Serious mental health conditions and exposure to adulthood trauma in low- and middle-income countries: a scoping review Authors: Anne Stevenson, Engida Girma, Benon Kabale Kitafuna, Boniface Harerimana, Karestan C. Koenen, Soraya Seedat Published: 2024 Impact statement The impact of this scoping review, a type of analysis that looks at all the research on a specific topic, shows much of our data on adults with preexisting serious mental health conditions (SMHCs) who are then exposed to trauma in low- and middle-income countries (LMICs) come from a small subset of countries and certain types of interpersonal violence. History of the topic: Trauma exposure and SMHCs, such as schizophrenia and bipolar disorder, have been shown to be interconnected. Most of the research in this area has focused on childhood trauma as a risk factor for developing an SMHC, and the bulk of this work has come from studies in high-income countries in North America and Western Europe. While it is often stated that people with SMHCs are more vulnerable to experiencing traumatic events than the general public, there is less research on having an SMHC as a risk factor for exposure to traumatic events, and particularly on populations in LMICs, which have been underrepresented in research to date. This scoping review aims to fill this gap and to synthesize what trauma types adults with SMHCs face in LMICs. Main takeaways: After a thorough search, we found that studies on this topic came from only 10 countries (out of more than 130 LMICs) and two-thirds of these were from India, Ethiopia and China. The vast majority of these studies were about interpersonal violence, such as sexual violence, intimate partner violence and physical violence, and more than one-fifth of these were on physical restraint, such as chaining. There were no studies on man-made or natural disasters. The findings support that adulthood trauma exposure among people with SMHCs is an existing issue in LMICs and is a call to action to eradicate violent practices in both community and healthcare settings. Introduction There is an established link between trauma exposure and serious mental health conditions (SMHCs), such as schizophrenia, schizoaffective disorder and bipolar disorder (Grubaugh et al., 2011; Mauritz et al., 2013). Many studies have shown that trauma exposure, often defined as exposure to actual or threatened death, serious injury or sexual violence (American Psychiatric Association, 2013), poses a significant risk for the development of SMHCs (Read et al., 2005; Varese et al., 2012; Woolway et al., 2022). Most of the research on trauma and SMHCs has focused on childhood adversity and in samples from high-income countries in Europe, North America and Australasia (Read et al., 2005; Varese et al., 2012; Woolway et al., 2022). Multiple models have been proposed to explain why exposure to trauma, and particularly childhood adversity, may increase the risk for developing an SMHC, including the stress-vulnerability model (Zubin and Spring, 1977). This model posits that some people have a biological vulnerability to SMHCs. While individuals may be able to withstand a certain amount of stress, once a certain threshold of stress is reached or a stressor is particularly intense, people become more susceptible to developing SMHCs. Likewise, researchers have also argued that people with SMHCs may be more vulnerable to exposure to trauma than those without such conditions. Population studies have found people with SMHCs report more victimization than the general public (Maniglio, 2009; de Vries et al., 2019). One theory for this is that people with SMHCs may have cognitive impairments that impact their decision-making and problem-solving abilities, restricting their ability to navigate potentially dangerous situations (de Vries et al., 2019). People with SMHCs may also be exposed to trauma through settings due to their mental health conditions, such as inpatient and outpatient healthcare centers. Patients with SMHCs have reported high rates of trauma exposure in psychiatric settings, including physical and sexual assault by staff, police and other patients (Frueh et al., 2005; Lundberg et al., 2012a). Like the childhood adversity literature, the research on having an SMHC as a risk factor for trauma exposure in adulthood has also primarily been from samples in high-income countries (Maniglio, 2009; de Vries et al., 2019). Deepening research in LMICs may shed light on the possibility of unique traumas in this population. Furthermore, identifying and codifying trauma exposure in this population may raise awareness about such incidents and lead to healthcare and human rights policy changes for people with SMHCs (Guan et al., 2015; Hidayat et al., 2023). This scoping review aims to fill this gap by synthesizing the literature on trauma that adults with a diagnosis of schizophrenia, bipolar disorder, schizoaffective disorder, delusional disorder or depression with psychotic features are exposed to in LMICs, which have been less represented in research to date. We framed this exploratory study around Peters et al.’s population, concept, context framework for scoping reviews (Peters et al., 2021). Our primary research question aimed to establish the evidence base on adult trauma exposure experienced by people with SMHCs in LMICs. In other words, what is the extent, range, and nature of research about adults with SMHCs (population) being exposed to a traumatic event in adulthood (concept) in an LMIC (context)? Our secondary research question was to ascertain whether there were any trends in the data; for example, whether there were trauma types or demographics that were more commonly studied than others. We expected most trauma exposures would be types of interpersonal violence (use of physical, sexual or psychological force against another person or small group of people) (Mercy et al., 2017) and human rights violations, such as solitary confinement for multiple years. Methods We followed the JBI updated guidelines for the conduct of scoping reviews (Peters et al., 2021). As part of the review, we developed an extensive search query, retrieved all articles, conducted multiple phases of screening, extracted the data and synthesized the results. We outline each step below. Search strategy: We developed our search strategy to run in five databases: PubMed, Embase, PsycINFO, Web of Science Core Collection and Africa-Wide Information/NiPad. We started with PubMed and used a combination of medical subject headings (MeSH), some of which were “exploded” to retrieve any citations that fell under its subheadings, title and abstract (tiab) fields, and keyword (kw) fields. We used Boolean operators to develop a search query based on more than 300 terms and phrases connected to SMHCs, traumatic events and LMICs. For SMHCs, we included words that were often associated with this term, such as “schizophrenia,” “serious mental illness” and “psychotic disorder.” For trauma exposure, we pulled terms from two commonly used measures that assess potentially traumatic events: the Life Events Checklist for the Diagnostic and Statistical Manual of Mental Disorders (DSM), Fifth Edition (LEC-5) (Weathers et al., 2013) and the posttraumatic stress disorder module of the World Mental Health Survey version of the World Health Organization Composite International Diagnostic Interview (CIDI) (Kessler and Üstün, 2004), which is based on the International Classification of Diseases (ICD). We then expanded the terms from the LEC-5 and the CIDI to include specific examples of the type of trauma in question. For example, for “natural disasters,” we listed this term and also listed typhoons, landslides, tsunamis, etc. A priori, on the basis of qualitative studies with people with SMHCs in LMICs and commentaries by leaders in the field (Alem, 2000; Ametaj et al., 2021), we included events that may be more common in LMICs than in high-income countries, although they can happen anywhere, including human rights violations, mob justice, chaining and restraining. Additionally, we included the term “idiom(s) of distress” to potentially retrieve articles that might have focused on expressions of trauma not typically captured in “Western” terminology. For LMICs, we included the name of every country that was classified as an LMIC by the World Bank as of November 2023 (World Bank, 2023) (i.e., Afghanistan) as well as unrecognized countries or disputed territories that were not on the World Bank list but still met criteria as an LMIC (i.e., Somaliland). We also added overarching categories that are sometimes used to describe these groupings, such as “South Asia,” “developing country,” and “Global South.” We then adapted the search string to Embase, PsycINFO, Web of Science, and Africa-Wide Information/NiPad using their controlled subject vocabulary. See Supplemental File 1 for the full search strings for each database. Inclusion and exclusion criteria: Our inclusion and exclusion criteria were as follows: Inclusion:Adults aged ≥18 years.People with a diagnosis of schizophrenia, bipolar disorder, schizoaffective disorder, delusional disorder or depression with psychotic features.Trauma exposure must have occurred after the psychiatric diagnosis.Study population was in an LMIC.Observational studies only.Empirical studies in peer-reviewed journals, including case studies, qualitative studies and quantitative studies.There were no restrictions on the language in which the article was written.There were no restrictions on the date the article was published. Adults aged ≥18 years. Adults aged ≥18 years. People with a diagnosis of schizophrenia, bipolar disorder, schizoaffective disorder, delusional disorder or depression with psychotic features. People with a diagnosis of schizophrenia, bipolar disorder, schizoaffective disorder, delusional disorder or depression with psychotic features. Trauma exposure must have occurred after the psychiatric diagnosis. Trauma exposure must have occurred after the psychiatric diagnosis. Study population was in an LMIC. Study population was in an LMIC. Observational studies only. Observational studies only. Empirical studies in peer-reviewed journals, including case studies, qualitative studies and quantitative studies. Empirical studies in peer-reviewed journals, including case studies, qualitative studies and quantitative studies. There were no restrictions on the language in which the article was written. There were no restrictions on the language in which the article was written. There were no restrictions on the date the article was published. There were no restrictions on the date the article was published. Exclusion:Studies with data collected in high-income countries.Studies where it was not clear when the trauma occurred and the diagnosis of the mental health condition occurred.Studies where the trauma only occurred in childhood, adolescence or youth.Children (people aged <18 years).Intervention studies (i.e., randomized control trials).Conference papers and abstracts, commentaries, letters, news articles, scoping reviews, systematic reviews, meta-analyses and gray literature.Biological process studies such as biomarkers, genetic studies, biospecimens and circuitry.Animal studies. Studies with data collected in high-income countries. Studies with data collected in high-income countries. Studies where it was not clear when the trauma occurred and the diagnosis of the mental health condition occurred. Studies where it was not clear when the trauma occurred and the diagnosis of the mental health condition occurred. Studies where the trauma only occurred in childhood, adolescence or youth. Studies where the trauma only occurred in childhood, adolescence or youth. Children (people aged <18 years). Children (people aged <18 years). Intervention studies (i.e., randomized control trials). Intervention studies (i.e., randomized control trials). Conference papers and abstracts, commentaries, letters, news articles, scoping reviews, systematic reviews, meta-analyses and gray literature. Conference papers and abstracts, commentaries, letters, news articles, scoping reviews, systematic reviews, meta-analyses and gray literature. Biological process studies such as biomarkers, genetic studies, biospecimens and circuitry. Biological process studies such as biomarkers, genetic studies, biospecimens and circuitry. Animal studies. Animal studies. There were a few caveats to these criteria. If we could not distinguish whether an article included participants with depression vs. depression with psychotic features, we kept the article in. If an article was about “psychiatric conditions,” but we could not split out data from the disorders that met our eligibility criteria (i.e., schizophrenia) from the overall grouping, we excluded the article. If an article included participants under 17.9 years and we could separate out data for participants ≥18, we included the article and only reported on data pertaining to adults. If we could not split out participants under 17.9 from ≥18 (for example, if the age range was 15–49), the article met all our other eligibility criteria and the mean age of participants was ≥18, we included the article. Reviewing retrieved articles We ran our search on December 20, 2023, downloaded the citations from the retrieved articles and imported them into Covidence (www.covidence.org [Veritas Health Innovation Ltd, 2024]). Following the removal of duplicate articles in Covidence, we followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) (Tricco et al., 2018) process and created a flowchart to capture each stage (see Figure 1). We screened the remaining articles in two stages. First, two reviewers (AS and EG) independently screened the titles and abstracts for eligibility according to the inclusion/exclusion criteria. The same reviewers then independently screened the full text of the remaining articles to assess eligibility. At both stages, for articles that were in conflict, the two reviewers discussed the articles and came to a consensus on whether they met eligibility. Following confirmation that an article met the criteria, we hand-searched its reference list to identify additional articles that should be reviewed (Dundar and Fleeman, 2017). We then moved all articles into the data extraction pool. We created a Google Sheet to extract a range of fields from each article: article information (i.e., first author and title), study design, sample size, sex/gender of participants (% female), mean age, racial/ethnic composition, participant psychiatric diagnosis, method of assessment for the psychiatric condition, type of trauma exposure, tool used to assess trauma exposure, and country and municipality where the participants were recruited. AS extracted data for each article, and another reviewer independently extracted data from the same article. They then cross-checked each other’s work to ensure accuracy. We held group discussions when there were divergent decisions and came to a consensus on the final answer. Search procedure validation In order to validate our search procedure and to confirm that we did not exclude any articles in the screening stages that should have been included, we used a machine learning tool, ASReview version 1.6.2 (https://asreview.nl/ [van de Schoot et al., 2021]) to additionally re-check the excluded articles. ASReview is an open-source software that utilizes active-learning to screen large amounts of data primarily for scoping and systematic reviews. For this process, we uploaded all the articles from the original screening phase (following the removal of the duplicates). We then selected one article that met inclusion criteria and one that did not train the software on which articles should be included/excluded. We screened 407 articles in total, labeling each one as “relevant” (likely met inclusion criteria) and “irrelevant” (did not meet inclusion criteria). From the articles we labeled as relevant, we cross-checked these against our final list of included articles and reread the full text of any articles we had originally excluded to determine whether they met inclusion criteria. Synthesis plan The synthesis plan was a narrative summary following guidelines from (Cumpston et al., 2022) in which we focused primarily on the types of studies included, countries where the data collection took place and types of trauma reported. We presented the charting results as text in a table and graphical summaries of the research data. We did not conduct a formal publication bias analysis as this is not typically done in scoping reviews (Peters et al., 2021). Protocol deposit The protocol for this study is available through Open Science Framework and can be found here: https://doi.org/10.17605/OSF.IO/JRSQP. Results We ran our search in PubMed, Embase, PsycINFO, Web of Science Core Collection and Africa-Wide Information/NiPad, which retrieved 4,111 articles. After uploading all search results into Covidence, we removed 1,744 duplicates and were left with 2,367 articles for title and abstract screening. AS and EG double screened all titles and abstracts and removed 2,290 irrelevant studies. Their agreement rate was 97.5%. We then reviewed the full text of 77 articles. Forty-six studies met the criteria for inclusion. For the hand-searching process, AS identified 39 citations for articles to screen; after going through the full screening process for these articles, AS found four that met eligibility. In addition, AS discovered an additional article that met the criteria during a separate literature review. AS added these five articles to the pool for extraction. Using ASReview, AS tagged 57 articles as relevant, of which 46 were already in our included pool. Upon rereading the additional 11, they still did not meet our inclusion criteria; thus, the software did not identify any erroneously excluded articles. We were left with 51 articles that met the criteria. (See Figure 1 for the PRISMA flow diagram that captures the search and selection process.) We then extracted data from all the remaining articles into a Google document (see Table 1 for the articles that met eligibility and the data we extracted from each). There was 1 case study, 1 mixed-methods study (both qualitative and quantitative measures), 12 qualitative studies and 37 quantitative studies. Of the quantitative studies, the majority were cross-sectional. There were nine studies reporting on mortality, all of which were cohort studies, some of which followed patients for years and some which used national or regional health records to determine mortality rates. The number of participants in the studies ranged from 1 to 72,021,918. The percentage of female participants ranged from 0% to 100%. For studies that provided a mean age of participants (n = 42), the average age ranged from 26.7 to 51.9 years. The most represented countries were India (n = 19), Ethiopia (n = 9) and China (n = 6), which made up two-thirds of the studies. Additional countries represented in the studies were Indonesia (n = 5), Nigeria (n = 3), Uganda (n = 3), Brazil (n = 2), Egypt (n = 2), Ghana (n = 1) and South Africa (n = 1). The most studied mental health condition was schizophrenia. The most studied traumatic event was interpersonal violence (76.5%–84.3%) 1 including intimate partner violence and sexual and physical victimization types. Of the types of interpersonal violence, more than one-fifth were about physical restraint, which ranged from shackling people by their ankles to a tree to chemical restraint in a hospital (23.3%–34.9%) 1. There were no studies on man-made or natural disasters.Table 1.Articles and their characteristics included in the scoping reviewFirst author – year of publicationArticle titleCountryRegionStudy designPopulation description# of participants% FemaleMean age (years)Race/ethnicity of participantsType of trauma(s) assessedHow traumas were assessed% of participants with SMHC who experienced traumatic eventMental health diagnosisHow mental health diagnosis was assessedAfe 2016 (Afe et al., 2016)Intimate partner violence, psychopathology and the women with schizophrenia in an outpatient clinic South–South, NigeriaNigeriaCross River State, South–SouthQuantitative: cross-sectionalFemale outpatients with SCZ77100%38.3Efik/Ibibio (49.0%), Ibos (34.0%), Yoruba (16.0%) and Hausas (1.0%)IPVWHO Violence Against Women Instrument75%SCZSCIDAluh 2022 (Aluh et al., 2022)Experiences and perceptions of coercive practices in mental health care among service users in Nigeria: a qualitative studyNigeriaNot reportedQualitativeMental health service users at two major psychiatric hospitals3036.7%34.67Not reportedCoercive practices in hospitals, for example, use of ropes, chains and handcuffs for mechanical restraint, being whippedFocus group discussions40% chaining and 16.6% whippingSCZ, SZA, BD, depression and other mental and behavioral disorders due to psychoactive substanceNot reportedAmetaj 2021 (Ametaj et al., 2021)Traumatic Events and Posttraumatic Stress Disorder in Individuals with Severe Mental Illness in a Non-Western Setting: Data from Rural EthiopiaEthiopiaSodo District, Gurage ZoneQualitativePatients with SMI, their caregivers, healthcare providers, and community and religious leaders in a rural community48 in full sample; 13 with SMI46.15% of people with SMI35.15 years for people with SMIGurage (100%)Multiple, including chaining, animal attacks, rape and physical assaultSemi-structured interviewN/ASMI: SCZ, BD and depression with psychotic featuresOPCRIT conducted by psychiatric nursesAn 2016 (An et al., 2016)Physical restraint for psychiatric patients and its associations with clinical characteristics and the National Mental Health Law in ChinaChinaBeijingQuantitative: cross-sectionalPatients with psychiatric conditions1,36463.9%36.2Not reportedPhysical restraint, for example, use of belts to fix a patient to a bedChart review and confirmed during interviews35.2% of people with SCZ and 42.1% with mood disordersSCZ, mood disorders and otherConfirmed in a clinical interviewAsher 2017 (Asher et al., 2017)“I cry every day and night, I have my son tied in chains”: physical restraint of people with schizophrenia in community settings in EthiopiaEthiopiaSodo and Butajira districts, Gurage ZoneQualitativePeople with SCZ, their caregivers, community leaders, and primary and community health workers in rural community50 in full sample; 4 with SCZ52.0% of full sample; 25% of people with SCZNot reportedN/APhysical violence and restraint in community settings, for example, beating, use of handcuffs, chains and iron barsFocus group discussions and in-depth interviewsN/ASCZOPCRIT conducted by a psychiatric nurseBagewadi 2016 (Bagewadi et al., 2016)Standardized Mortality Ratio in Patients with schizophrenia – Findings from Thirthahalli: A Rural South Indian CommunityIndiaThirthahalli, KarnatakaQuantitative: cohortPatients with SCZ who live in a community setting who are alive and those who died943 in full sample;12 deceased33.3% of the deceased patients45.1 for the deceased patientsNot reportedAccidents leading to deathChart review and/or verbal report from family members0.11% of the full cohort; 8.3% of the deceased patientsSCZClinical diagnosis by a psychiatrist, followed by the MINIBelete 2017 (Belete, 2017a)Leveling and abuse among patients with bipolar disorder at psychiatric outpatient departments in EthiopiaEthiopiaAddis AbabaQuantitative: cross-sectionalOutpatients with BD41157.4%34.35Amhara (34.55%), Oromo (32.85%), Gurage (20%), Tigray (5.8%) and others (6.8%)Verbal or physical abuseChart review and by asking the patients and/or from family members37.7%BDClinical diagnosis by a psychiatrist or mental health professional specialistBelete 2017 (Belete, 2017b)Use of physical restraints among patients with bipolar disorder in Ethiopian Mental Specialized Hospital, outpatient department: cross-sectional studyEthiopiaAddis AbabaQuantitative: cross-sectionalOutpatients with BD40057.2%32Amhara (34.7%), Oromo (33.0%), Gurage (19.7%), Tigray (5.8%) and others (6.8%)Physical restraintChart review and/or verbal report from family members65%BDClinical diagnosis by a psychiatrist or mental health professional specialistBhattacharya 2022 (Bhattacharya, 2022)“The Day I Die Is the Day I Will Find My Peace”: Narratives of Family, Marriage, and Violence Among Women Living with Serious Mental Illness in IndiaIndiaNot reportedQualitativeCurrent or former female inpatients with SMI living in an urban setting11100%Mid‑30s to early 60s, no mean age providedNot reportedDomestic and sexual violenceIn-depth interviewsN/ASMI: SCZ, SZA or BDNot reportedChandra 2003 (Chandra et al., 2003)A Cry from the Darkness: Women with Severe Mental Illness in India Reveal Their Experiences with Sexual CoercionIndiaBangaloreMixed-methods (qualitative and structured interviews)Female inpatients with SMI50100%30Not reportedSexual coercion, that is, forced sexual intercourse, unwanted sexual play and sexual experiences involving threats or use of physical forceSexual Experiences Survey (SES) and semi-structured interview34% coercive experiences after onset on mental illnessSMI: recurrent depressive disorder, SCZ spectrum disorder, BD and other disordersClinical diagnosis by a psychiatristChandra 2003 (Chandra et al., 2003)Sexual coercion and abuse among women with a severe mental illness in India: An exploratory investigationIndiaBangaloreQuantitative: cross-sectionalFemale inpatients with SMI146100%31.6Not reportedSexual coercion, that is, forced sexual intercourse, unwanted sexual play, and sexual experiences involving threats or use of physical forceSexual Experiences Survey (SES)16% of adult sexual coercion and 7% of both adult sexual coercion and child sexual abuseSCZ-spectrum disorders, BD and depressionClinical diagnosis by a psychiatristde Oliveira 2013 (de Oliveira et al., 2013)Physical violence against patients with mental disorders in Brazil: sex differences in a cross-sectional studyBrazilNationalQuantitative: cross-sectionalPatients with SMI using public mental health services2,47551.6%44.9 women and 51.9 menNot reportedPhysical violenceSemi-structured interview12.0% of women and 16.0% of men experienced physical violence within health institutions by other patients, employees or health professionals. (Other physical violence reported could not be split out from lifetime exposure.)SCZ and other psychosis, BD, depressive disorder and othersChart reviewEl Missiry 2019 (El Missiry et al., 2019)Rates and profile of victimization in a sample of Egyptian patients with major mental illnessEgyptCairoQuantitative: cross-sectionalInpatients and outpatients with major mental illness30052.0% of full sample34.3 for victimized sampleNot reportedCovert/relational victimization or physical victimization, that is, threatened or subjected to corporeal damageVictimization Questionnaire (VQ)43.3%Major mental illness: SCZ, BD and major depressionSCIDFekadu 2015 (Fekadu et al., 2015)Excess mortality in severe mental illness: 10-year population-based cohort study in rural EthiopiaEthiopiaButajira District, Gurage ZoneQuantitative: cohortPeople with SMI and those without in a rural community919 in full cohort; 121 patients who died37.8%15–49 at baseline, no mean age providedNot reportedAccidents leading to death and homicidesVerbal autopsy9.10%SMI: SCZ, BD and severe depressionSCAN conducted by a psychiatrist or mental health professionalFekry 2012 (Fekry, 2011)Clinical and sociodemographic profile of victimized versus nonvictimized Egyptian patients with bipolar mood disorderEgyptCairoQuantitative: cross-sectionalInpatients and outpatients with BD10036.0% of full sample; 41.7% of victimized group31Not reportedCovert/relational victimization or physical victimization, that is, threatened or subjected to corporeal damageVictimization Questionnaire (VQ)48%BDSCIDGilmoor 2020 (Gilmoor et al., 2020)“If somebody could just understand what I am going through, it would make all the difference”: Conceptualizations of trauma in homeless populations experiencing severe mental illnessIndiaKanchipuram and ChennaiQualitativePeople with SMI who were previously homeless or at risk of homelessness2676.9%47Not reportedMultiple, including abuse, man-made accidents, natural disasters, violence, illness and deathIn-depth interviews and questions based on previous culturally relevant free listing exerciseN/ASMI: SCZ, BD, psychosis NOS, substance induced delirium and major depressive disorderSelf-reportGowda 2018 (Gowda et al., 2018)Restraint prevalence and perceived coercion among psychiatric inpatients from South India: A prospective studyIndiaBangaloreQuantitative: cohortInpatients with SCZ or mood disorders20045%33.5Not reportedRestraint measures: physical restraints, chemical restraints, seclusion/isolation and involuntary medicationInterviews with participants and health records66.5% experienced 1+ restraint measures; 20% physical restraint, 58% chemical restraint, 18% seclusion and 32% involuntary medicationSCZ or other psychotic disorders and mood disordersMINIHu 2022 (Hu et al., 2022)A Retrospective Analysis of Death Among Chinese Han Patients with schizophrenia from ShandongChinaShandongQuantitative: cohortPatients with SCZ who are alive and those who died72,102 in full sample; 11,766 deceased48.21%47.21Han (100%)Accidents leading to death and homicidesDeath certificates or forensic specialist6.97%SCZClinical diagnosis by a psychiatristJakhar 2015 (Jakhar et al., 2015)A cross sectional study of prevalence and correlates of current and past risks in schizophreniaIndiaNew DelhiQuantitative: cross-sectionalPatients with SCZ from the psychiatric ward of large mental hospitals27035.2%34.0Not reported“Risk from others”Ram Manohar Lohia Risk Assessment Interview11%SCZReferral by treating a clinician followed by Diagnostic Interview for Genetic Studies – Hindi version (DIGS)Lundberg 2012 (Lundberg et al., 2012b)Sexual Risk Behaviors and Sexual Abuse in Persons with Severe Mental Illness in Uganda: A Qualitative StudyUgandaKampalaQualitativeInpatients cleared for discharge and outpatients2065%18–49, no mean age providedNot reportedSexual abuseSemi-structured interviewN/ASMI: SCZ, BD or depressionChart reviewLundberg 2015 (Lundberg et al., 2015)Sexual Risk Behavior, Sexual Violence, and HIV in Persons with Severe Mental Illness in Uganda: Hospital-Based Cross-Sectional Study and National Comparison DataUgandaKampalaQuantitative: cross-sectionalFormer inpatients with SMI60257.0% (note: only women assessed for sexual violence)18–49, no mean age providedNot reportedSexual violenceWHO Violence Against Women Instrument24.2% of women experienced sexual violence by a partner; 10.5% of women experienced sexual violence by a non-partnerSMI: BD, nonaffective psychosis and major depressionChart reviewManjunatha 2019 (Manjunatha et al., 2019)Mortality in schizophrenia: A study of verbal autopsy from cohorts of two rural communities of South IndiaIndiaThirthahalli and Turuvekere, KarnatakaQuantitative: cohortPeople with SCZ from rural communities who died5347.1%50.5Not reportedDeath due to road traffic accidentsVerbal autopsy5.6%SCZClinical diagnosis by a psychiatrist, followed by the MINIMelo 2022 (Melo et al., 2022)All-cause and cause-specific mortality among people with severe mental illness in Brazil’s public health system, 2000–2015: a retrospective studyBrazilNationalQuantitative: cohortInpatients with SMI and those without in the public health system72,021,918 in full sample; 749,720 with SMI56.9% of the full sample and 50.5% of the people with SMI41.1 for full sampleNot reportedInjuries including interpersonal violence and unintentional causes such as fires, drowning, foreign body, road injuries and fallsChart review19.9% of deaths in the population with SMISMI: SCZ, SZA, BD or depressive disorderChart reviewMinas 2008 (Minas and Diatri, 2008)Pasung: Physical restraint and confinement of the mentally ill in the communityIndonesiaSamosirQuantitative: cross-sectionalPeople with SMI in the community1546.7%25–56 years, no mean age providedNot reportedPhysical restraint and confinement, that is, by wooden stocks, confined in a small room or hut, tied with rope or chainedPhysical identification (case finding)100%SCZ and otherSelf-report and reports by family or community membersMojtabai 2001 (Mojtabai et al., 2001)Mortality and long-term course in schizophrenia with a poor 2-year course: A study in a developing countryIndiaUrban and rural ChandigarhQuantitative: cohortPatients with SCZ in an urban and a rural setting171; 24 participants who died47% of full sample26.7 at baselineNot reportedDeath due to traffic accidentsVerbal autopsy and/or medical records0.58% of full sample; 4.2% of the deceased groupSCZPresent State Examination (PSE)Moodley 2023 (Moodley et al., 2023)The missed pandemic: Intimate partner violence in female mental-health-care-users during the COVID‑19 pandemicSouth AfricaKwaZulu-NatalQuantitative: cross-sectionalFemale outpatients with SMI154100%42.7Black (50.3%), White (17.2%), Indian (21.2%) and colored (11.3%)Interpersonal violenceWomen Abuse Screening Tool (WAST)46.6%SMI: SCZ spectrum disorder, BD, major depressive disorder, anxiety disorders and personality disordersDiagnoses were confirmed clinically using DSM‑5 criteria by the researcher and chart reviews provided by the treating doctorMpango 2023 (Mpango et al., 2023)Physical and sexual victimization of persons with severe mental illness seeking care in central and southwestern UgandaUgandaKampala and MasakaQuantitative: cross-sectionalOutpatients with SMI1,20154.5%36Not reportedPhysical and sexual victimizationAdverse life events module of the European Para-suicide Interview Schedule (EPSIS I)13.6% physical victimization and 8.6% sexual victimization in the last 12 monthsSMI: SCZ, BD and recurrent major depressive disorderChart review followed by the MININair 2020 (Nair et al., 2020)Prevalence and clinical correlates of intimate partner violence (IPV) in women with severe mental illness (SMI)India“Southern part of India”Quantitative: cross-sectionalFemale inpatients with SMI who cohabitate with a partner100100%34.2Not reportedIPVIndian Family Violence and Control Scale (IFVCS)20% in the last 12 monthsPsychosis and BDChart reviewNg 2019 (Ng et al., 2019)Trauma exposure, depression, suicidal ideation, and alcohol use in people with severe mental disorder in EthiopiaEthiopiaSodo District, Gurage zoneQuantitative: cross-sectionalPeople with severe mental disorders in a rural community30043%36Not reportedPhysical restraint, assault, being beaten, rape, and moreList of Threatening Experiences scale (LTE); 2 questions on restraint; 13 questions on locally established traumatic events, for example, rape, beaten and hit by car26.76% serious illness/injury-self; 46.33% restrained; 35.67% locally relevant traumatic eventsPrimary psychosis, BD and depression with psychotic featuresOPCRIT conducted by psychiatric nursesOpekitan 2017 (Opekitan et al., 2017)Socio-Demographic Characteristics, Partner Characteristics, Socioeconomic Variables, and Intimate Partner Violence in Women with schizophrenia in South–South NigeriaNigeriaSouth–SouthQuantitative: cross-sectionalFemale outpatients with SCZ79100%38.3Efik/Ibibio (51.0%), Ibos (33.0%), Yorubas (15.0%) and Hausas (1.0%)IPVWHO Violence Against Women Instrument73.0%SCZSCIDPaul 2018 (Paul, 2018)Are we doing enough? Stigma, discrimination and human rights violations of people living with schizophrenia in India: Implications for social work practiceIndiaMumbaiQualitativePeople with SCZ or SCZ-related disorders, family members and mental health professionals40 in full sample; 20 with SCZ-related disorder50% of people with SCZ40.85 for people with SCZNot reportedStigma, discrimination and human rights violations, including physical and emotional violenceIn-depth interviews and focus group discussionsN/ASCZ and/or a SCZ-related disorderNot reportedPonnudurai 2006 (Ponnudurai et al., 2006)Assessment of mortality and marital status of schizophrenic patients over a period of 13 yearsIndiaTamil NaduQuantitative: cohortOutpatients with SCZ who died6011.67%27 for men and 26.7 for women at baselineNot reportedDeath due to accidentsCase notes at the hospital and verbal autopsies confirmed by police reports1.67%SCZIndependently confirmed by two psychiatristsRan 2007 (Ran et al., 2007)Mortality in people with schizophrenia in rural China: 10-year cohort studyChinaXinjin County, ChengduQuantitative: cohortPatients with SCZ in a rural community who are alive and those who died500; 98 participants who died53.4% of full sample44.7Not reportedDeath due to accidentsDeath certificates and verbal autopsies2.6% of full sample; 13.3% of the deceased groupSCZPresent State Examination (PSE)Rani 2023 (Rani et al., 2023)A Qualitative Study to Understand the Nature of Abuse Experienced by Persons with Severe Mental IllnessIndiaBengaluruQualitativeOutpatients with SMIs, their caregivers and experts from the community51 in full sample; 14 with SMI50% of people with SMI18–60+, no mean age providedNot reportedPhysical abuse, psychological abuse, sexual abuse, social abuse and trauma in formal careKey informant interviews and focus group discussionsN/ASMI: BD, SCZ, SZA and recurrent depressive disorderNot reportedRani 2023 (Rani et al., 2023)Profiles of Victimized Outpatients with Severe Mental Illness in IndiaIndiaNot reportedQuantitative: cross-sectionalOutpatients with SMI15056%36.16Not reportedEmotional abuse, severe combined abuse, physical abuse and harassmentComposite Abuse Scale (CAS)100% experienced emotional abuse, 94% severe combined abuse, 92.7% physical abuse and 54% harassment in the past 12 monthsSMI: BD, SCZ, SZA and recurrent depressive disorderChart reviewRead 2009 (Read et al., 2009)Local suffering and the global discourse of mental health and human rights: An ethnographic study of responses to mental illness in rural GhanaGhanaKintampoQualitative and ethnographic methodsPeople with SMI, their families, healing practitioners and religious leaders within rural communities67 in full sample; 25 with SMINot reportedNot reportedNot reportedChaining, beating and withholding of foodObservation, semi-structured interviews and focus group discussionsN/ASMINot reportedReddy 2020 (Reddy et al., 2020)Childhood abuse and intimate partner violence among women with mood disordersIndiaBangaloreQuantitative: cross-sectionalWomen with SMI vs. healthy women with intimate partners251 with SMI and 72 healthy women100%35.35 for people with SMINot reportedChildhood abuse and IPVComposite Abuse Scale to assess the adulthood IPV29.1% severe combined abuse; 37.5% emotional abuse; 47.8% physical abuse; 25.9% harassment in the population with SMISMI: unipolar depression and BDClinical diagnosis by a psychiatrist, followed by the MINISadath 2014 (Sadath et al., 2014)Human rights violation in mental health: A case report from IndiaIndia“South India”Case reportMan with SCZ10%41Not reportedSolitary confinement in a single room/terrace with no ventilation, sanitary facilities, fed through window and personal care not given for >15 yearsPhysical identification (case finding)100%Paranoid SCZNot reportedSam 2019 (Sam et al., 2019)Stressful Life Events and Relapse in Bipolar Affective Disorder: A Cross-Sectional Study from a Tertiary Care Center of Southern IndiaIndiaKolenchery, KeralaQuantitative: cross-sectionalInpatients with BD12843.0%40.19Not reportedMajor personal illness or injuryPresumptive Stressful Life Events Scale (PSLES)3.4%BD: mania and depressionNot reported, beyond stating “ICD‑10 criteria”Subramanian 2017 (Subramanian et al., 2017)Role of stressful life events and kindling in bipolar disorder: Converging evidence from a mania-predominant illness courseIndia“Southern India”Quantitative: cross-sectionalInpatients and outpatients with BD14952.3%37.7Not reportedMajor personal illness or injuryPresumptive Stressful Life Events Scale (PSLES)0.8%BDSCIDSuryani 2011 (Suryani et al., 2011)Treating the untreated: Applying a community-based, culturally sensitive psychiatric intervention to confined and physically restrained mentally ill individuals in Bali, IndonesiaIndonesiaKarangasem regency, BaliQuantitative: cross-sectionalPeople with SCZ-spectrum disorders in the community2313%45.1Not reportedRestraint with iron shackles, ropes, wooden stocks or wooden cagesPhysical identification (case finding)91.3% restraint with iron shackles, ropes and wooden stocks; 8.7% confined to wooden cagesSCZ-spectrum disorderClinical diagnosis by a psychiatristTeferra 2011 (Teferra et al., 2011)Five-year mortality in a cohort of people with schizophrenia in EthiopiaEthiopiaButajira District, Gurage zoneQuantitative: cohortPatients with SCZ in a rural community who are alive and those who died307; 38 participants who died17.9% of the full sample; 10.5% of the deceased group35 for the deceased groupNot reportedDeath due to traffic accidentsVerbal autopsy0.7% of full sample; 5.3% of the deceased groupSCZSCAN conducted by a psychiatrist or mental health professionalTsigebrhan 2014 (Tsigebrhan et al., 2014)Violence and violent victimization in people with severe mental illness in a rural low-income country setting: A comparative cross-sectional community studyEthiopiaButajira District, Gurage ZoneQuantitative: cross-sectionalPeople with SMI and those without in a rural community401 in full sample; 201 with SMI and 200 without38.8% in the full sample40 for the full sample; 40.3 for people with SMINot reportedViolent victimization, for example, being kicked, dragged, chained or beaten; threatened or attacked with a weapon; forced to have sexual intercourseAdapted version of the McArthur Violence Interview17.4% in the past 12 monthsSMI: SCZ, SZA and BDSCAN conducted by a psychiatrist or mental health professionalVijayalakshmi 2012 (Vijayalakshmi et al., 2012)Gender-Related Differences in the Human Rights Needs of Patients with Mental IllnessIndiaNot reportedQuantitative: cross-sectionalAsymptomatic outpatients with SCZ or mood disorders10047.0%34.68Not reportedTorture, cruel, inhuman or degrading treatment or punishment; not allowed to leave home; sexual advances by family membersNeeds assessment questionnaire based on Universal Declaration of Human Rights61.0% sexual advances by family members and 35.0% not allowed to leave their homeMood disorders or SCZ disordersNot reportedWang 2020 (Wang et al., 2020)Frequency and correlates of violence against patients with schizophrenia living in rural ChinaChinaLuoding County, GuangdongQuantitative: cross-sectionalOutpatients with SCZ in a rural community48735.9%42.36Not reportedViolent victimization in the past 6 months, for example, sexual assault with violence, sexual harassment with physical contacts, verbal harassment with sexual content, nonsexual physical violence and verbal threat and abuse by family members or othersAsked participants18.9% experienced 1+ event in the past 6 monthsSCZChart reviewed followed by a clinical interviewWindarwati 2021 (Windarwati et al., 2021)A Journey of Hidden Outburst of Anger Shackling a Person with schizophrenia: The Indonesian ContextIndonesiaEast JavaQualitativePeople with SCZ, family members, volunteers, prominent figures and nurses23 in full sample; 5 with SCZ0% of people with SCZ> 40, no mean age providedNot reportedShackling by anklesRecords from regional health office and grounded-theory approach100%SCZNot reportedYosep 2021 (Yosep et al., 2021)How patients with schizophrenia “as a Victim” cope with violence in Indonesia: a qualitative studyIndonesiaWest JavaQualitativePatients with SCZ from the psychiatric ward of large mental hospitals4035.0%35.8Not reportedPushing, punching, kicking and restrainingSemi-structured interview100%SCZConfirmed diagnosis by a physicianYosep 2021 (Yosep et al., 2021)Experiences of Violence Among Individuals With schizophrenia in Indonesia a Phenomenological StudyIndonesiaWest JavaQualitativePatients with SCZ recruited from referral hospitals4037.5%35.6Not reportedVictimization by nurses and family members, including physical and verbal violenceFocus group discussionsN/ASCZNot reportedZerihun 2021 (Zerihun et al., 2021)Intimate partner violence among reproductive-age women with chronic mental illness attending a psychiatry outpatient department: cross-sectional facility-based study, Addis Ababa, EthiopiaEthiopiaAddis AbabaQuantitative: cross-sectionalFemale outpatients of reproductive age422100%32.1Not reportedIPVWHO Violence Against Women Instrument44.1% in the past 12 monthsSCZ, BD, and severe major depressionParticipants recruited from outpatient departments, but diagnosis process not statedZhang 2018 (Zhang et al., 2018)Long-Term Outcomes of Unlocking Chinese Patients with Severe Mental IllnessChinaHebeiQuantitative: cohortPatients with SMI10716.82% at baseline35.9 at baselineNot reportedPhysical restraint, for example, use of iron chains, iron cage, rope or a separate room or shedNot reported100% at baseline; 19.6% at Time 2 (year 2012); 17.8% at Time 3 (year 2016)SMI: SCZ, paranoid psychosis and BDNot reported, beyond stating “ICD‑10 criteria”Zhu 2014 (Zhu et al., 2014)Frequency of Physical Restraint and Its Associations with Demographic and Clinical Characteristics in a Chinese Psychiatric InstitutionChinaHunanQuantitative: cross-sectionalInpatients with psychiatric conditions160; 85 with SCZ and BP50.6% of full sample30.0Not reportedPhysical restraintMedical records and confirmed by interview51.3%Of the full sample: SCZ, mood disorders and othersChart review and confirmed by clinical interviewAcronyms: BP = bipolar disorder; DSM = Diagnostic and Statistical Manual of Mental Disorders; ICD = International Classification of Diseases; IPV = intimate partner violence; MINI = Mini International Neuropsychiatric Interview; N/A = not applicable; OPCRIT = Operational Criteria Checklist for Psychotic Illness; PSE = Present State Examination; SCID = Structured Clinical Interview for DSM-IV; SCZ = schizophrenia; SMI = severe mental illness or serious mental illness; SZA = schizoaffective disorder; WHO = World Health Organization.Note: the terminology in the table reflects the language the researchers used in their articles, such as SMI. Figure 1.PRISMA flow chart of the search and selection process for the scoping review. Table 1.Articles and their characteristics included in the scoping reviewFirst author – year of publicationArticle titleCountryRegionStudy designPopulation description# of participants% FemaleMean age (years)Race/ethnicity of participantsType of trauma(s) assessedHow traumas were assessed% of participants with SMHC who experienced traumatic eventMental health diagnosisHow mental health diagnosis was assessedAfe 2016 (Afe et al., 2016)Intimate partner violence, psychopathology and the women with schizophrenia in an outpatient clinic South–South, NigeriaNigeriaCross River State, South–SouthQuantitative: cross-sectionalFemale outpatients with SCZ77100%38.3Efik/Ibibio (49.0%), Ibos (34.0%), Yoruba (16.0%) and Hausas (1.0%)IPVWHO Violence Against Women Instrument75%SCZSCIDAluh 2022 (Aluh et al., 2022)Experiences and perceptions of coercive practices in mental health care among service users in Nigeria: a qualitative studyNigeriaNot reportedQualitativeMental health service users at two major psychiatric hospitals3036.7%34.67Not reportedCoercive practices in hospitals, for example, use of ropes, chains and handcuffs for mechanical restraint, being whippedFocus group discussions40% chaining and 16.6% whippingSCZ, SZA, BD, depression and other mental and behavioral disorders due to psychoactive substanceNot reportedAmetaj 2021 (Ametaj et al., 2021)Traumatic Events and Posttraumatic Stress Disorder in Individuals with Severe Mental Illness in a Non-Western Setting: Data from Rural EthiopiaEthiopiaSodo District, Gurage ZoneQualitativePatients with SMI, their caregivers, healthcare providers, and community and religious leaders in a rural community48 in full sample; 13 with SMI46.15% of people with SMI35.15 years for people with SMIGurage (100%)Multiple, including chaining, animal attacks, rape and physical assaultSemi-structured interviewN/ASMI: SCZ, BD and depression with psychotic featuresOPCRIT conducted by psychiatric nursesAn 2016 (An et al., 2016)Physical restraint for psychiatric patients and its associations with clinical characteristics and the National Mental Health Law in ChinaChinaBeijingQuantitative: cross-sectionalPatients with psychiatric conditions1,36463.9%36.2Not reportedPhysical restraint, for example, use of belts to fix a patient to a bedChart review and confirmed during interviews35.2% of people with SCZ and 42.1% with mood disordersSCZ, mood disorders and otherConfirmed in a clinical interviewAsher 2017 (Asher et al., 2017)“I cry every day and night, I have my son tied in chains”: physical restraint of people with schizophrenia in community settings in EthiopiaEthiopiaSodo and Butajira districts, Gurage ZoneQualitativePeople with SCZ, their caregivers, community leaders, and primary and community health workers in rural community50 in full sample; 4 with SCZ52.0% of full sample; 25% of people with SCZNot reportedN/APhysical violence and restraint in community settings, for example, beating, use of handcuffs, chains and iron barsFocus group discussions and in-depth interviewsN/ASCZOPCRIT conducted by a psychiatric nurseBagewadi 2016 (Bagewadi et al., 2016)Standardized Mortality Ratio in Patients with schizophrenia – Findings from Thirthahalli: A Rural South Indian CommunityIndiaThirthahalli, KarnatakaQuantitative: cohortPatients with SCZ who live in a community setting who are alive and those who died943 in full sample;12 deceased33.3% of the deceased patients45.1 for the deceased patientsNot reportedAccidents leading to deathChart review and/or verbal report from family members0.11% of the full cohort; 8.3% of the deceased patientsSCZClinical diagnosis by a psychiatrist, followed by the MINIBelete 2017 (Belete, 2017a)Leveling and abuse among patients with bipolar disorder at psychiatric outpatient departments in EthiopiaEthiopiaAddis AbabaQuantitative: cross-sectionalOutpatients with BD41157.4%34.35Amhara (34.55%), Oromo (32.85%), Gurage (20%), Tigray (5.8%) and others (6.8%)Verbal or physical abuseChart review and by asking the patients and/or from family members37.7%BDClinical diagnosis by a psychiatrist or mental health professional specialistBelete 2017 (Belete, 2017b)Use of physical restraints among patients with bipolar disorder in Ethiopian Mental Specialized Hospital, outpatient department: cross-sectional studyEthiopiaAddis AbabaQuantitative: cross-sectionalOutpatients with BD40057.2%32Amhara (34.7%), Oromo (33.0%), Gurage (19.7%), Tigray (5.8%) and others (6.8%)Physical restraintChart review and/or verbal report from family members65%BDClinical diagnosis by a psychiatrist or mental health professional specialistBhattacharya 2022 (Bhattacharya, 2022)“The Day I Die Is the Day I Will Find My Peace”: Narratives of Family, Marriage, and Violence Among Women Living with Serious Mental Illness in IndiaIndiaNot reportedQualitativeCurrent or former female inpatients with SMI living in an urban setting11100%Mid‑30s to early 60s, no mean age providedNot reportedDomestic and sexual violenceIn-depth interviewsN/ASMI: SCZ, SZA or BDNot reportedChandra 2003 (Chandra et al., 2003)A Cry from the Darkness: Women with Severe Mental Illness in India Reveal Their Experiences with Sexual CoercionIndiaBangaloreMixed-methods (qualitative and structured interviews)Female inpatients with SMI50100%30Not reportedSexual coercion, that is, forced sexual intercourse, unwanted sexual play and sexual experiences involving threats or use of physical forceSexual Experiences Survey (SES) and semi-structured interview34% coercive experiences after onset on mental illnessSMI: recurrent depressive disorder, SCZ spectrum disorder, BD and other disordersClinical diagnosis by a psychiatristChandra 2003 (Chandra et al., 2003)Sexual coercion and abuse among women with a severe mental illness in India: An exploratory investigationIndiaBangaloreQuantitative: cross-sectionalFemale inpatients with SMI146100%31.6Not reportedSexual coercion, that is, forced sexual intercourse, unwanted sexual play, and sexual experiences involving threats or use of physical forceSexual Experiences Survey (SES)16% of adult sexual coercion and 7% of both adult sexual coercion and child sexual abuseSCZ-spectrum disorders, BD and depressionClinical diagnosis by a psychiatristde Oliveira 2013 (de Oliveira et al., 2013)Physical violence against patients with mental disorders in Brazil: sex differences in a cross-sectional studyBrazilNationalQuantitative: cross-sectionalPatients with SMI using public mental health services2,47551.6%44.9 women and 51.9 menNot reportedPhysical violenceSemi-structured interview12.0% of women and 16.0% of men experienced physical violence within health institutions by other patients, employees or health professionals. (Other physical violence reported could not be split out from lifetime exposure.)SCZ and other psychosis, BD, depressive disorder and othersChart reviewEl Missiry 2019 (El Missiry et al., 2019)Rates and profile of victimization in a sample of Egyptian patients with major mental illnessEgyptCairoQuantitative: cross-sectionalInpatients and outpatients with major mental illness30052.0% of full sample34.3 for victimized sampleNot reportedCovert/relational victimization or physical victimization, that is, threatened or subjected to corporeal damageVictimization Questionnaire (VQ)43.3%Major mental illness: SCZ, BD and major depressionSCIDFekadu 2015 (Fekadu et al., 2015)Excess mortality in severe mental illness: 10-year population-based cohort study in rural EthiopiaEthiopiaButajira District, Gurage ZoneQuantitative: cohortPeople with SMI and those without in a rural community919 in full cohort; 121 patients who died37.8%15–49 at baseline, no mean age providedNot reportedAccidents leading to death and homicidesVerbal autopsy9.10%SMI: SCZ, BD and severe depressionSCAN conducted by a psychiatrist or mental health professionalFekry 2012 (Fekry, 2011)Clinical and sociodemographic profile of victimized versus nonvictimized Egyptian patients with bipolar mood disorderEgyptCairoQuantitative: cross-sectionalInpatients and outpatients with BD10036.0% of full sample; 41.7% of victimized group31Not reportedCovert/relational victimization or physical victimization, that is, threatened or subjected to corporeal damageVictimization Questionnaire (VQ)48%BDSCIDGilmoor 2020 (Gilmoor et al., 2020)“If somebody could just understand what I am going through, it would make all the difference”: Conceptualizations of trauma in homeless populations experiencing severe mental illnessIndiaKanchipuram and ChennaiQualitativePeople with SMI who were previously homeless or at risk of homelessness2676.9%47Not reportedMultiple, including abuse, man-made accidents, natural disasters, violence, illness and deathIn-depth interviews and questions based on previous culturally relevant free listing exerciseN/ASMI: SCZ, BD, psychosis NOS, substance induced delirium and major depressive disorderSelf-reportGowda 2018 (Gowda et al., 2018)Restraint prevalence and perceived coercion among psychiatric inpatients from South India: A prospective studyIndiaBangaloreQuantitative: cohortInpatients with SCZ or mood disorders20045%33.5Not reportedRestraint measures: physical restraints, chemical restraints, seclusion/isolation and involuntary medicationInterviews with participants and health records66.5% experienced 1+ restraint measures; 20% physical restraint, 58% chemical restraint, 18% seclusion and 32% involuntary medicationSCZ or other psychotic disorders and mood disordersMINIHu 2022 (Hu et al., 2022)A Retrospective Analysis of Death Among Chinese Han Patients with schizophrenia from ShandongChinaShandongQuantitative: cohortPatients with SCZ who are alive and those who died72,102 in full sample; 11,766 deceased48.21%47.21Han (100%)Accidents leading to death and homicidesDeath certificates or forensic specialist6.97%SCZClinical diagnosis by a psychiatristJakhar 2015 (Jakhar et al., 2015)A cross sectional study of prevalence and correlates of current and past risks in schizophreniaIndiaNew DelhiQuantitative: cross-sectionalPatients with SCZ from the psychiatric ward of large mental hospitals27035.2%34.0Not reported“Risk from others”Ram Manohar Lohia Risk Assessment Interview11%SCZReferral by treating a clinician followed by Diagnostic Interview for Genetic Studies – Hindi version (DIGS)Lundberg 2012 (Lundberg et al., 2012b)Sexual Risk Behaviors and Sexual Abuse in Persons with Severe Mental Illness in Uganda: A Qualitative StudyUgandaKampalaQualitativeInpatients cleared for discharge and outpatients2065%18–49, no mean age providedNot reportedSexual abuseSemi-structured interviewN/ASMI: SCZ, BD or depressionChart reviewLundberg 2015 (Lundberg et al., 2015)Sexual Risk Behavior, Sexual Violence, and HIV in Persons with Severe Mental Illness in Uganda: Hospital-Based Cross-Sectional Study and National Comparison DataUgandaKampalaQuantitative: cross-sectionalFormer inpatients with SMI60257.0% (note: only women assessed for sexual violence)18–49, no mean age providedNot reportedSexual violenceWHO Violence Against Women Instrument24.2% of women experienced sexual violence by a partner; 10.5% of women experienced sexual violence by a non-partnerSMI: BD, nonaffective psychosis and major depressionChart reviewManjunatha 2019 (Manjunatha et al., 2019)Mortality in schizophrenia: A study of verbal autopsy from cohorts of two rural communities of South IndiaIndiaThirthahalli and Turuvekere, KarnatakaQuantitative: cohortPeople with SCZ from rural communities who died5347.1%50.5Not reportedDeath due to road traffic accidentsVerbal autopsy5.6%SCZClinical diagnosis by a psychiatrist, followed by the MINIMelo 2022 (Melo et al., 2022)All-cause and cause-specific mortality among people with severe mental illness in Brazil’s public health system, 2000–2015: a retrospective studyBrazilNationalQuantitative: cohortInpatients with SMI and those without in the public health system72,021,918 in full sample; 749,720 with SMI56.9% of the full sample and 50.5% of the people with SMI41.1 for full sampleNot reportedInjuries including interpersonal violence and unintentional causes such as fires, drowning, foreign body, road injuries and fallsChart review19.9% of deaths in the population with SMISMI: SCZ, SZA, BD or depressive disorderChart reviewMinas 2008 (Minas and Diatri, 2008)Pasung: Physical restraint and confinement of the mentally ill in the communityIndonesiaSamosirQuantitative: cross-sectionalPeople with SMI in the community1546.7%25–56 years, no mean age providedNot reportedPhysical restraint and confinement, that is, by wooden stocks, confined in a small room or hut, tied with rope or chainedPhysical identification (case finding)100%SCZ and otherSelf-report and reports by family or community membersMojtabai 2001 (Mojtabai et al., 2001)Mortality and long-term course in schizophrenia with a poor 2-year course: A study in a developing countryIndiaUrban and rural ChandigarhQuantitative: cohortPatients with SCZ in an urban and a rural setting171; 24 participants who died47% of full sample26.7 at baselineNot reportedDeath due to traffic accidentsVerbal autopsy and/or medical records0.58% of full sample; 4.2% of the deceased groupSCZPresent State Examination (PSE)Moodley 2023 (Moodley et al., 2023)The missed pandemic: Intimate partner violence in female mental-health-care-users during the COVID‑19 pandemicSouth AfricaKwaZulu-NatalQuantitative: cross-sectionalFemale outpatients with SMI154100%42.7Black (50.3%), White (17.2%), Indian (21.2%) and colored (11.3%)Interpersonal violenceWomen Abuse Screening Tool (WAST)46.6%SMI: SCZ spectrum disorder, BD, major depressive disorder, anxiety disorders and personality disordersDiagnoses were confirmed clinically using DSM‑5 criteria by the researcher and chart reviews provided by the treating doctorMpango 2023 (Mpango et al., 2023)Physical and sexual victimization of persons with severe mental illness seeking care in central and southwestern UgandaUgandaKampala and MasakaQuantitative: cross-sectionalOutpatients with SMI1,20154.5%36Not reportedPhysical and sexual victimizationAdverse life events module of the European Para-suicide Interview Schedule (EPSIS I)13.6% physical victimization and 8.6% sexual victimization in the last 12 monthsSMI: SCZ, BD and recurrent major depressive disorderChart review followed by the MININair 2020 (Nair et al., 2020)Prevalence and clinical correlates of intimate partner violence (IPV) in women with severe mental illness (SMI)India“Southern part of India”Quantitative: cross-sectionalFemale inpatients with SMI who cohabitate with a partner100100%34.2Not reportedIPVIndian Family Violence and Control Scale (IFVCS)20% in the last 12 monthsPsychosis and BDChart reviewNg 2019 (Ng et al., 2019)Trauma exposure, depression, suicidal ideation, and alcohol use in people with severe mental disorder in EthiopiaEthiopiaSodo District, Gurage zoneQuantitative: cross-sectionalPeople with severe mental disorders in a rural community30043%36Not reportedPhysical restraint, assault, being beaten, rape, and moreList of Threatening Experiences scale (LTE); 2 questions on restraint; 13 questions on locally established traumatic events, for example, rape, beaten and hit by car26.76% serious illness/injury-self; 46.33% restrained; 35.67% locally relevant traumatic eventsPrimary psychosis, BD and depression with psychotic featuresOPCRIT conducted by psychiatric nursesOpekitan 2017 (Opekitan et al., 2017)Socio-Demographic Characteristics, Partner Characteristics, Socioeconomic Variables, and Intimate Partner Violence in Women with schizophrenia in South–South NigeriaNigeriaSouth–SouthQuantitative: cross-sectionalFemale outpatients with SCZ79100%38.3Efik/Ibibio (51.0%), Ibos (33.0%), Yorubas (15.0%) and Hausas (1.0%)IPVWHO Violence Against Women Instrument73.0%SCZSCIDPaul 2018 (Paul, 2018)Are we doing enough? Stigma, discrimination and human rights violations of people living with schizophrenia in India: Implications for social work practiceIndiaMumbaiQualitativePeople with SCZ or SCZ-related disorders, family members and mental health professionals40 in full sample; 20 with SCZ-related disorder50% of people with SCZ40.85 for people with SCZNot reportedStigma, discrimination and human rights violations, including physical and emotional violenceIn-depth interviews and focus group discussionsN/ASCZ and/or a SCZ-related disorderNot reportedPonnudurai 2006 (Ponnudurai et al., 2006)Assessment of mortality and marital status of schizophrenic patients over a period of 13 yearsIndiaTamil NaduQuantitative: cohortOutpatients with SCZ who died6011.67%27 for men and 26.7 for women at baselineNot reportedDeath due to accidentsCase notes at the hospital and verbal autopsies confirmed by police reports1.67%SCZIndependently confirmed by two psychiatristsRan 2007 (Ran et al., 2007)Mortality in people with schizophrenia in rural China: 10-year cohort studyChinaXinjin County, ChengduQuantitative: cohortPatients with SCZ in a rural community who are alive and those who died500; 98 participants who died53.4% of full sample44.7Not reportedDeath due to accidentsDeath certificates and verbal autopsies2.6% of full sample; 13.3% of the deceased groupSCZPresent State Examination (PSE)Rani 2023 (Rani et al., 2023)A Qualitative Study to Understand the Nature of Abuse Experienced by Persons with Severe Mental IllnessIndiaBengaluruQualitativeOutpatients with SMIs, their caregivers and experts from the community51 in full sample; 14 with SMI50% of people with SMI18–60+, no mean age providedNot reportedPhysical abuse, psychological abuse, sexual abuse, social abuse and trauma in formal careKey informant interviews and focus group discussionsN/ASMI: BD, SCZ, SZA and recurrent depressive disorderNot reportedRani 2023 (Rani et al., 2023)Profiles of Victimized Outpatients with Severe Mental Illness in IndiaIndiaNot reportedQuantitative: cross-sectionalOutpatients with SMI15056%36.16Not reportedEmotional abuse, severe combined abuse, physical abuse and harassmentComposite Abuse Scale (CAS)100% experienced emotional abuse, 94% severe combined abuse, 92.7% physical abuse and 54% harassment in the past 12 monthsSMI: BD, SCZ, SZA and recurrent depressive disorderChart reviewRead 2009 (Read et al., 2009)Local suffering and the global discourse of mental health and human rights: An ethnographic study of responses to mental illness in rural GhanaGhanaKintampoQualitative and ethnographic methodsPeople with SMI, their families, healing practitioners and religious leaders within rural communities67 in full sample; 25 with SMINot reportedNot reportedNot reportedChaining, beating and withholding of foodObservation, semi-structured interviews and focus group discussionsN/ASMINot reportedReddy 2020 (Reddy et al., 2020)Childhood abuse and intimate partner violence among women with mood disordersIndiaBangaloreQuantitative: cross-sectionalWomen with SMI vs. healthy women with intimate partners251 with SMI and 72 healthy women100%35.35 for people with SMINot reportedChildhood abuse and IPVComposite Abuse Scale to assess the adulthood IPV29.1% severe combined abuse; 37.5% emotional abuse; 47.8% physical abuse; 25.9% harassment in the population with SMISMI: unipolar depression and BDClinical diagnosis by a psychiatrist, followed by the MINISadath 2014 (Sadath et al., 2014)Human rights violation in mental health: A case report from IndiaIndia“South India”Case reportMan with SCZ10%41Not reportedSolitary confinement in a single room/terrace with no ventilation, sanitary facilities, fed through window and personal care not given for >15 yearsPhysical identification (case finding)100%Paranoid SCZNot reportedSam 2019 (Sam et al., 2019)Stressful Life Events and Relapse in Bipolar Affective Disorder: A Cross-Sectional Study from a Tertiary Care Center of Southern IndiaIndiaKolenchery, KeralaQuantitative: cross-sectionalInpatients with BD12843.0%40.19Not reportedMajor personal illness or injuryPresumptive Stressful Life Events Scale (PSLES)3.4%BD: mania and depressionNot reported, beyond stating “ICD‑10 criteria”Subramanian 2017 (Subramanian et al., 2017)Role of stressful life events and kindling in bipolar disorder: Converging evidence from a mania-predominant illness courseIndia“Southern India”Quantitative: cross-sectionalInpatients and outpatients with BD14952.3%37.7Not reportedMajor personal illness or injuryPresumptive Stressful Life Events Scale (PSLES)0.8%BDSCIDSuryani 2011 (Suryani et al., 2011)Treating the untreated: Applying a community-based, culturally sensitive psychiatric intervention to confined and physically restrained mentally ill individuals in Bali, IndonesiaIndonesiaKarangasem regency, BaliQuantitative: cross-sectionalPeople with SCZ-spectrum disorders in the community2313%45.1Not reportedRestraint with iron shackles, ropes, wooden stocks or wooden cagesPhysical identification (case finding)91.3% restraint with iron shackles, ropes and wooden stocks; 8.7% confined to wooden cagesSCZ-spectrum disorderClinical diagnosis by a psychiatristTeferra 2011 (Teferra et al., 2011)Five-year mortality in a cohort of people with schizophrenia in EthiopiaEthiopiaButajira District, Gurage zoneQuantitative: cohortPatients with SCZ in a rural community who are alive and those who died307; 38 participants who died17.9% of the full sample; 10.5% of the deceased group35 for the deceased groupNot reportedDeath due to traffic accidentsVerbal autopsy0.7% of full sample; 5.3% of the deceased groupSCZSCAN conducted by a psychiatrist or mental health professionalTsigebrhan 2014 (Tsigebrhan et al., 2014)Violence and violent victimization in people with severe mental illness in a rural low-income country setting: A comparative cross-sectional community studyEthiopiaButajira District, Gurage ZoneQuantitative: cross-sectionalPeople with SMI and those without in a rural community401 in full sample; 201 with SMI and 200 without38.8% in the full sample40 for the full sample; 40.3 for people with SMINot reportedViolent victimization, for example, being kicked, dragged, chained or beaten; threatened or attacked with a weapon; forced to have sexual intercourseAdapted version of the McArthur Violence Interview17.4% in the past 12 monthsSMI: SCZ, SZA and BDSCAN conducted by a psychiatrist or mental health professionalVijayalakshmi 2012 (Vijayalakshmi et al., 2012)Gender-Related Differences in the Human Rights Needs of Patients with Mental IllnessIndiaNot reportedQuantitative: cross-sectionalAsymptomatic outpatients with SCZ or mood disorders10047.0%34.68Not reportedTorture, cruel, inhuman or degrading treatment or punishment; not allowed to leave home; sexual advances by family membersNeeds assessment questionnaire based on Universal Declaration of Human Rights61.0% sexual advances by family members and 35.0% not allowed to leave their homeMood disorders or SCZ disordersNot reportedWang 2020 (Wang et al., 2020)Frequency and correlates of violence against patients with schizophrenia living in rural ChinaChinaLuoding County, GuangdongQuantitative: cross-sectionalOutpatients with SCZ in a rural community48735.9%42.36Not reportedViolent victimization in the past 6 months, for example, sexual assault with violence, sexual harassment with physical contacts, verbal harassment with sexual content, nonsexual physical violence and verbal threat and abuse by family members or othersAsked participants18.9% experienced 1+ event in the past 6 monthsSCZChart reviewed followed by a clinical interviewWindarwati 2021 (Windarwati et al., 2021)A Journey of Hidden Outburst of Anger Shackling a Person with schizophrenia: The Indonesian ContextIndonesiaEast JavaQualitativePeople with SCZ, family members, volunteers, prominent figures and nurses23 in full sample; 5 with SCZ0% of people with SCZ> 40, no mean age providedNot reportedShackling by anklesRecords from regional health office and grounded-theory approach100%SCZNot reportedYosep 2021 (Yosep et al., 2021)How patients with schizophrenia “as a Victim” cope with violence in Indonesia: a qualitative studyIndonesiaWest JavaQualitativePatients with SCZ from the psychiatric ward of large mental hospitals4035.0%35.8Not reportedPushing, punching, kicking and restrainingSemi-structured interview100%SCZConfirmed diagnosis by a physicianYosep 2021 (Yosep et al., 2021)Experiences of Violence Among Individuals With schizophrenia in Indonesia a Phenomenological StudyIndonesiaWest JavaQualitativePatients with SCZ recruited from referral hospitals4037.5%35.6Not reportedVictimization by nurses and family members, including physical and verbal violenceFocus group discussionsN/ASCZNot reportedZerihun 2021 (Zerihun et al., 2021)Intimate partner violence among reproductive-age women with chronic mental illness attending a psychiatry outpatient department: cross-sectional facility-based study, Addis Ababa, EthiopiaEthiopiaAddis AbabaQuantitative: cross-sectionalFemale outpatients of reproductive age422100%32.1Not reportedIPVWHO Violence Against Women Instrument44.1% in the past 12 monthsSCZ, BD, and severe major depressionParticipants recruited from outpatient departments, but diagnosis process not statedZhang 2018 (Zhang et al., 2018)Long-Term Outcomes of Unlocking Chinese Patients with Severe Mental IllnessChinaHebeiQuantitative: cohortPatients with SMI10716.82% at baseline35.9 at baselineNot reportedPhysical restraint, for example, use of iron chains, iron cage, rope or a separate room or shedNot reported100% at baseline; 19.6% at Time 2 (year 2012); 17.8% at Time 3 (year 2016)SMI: SCZ, paranoid psychosis and BDNot reported, beyond stating “ICD‑10 criteria”Zhu 2014 (Zhu et al., 2014)Frequency of Physical Restraint and Its Associations with Demographic and Clinical Characteristics in a Chinese Psychiatric InstitutionChinaHunanQuantitative: cross-sectionalInpatients with psychiatric conditions160; 85 with SCZ and BP50.6% of full sample30.0Not reportedPhysical restraintMedical records and confirmed by interview51.3%Of the full sample: SCZ, mood disorders and othersChart review and confirmed by clinical interviewAcronyms: BP = bipolar disorder; DSM = Diagnostic and Statistical Manual of Mental Disorders; ICD = International Classification of Diseases; IPV = intimate partner violence; MINI = Mini International Neuropsychiatric Interview; N/A = not applicable; OPCRIT = Operational Criteria Checklist for Psychotic Illness; PSE = Present State Examination; SCID = Structured Clinical Interview for DSM-IV; SCZ = schizophrenia; SMI = severe mental illness or serious mental illness; SZA = schizoaffective disorder; WHO = World Health Organization.Note: the terminology in the table reflects the language the researchers used in their articles, such as SMI. Articles and their characteristics included in the scoping review Acronyms: BP = bipolar disorder; DSM = Diagnostic and Statistical Manual of Mental Disorders; ICD = International Classification of Diseases; IPV = intimate partner violence; MINI = Mini International Neuropsychiatric Interview; N/A = not applicable; OPCRIT = Operational Criteria Checklist for Psychotic Illness; PSE = Present State Examination; SCID = Structured Clinical Interview for DSM-IV; SCZ = schizophrenia; SMI = severe mental illness or serious mental illness; SZA = schizoaffective disorder; WHO = World Health Organization. Note: the terminology in the table reflects the language the researchers used in their articles, such as SMI. PRISMA flow chart of the search and selection process for the scoping review. Discussion This scoping review aimed to synthesize the existing literature on trauma exposure experienced by adults with preexisting SMHCs in LMICs. By including multiple types of studies, we captured a breadth of trauma types and SMHCs that inform the different ways we conceptualize trauma exposure across mental health diagnoses in LMIC settings. The range of sample sizes, sexes represented, age of participants and variation in diagnoses also illustrates the diversity of studies in the field, which may make the findings more reflective of real-world experiences of patients with SMHCs. Of the 51 included articles, India, Ethiopia and China made up more than 66% of the studies; thus, even though there was a broad range of types of studies and demographics, our understanding of the field derives from a small subset of countries. For example, eight of the nine mortality studies were from India, Ethiopia and China, providing data on traumas that led to death in these populations. Furthermore, with only 10 countries represented in total, there were no data from the more than 120 other LMICs. Though the additional articles outside of India, Ethiopia and China included studies from Southeast Asia, South America and Africa, these regions and the countries within them have diverse populations, geographies and cultures. As such, the studies included may not reflect the full scope of types and frequencies of traumas that people with SMHCs face within and across such heterogeneous LMICs. Interpersonal violence, which accounted for the majority of studies, emerged as a significant category of traumatic events. Such studies support the notion that people with SMHCs are at additional risk for victimization given their mental health conditions. For example, having an SMHC is associated with social and physical isolation and potential cognitive impairments and/or symptoms (i.e., positive and negative symptoms), which increase the likelihood of being targeted by a perpetrator (de Vries et al., 2019). Alternatively, in many studies family members said they had tied up their loved ones in order to protect them from harm in the community and had no other options given the lack of treatment options and accessible and affordable health care (Asher et al., 2017). One type of interpersonal violence, “chaining” and “restraining,” represented a significant percentage of events. However, the type of restraint differed widely. While some participants were kept in wooden cages in their community for years, some were in solitary isolation as inpatients for a short period. In both high-income countries (the United States) and in LMICs (Ethiopia and China), people with SMHCs have described restraint as traumatizing (Frueh et al., 2005; Zhu et al., 2014; Ametaj et al., 2021). In Uganda, the United Nations Office of the High Commissioner for Human Rights has described the use of physical, mechanical and chemical restraints and isolation/seclusion in psychiatric facilities in conflict with the Convention on the Rights of Persons with Disabilities’ Article 15 (“Freedom from torture and cruel, inhuman or degrading treatment or punishment” for all people) (OHCHR, 2016). Even in nonpsychiatric contexts, there is evidence that physical restraint can lead to physical and psychological harm, including PTSD (Franks et al., 2021). The concept of restraint can be controversial, however, because it can be deemed necessary by hospital staff for a patient’s safety depending on an institution’s clinical guidelines. Restraint is also not captured by measures that are often used to assess trauma exposure such as the LEC and the CIDI. Such standardized measures can sometimes fail to capture a wider range of exposures as well as the subjective nature of what is considered traumatic. This may have affected how participants reported traumatic events in these countries. This is particularly relevant in non-Western settings. Some researchers maintain that the acknowledgment of traumatic events and subsequent traumatic stress is not simply a response to objective experiences, but is shaped by the culture in which one exists and is understood through what is considered traumatic in their setting (Chentsova-Dutton and Maercker, 2019). The value of including qualitative research is that it can augment quantitative reports by offering insights into personal trauma experiences and contextual factors in these settings. Surprisingly, there were no studies about man-made or natural disasters, such as war, fires and landslides. There is literature about vulnerable populations being left behind in dangerous situations because they are incapable of fleeing, such as the elderly in conflict zones (Murthy and Lakshminarayana, 2006) or during environmental crises like Hurricane Katrina (Dosa et al., 2010). However, none of these trauma types were elicited in this search. It is possible that this is because it is harder to conduct research during events like wars (Cohen and Arieli, 2011), and this absence partially reflects a survival bias. That is, people with SMHCs who died during disasters may not be represented in studies. Despite this, there were mortality studies in the scoping review that captured types of traumas that led to death, somewhat counterbalancing this potential survival bias. Such studies were conducted in more controlled environments in which populations could be followed for years or in countries where there were strong national or regional electronic medical records where accurate patient outcome data could be pulled more easily. Limitations There are some limitations that should be noted here. Although we tried to capture an extensive range of trauma types, we know that there are traumas that are culturally relevant in the countries we looked at that might not meet DSM-5 or ICD-11 criteria. While we tried to account for this by including “idioms of distress” and terminology about human rights violations and chaining/restraint, it is likely that there were trauma types that have not been studied by researchers in these contexts or that used different terms from ours that we did not retrieve. In addition, if an article did not include the name of the country in its title or abstract or a grouping term for the countries that met our criteria (i.e., “LMIC” or “East Africa”), we would likely have missed it. We may have missed articles that were not indexed in the American or European databases used here, as some articles from LMICs do not get included in these databases. However, we included Africa-Wide Information/NiPad in order to capture some of the articles that might have been otherwise missed. We limited inclusion to published studies in the interest of completing this scoping review in a timely manner. We acknowledge that excluding gray literature may potentially introduce publication bias. Inclusion of gray literature to address this research question should be considered in a future systematic review. Lastly, we did not include articles with potential mediators between SMHCs and trauma exposure, such as substance use or economic disadvantage. Studies with mediators would be worthwhile to look at in the future. Conclusion Future research on trauma types that people with SMHCs are exposed to in LMICs should expand beyond 10 countries to more than 120 additional countries that qualify as LMICs. Research on man-made and natural disasters in this population is warranted. Additional qualitative research may help identify locally relevant trauma types as well as understand people’s lived experiences. Supporting information Stevenson et al. supplementary material Stevenson et al. supplementary material In addition to the reviewer comments, please also add a methodological guideline in the methods section, for instance, Arksey and O’Malley, JBI etc. No accompanying comment. No accompanying comment. No accompanying comment.