthan Meena and other tribes	Understand the health issues related to the use of opium	Qualitative	Traditional opium users (200) from six villages, ethnographic information of opium use	Majority of the addicts were between 40 and 60 years of age. Consumption starts with 10 g/day. Some were hard-core users who consumed around 100 g a day or 250-300 g
Aparajitaet al., 1996[36]	Ganjam district, Orissa	Assessing social support network and the satisfaction of the children’s needs belonging to high and low sociocultural status families	Cross-sectional	Disadvantaged group (300) and advantaged group (150) and equal number from both genders from the 8th, 9th, and 10th grades were taken as samples	Children from advantaged socio-cultural environment were found to have health and enriching family climate, whereas children from socioculturally disadvantaged environment were deprived of getting necessary interpersonal and intra family support. In spite of getting negative support and responsibilities from their families, the need satisfaction rate was found to be more in disadvantaged children. Girls received more negative response from their family members than the boys. This paper confirmed the continuous positive social support in satisfying children’s needs in the Indian social system
	Mixed Population			Structured questionnaire (35 item questionnaire)
Chaturvedi et al., 2013[18]	Changlang district, Arunachal Pradesh	Assesses the types of substance use	Cross-sectional	Households (1092) respondents, age ≥10 years (5135)	Prevalence of opium use was more among males. Usage was higher in higher altitudes
	Tangsa, Singpho, Khamti, and Tutsa			Structured pretested questionnaire
Prabhakar and Manoharan, 2005[32]	Tamil Nadu Malayali and Lambadi	Evaluate health system to examine the health status of the target population	Cross-sectional	Villages (21), respondents (2785), Examine the health system from the perspective of the base hospital	Gender and age susceptibility patterns revealed specific age intervals for mental health disorders
Sushila et al., 2005[25]	Southern Rajasthan	Explore factors responsible for physical and mental discomfort; availability of health care facilities; preferred system to cure such problems	Qualitative	Households (156), Bhil, households of village Madri and tribal households from village Jamun	Services of traditional healers are used by the people in all kind of physical and mental discomforts
	Bhils			The perceptions of illness, socio-cultural beliefs, and practices regarding illness Mixed-methods study
Hackett et al., 2007[42]	Wayanad, Kerala	Examine association between CMD, anemia, malnutrition, and physical symptoms	Cross-sectional	Tribes (721) seeking treatment at Swami Vivekananda medical mission	CMD was not associated with anemia, malnutrition and physical symptoms
	NR			Quantitative data collection by interview and estimation of hemoglobin from blood samples
Giri et al., 2007[40]	Jharkhand	Study the sociodemographic and clinical profile of cases through a retrospective case record analysis of tribal populations and compare it with nontribals	Cross-sectional	All the patients registered (1752) in the three community outreach centers (Jonha, Khunti, Saraikella-kharsawan) from November 2005 to April 2006 were included in the study	About half of the cases from both groups are of age-group 20-39 years with gradual decline. Psychiatric morbidity among males was more than females in both tribals and nontribals. Patients with epilepsy were higher in tribal group. Tribals were more irregular with substantial number of dropouts
	Mixed population of tribals and non tribals			Sociodemographic profile and service utilization were recorded by reviewing the case records of the participants during that time
Sobhanjan and Mukhopadhyay, 2007[26]	Sikkim	Examine if perceived stress affects BP, lipids and obesity	Cross-sectional	Healthy volunteers (398) (age ≥20 years, urban males: 100; urban females (100); rural males (103) rural females (95)	Urban Bhutias experienced perceived stress to a significantly higher extent (mean±SD of PSSI value: male: 0.48±0.06, female: 0.48±0.07) than that of rural Bhutias (male: 0.22±0.07, females: 0.20±0.05)
	Bhutia			Structured questionnaire
Chowdhury et al. 2008[37]	Sundarban Delta Not mentioned	Examine the extent and impact of human-animal conflicts vis-a-vis psychosocial stressors and mental health of the affected people	Cross-sectional	3082 households (Satjelia, 1572, Lahiripur, 1512), were surveyed among the mixed population of tribals and nontribals Survey, FGD, IDIs, and medical clinics	During the last 15 years, 111 persons (male 83, female 28) became victims of animal attacks, viz., tiger (82%), crocodile (10.8%), and shark (7.2%), of which 73.9% died. In 94.5% cases, the conflict took place in and around the SRF during livelihood activities. Tracking of 66 widows, resulted from these conflicts, showed that majority of them (51.%) were either disabled or in a very poor health condition, 40.9% were in extreme economic stress and only 10.6% remarried. 1 widow committed suicide and 3 attempted suicide. A total of 178 persons (male 82, female 96) attended the community mental health clinics. Maximum cases were major depressive disorder (14.6%), followed by somatoform pain disorder (14.0%), posttraumatic stress disorder-animal attack related (9.6%), and adjustment disorder (9%). 11.2% cases had a history of deliberate self-harm attempt, of which 55% used pesticides
Tripathy et al. 2010[35]	Jharkhand and Odisha	Assess the effect of a participatory intervention with women’s groups on birth outcomes and maternal depression	Cluster randomized controlled trial	Intervention (2457) and control (2235). Women between the age group of 15 and 49 years	NMR was 32% lower in intervention clusters adjusted for clustering, stratification and baseline differences during the 3 year period and 45% lower in 2 and 3 years. No significant effect on maternal depression was noted
	Not mentioned			Primary outcome was to see the reduction in neonatal mortality rate and maternal depression scores after implementation of strategies to address the above-mentioned problems in the intervention arm compared to the control arm
Mohindra et al., 2011[19]	Waynad District, Kerala/	Understand the reasons, concerns, and consequences of consumption of alcohol	Qualitative	Households (393), age >15 years	Paniyas reported consumption of alcohol as a problem and is increasing among younger men
	Paniya			FGDs and semi-structured interviews	Reasons for consumption Easily available, produced illicitly in some colonies, employers using this as a strategy to attract Paniyas to work. The other reasons are range of socioeconomic consequences that are rooted in historic oppression and social discrimination
Sreeraj et al., 2012[20]	Ranchi, Jharkhand	Examine the reasons for alcohol intake, belief about addiction, their effect on the severity of addiction in people with different ethnic background	Cross-sectional	Tribal (40) and nontribal (20) communities	Patients from both the groups had a similar age of onset of substance intake, duration of intake in a dependence pattern, and duration of incubation from first intake to intake in dependence pattern. In spite of these similarities problems related to alcohol were more in tribals. Social enhancement, to cope with distressing emotions and peer pressure were some of the reasons for alcohol intake
	NR			Structured questionnaire through an interview
Yalsangi, 2012[34]	Trivandrum, Kerala Paniya Kattunaicken Bettakurumba Mullukurumba Irula	Assess the community health program run in a tribal area in Nilgiris	Cross-sectional- Program Evaluation	218, ST with an age of ≥25 and more were selected. No upper limit. Mixed-methods study	The intervention area had better awareness score (5.13±2.27) than that of control area (1.57±2.82)
Manimunda et al. 2012[23]	The Andaman and Nicobar Islands	Estimate the prevalence and determinants of tobacco use and nicotine dependency	Cross-sectional	18,018, both ST and non-ST population with an age group of ≥14 years	Prevalence of current tobacco use was 48.9%. Tobacco chewing alone was prevalent in 40.9% of the population. One-tenth (9%) of the males were nicotine dependent, while it was 3% in females. Three-fourths of the tobacco users initiated use of tobacco before reaching 21 years of age. Age, current use of alcohol, poor educational status, marital status, socioeconomic groups, and comorbidities were the main determinants of tobacco use and nicotine dependence
	Nicobarese tribe, Ranchi tribes			Structured questionnaire FTND test was used to estimate nicotine dependence
Diwan, 2012[28]	Ranchi, Jharkhand Not mentioned	Examine the main and interaction effects of ethnicity, marital status, and stress on mental health of tribal school teachers	Cross-sectional study	400, female school teachers of Ranchi town (160 tribal and 160 non tribal) GHQ-12, Stress scale was used to collect the data	Out of the three factors namely stress, marital status, and ethnicity, only ethnicity was found to produce main effect on mental health. Neither second-order interaction nor third-order interaction was found to be significant
Diwan., 2012[46]	Ranchi, Jharkhand	Know the impact of gender, socio-economic status, and age upon the mental health of tribal factory workers	Cross-sectional	400, tribal female workers from different factories	Out of the three factors namely gender, socio economic status, and age, gender was found to produce main effect on mental health
	Not mentioned			Personal data schedule, GHQ-12
Singh et al., 2013[24]	Roing and Anini districts, Arunachal Pradesh	Evaluate psychological traits of Idu Mishmi tribes to validate earlier report of high suicide rates	Cross-sectional, qualitative	218, unrelated school children aged 13-19 years, family members of unrelated individuals aged 19-85 years who had committed suicide	Suicide attempt was higher in Idu Mishmi population (14.22%) than urban population (0.4%-4.2%). Females were at higher risk. Depression (8.26%) was comparable with earlier reports, whereas anxiety syndrome (6.42%), alcohol abuse (36.24%), and eating disorder such as binge eating (6.42%) and bulimia nervosa (1.38%) were also recorded in the population
	Idu Mishmi			Data collection done using mixed-methods approach
Chaturvedi et al., 2013[18]	Arunachal Pradesh Tangsa, Singpho, Khamti and Tutsa	Estimate prevalence of opium use among tribes, association between sociodemographic factors and opium use	Cross-sectional	Age >15 years (3421), participated in substance survey, secondary data were used which were collected in a previous survey which assessed the types of substance use	Higher prevalence of opium use in men (10.6%) compared to women (2.1%). Opium use was significantly higher among Singpho and Khamti tribes. Variation seen according to age, educational level, occupation, marital status, and religion of the respondents
Raina et al., 2013[43]	Bharmour, Himachal Pradesh	Systematic methods for developing cognitive screening instrument for tribals	Cross-sectional	50, 60-75+ age groups, trained sample randomly picked	Modifications and testing of modified version of MMSE questionnaire resulted in an effective customized screening tool exclusive for Brahmouri population
	Gaddi			Different phases for development of MMSE questionnaire relevant for Bharmouri population
Longkumer et al., 2013[29]	Nagaland	Explore the existing knowledge and attitudes regarding mental disorders and see whether formal education has any relationship with their attitudes toward such disorders	Cross-sectional	Christian males (500) and (272) females in the age above 21 years	A great majority recognized mental health problem in the case vignette but used general terms such as psychosocial problem/mental problem/mental illness. Majority attributed the problem to psychosocial problems and chose a psychiatrist/psychologist over other options. However, a considerable number of participants reported evil spirit possession as the cause of mental disorders and preferred seeking for divine intervention as a treatment mode
	Nagas/Ao Naga Tribes			A brief instruction, respondent’s personal identification chart, a case vignette and a questionnaire based on the vignette
Raina et al., 2014[44]	Himachal Pradesh	Know the prevalence of dementia and to generate a hypothesis on the differential distribution across populations	Cross-sectional	2000, 60 years and above age, Two-phase study; screening and clinical phase	No case of dementia reported in tribal population
	Not mentioned			Screening - urban, rural, and migrant populations using HMSE questionnaire For tribal population modified version of MMSE was used Clinical evaluation- involved a psychiatrist and public health expert
Nizamie et al., 2015[31]	Jharkhand NR	Develop an effective health-care delivery model for epilepsy to reduce treatment gap in a rural community	Cross-sectional	114,068 6 health workers, traditional practitioners (267 faith healers and qualified practitioners, local practitioners) Involved training, awareness campaigns, diagnosis, treatment delivery, follow-up, and free medication	213 patients enrolled in a study completed 12 months treatment leaving 75% seizure free. The model was successful
Nimgaonkar and Menon, 2015[33]	Tamil Nadu Bettakurumba, Mullukurumba and other ST	Improve the health-care delivery through task shifting	Feasibility study	542, from 184 villages	Low cost task shifting was successfully implemented. Patients were well treated and they volunteered to increase the acceptance
Ozer, 2015[39]	Ladakh	To assess how two groups of Ladakhi college students navigate through different degrees of exposure to acculturation and how this affects their mental health	Mixed methods study (cross-sectional)	292 - quantitative and 12 - qualitative	Students with less acculturation exposure were more oriented toward ethnic culture and to a greater extent experienced impaired mental health when compared with the sample with more acculturation. Most prevalent among the students (34.2%) was a bicultural orientation, integrating both ethnic and mainstream culture. In general, acculturation orientation was not associated with quantitative measures of depression or anxiety. The qualitative analysis revealed agency and cultural identity to be pivotal factors in the process of reproducing culture and negotiating cultural change
	Mixed population			Structured data collection and IDIs
Jeffrey GS et al., 2016[38]	Central India	Understand human displacement’s mental health toll as well as the displacement-related changes that help explain such emotional suffering	Cross-sectional	Heads of the households (159) from Mazira and Behruda villages	Loss of homeland compromises mental health in all aspects
	Sahariyas			Ethnographic information and semi structured interview
Raina et al., 2016[45]	Himachal Pradesh Not mentioned	Explore the feasibility of using EASI as an alternative to HMSE and its modifications	Cross-sectional	60 years and above age (2000). Secondary data analysis	As the scores on EASI rise, the scores on HMSE fall both pointing to identification of the same clinical diagnosis, that is, dementia. EASI may be used as alternative to mental state examination
Lakhan et al., 2016[30]	Chikalia, Madhya Pradesh	Prevalence of Down’s syndrome in a tribal population and (2) its comorbidity with ID in tribal population		ST mothers (2767)	All mothers of all identified DS children were in young age (18-24 years) when they had babies with DS
	Not mentioned			Screening for ID (intellectual through a household survey). Identified cases evaluated by therapists in IDs for diagnosis
Janakiram et al., 2016[22]	Tribal colonies in Kalapetta, Kerala	Find out dependency of tobacco use in indigenous population of Waynad, India	Cross-sectional	103, individuals above age of 15 years in four colonies of Kalapetta	Prevalence of tobacco use in this population was 73.8%. Majority of them (92%) use smokeless forms of tobacco. The mean score for nicotine dependency was 3.85% for smoked tobacco and 4.61% was for smokeless tobacco which denote moderate dependency of tobacco use. Average age of onset of tobacco use was 16.41 years for smoked and 17.53 years for smokeless forms
	Adivasis			A structured questionnaire, modified and adapted from NIMHANS - the tobacco cessation questionnaire - was done
Ali et al., 2016[27]	Ranchi, Jharkhand	Find out the mental health status (emotional, hyperactivity, relationships and conduct problems and pro-social behavior’s) among school-going tribal adolescents	Cross-sectional study	Males (780) in the age range of 13-17 years going to school, belonging to tribal community	Out of the total participants, 5.12% of the students had emotional symptoms, 9.61% had conduct problems, 4.23% had hyperactivity, and 1.41% had significant peer problems
	Not mentioned			Semi structured sociodemographic data and strengths and difficulties questionnaire to assess the emotional and behavioral disorders were collected
Maulik et al., 2017[14]	Andhra Pradesh	Understand the feasibility and acceptability of mental health service utilization in Remote areas using mobile technology Evaluation of the SMART Mental Health project in rural India	Cross-sectional	Age >18 years (5007), participated in survey	Training was imparted to 21 ASHAs and 2 primary care doctors. 5007 of 5167 eligible individuals were screened, and 238 were
	Koya			Development of mobile technology-based EDSS	Identified as being positive for CMDs and referred to the primary care doctors for further management
				Interactive voice response system Stigma reduction campaign	Out of the screened positive, 2 (0.8%) had previously utilized mental health services. During the intervention period, 30 (12.6%) visited the primary care doctor for further diagnosis and treatment, as advised. There was a significant reduction in the depression and anxiety scores between start and end of the intervention among those who had screened positive at the beginning
				Baseline household survey	Stigma and mental health awareness in the broader community improved during the project Intervention Postintervention

CMD – Common mental disorder; BP – Blood pressure; SD – Standard deviation; PSSI – Permanent Shear Stability Index; SRF – Sundarbans reserve forest; IDIs – In-depth interviews; FGDs – Focus group discussions; FTND – Fagerström Test for Nicotine Dependence; GHQ – General Health Questionnaire; MMSE – Mini–Mental State Examination; HMSE – Hindi Mental State Examination; NR – Not reported; EASI – Everyday Abilities Scale for India; SMART – Systematic Medical Appraisal, Referral and Treatment; EDSS – Expanded Disability Status Scale; ASHA – Accredited social health activist; ST – Scheduled Tribes; ID – Intellectual Disability, DS – Down Syndrome, NMR – Neonatal Mortality Rate