Skip to main content
Schizophrenia Bulletin logoLink to Schizophrenia Bulletin
. 2025 Oct 31;52(5):sbaf195. doi: 10.1093/schbul/sbaf195

Prevalence, Correlates, and Treatment Gap of Schizophrenia Among Adults in Bangladesh: Findings From a Nationwide Household Survey

Ahsan Aziz Sarkar 1,✉, Md Faruq Alam 2, Mohammad Tariqul Alam 3, Helal Uddin Ahmed 4, Niaz Mohammad Khan 5, Munshi Mohammad Jalal Uddin 6
PMCID: PMC13645834  PMID: 41172138

Abstract

Background and Hypothesis

Schizophrenia contributes significantly to the global burden of disease; however, the data from low- and middle-income countries (LMICs) remain limited. Bangladesh, characterized by distinct sociodemographic and nutritional risk patterns, provides a critical context for understanding the epidemiology of schizophrenia in such settings. We hypothesized that the prevalence of schizophrenia would be higher among individuals residing in rural areas and those with lower levels of education and economic status.

Study Design

We conducted a cross-sectional, nationwide household (HH) survey as part of the Bangladesh National Mental Health Survey 2019. A stratified, multi-stage random sampling approach was used to recruit 7270 adults aged 18 years and older. Participants were initially screened using the 24-item Self-Reporting Questionnaire. Those screening positive underwent diagnostic assessment by trained psychiatrists using the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition criteria.

Study Results

The weighted prevalence of schizophrenia among adults was 0.6% (95% confidence interval: 0.4%-0.9%). Multivariate analyses revealed that schizophrenia was significantly associated with unemployment, a family history of mental illness, a family history of suicidal behavior and being divorced or separated. The prevalence did not differ between age, sex, residence (urban vs. rural), income, or educational level. Among individuals diagnosed with schizophrenia, 39.3% had received treatment.

Conclusions

This study provides the first nationwide HH survey data on schizophrenia in Bangladesh, identifying familial and marital disruptions as key correlates. These findings challenge commonly held assumptions about urbanicity and socioeconomic disadvantage as risk factors in LMIC contexts and highlight the urgent need to address the treatment gap.

Keywords: schizophrenia, prevalence, risk factors, treatment gap, Bangladesh, LMICs, survey

Introduction

Schizophrenia is a complex major psychiatric disorder that typically follows a chronic, lifelong course and imposes a substantial burden on individuals, families, and society. Globally, its median point, period, and lifetime prevalence stand at 0.46%, 0.33%, and 0.40%, respectively, 1 and it is associated with an average of 14.5 potential years of life lost.2 It ranks among the top fifteen causes of disability worldwide, contributing ~1.7% of the total years lived with disability and the burden of disability-adjusted life year is 4 times higher in low- and middle-income countries (LMICs) than in developed countries.3 In addition to its clinical impact, the disorder leads to marked impairments in social and occupational functioning, diminished quality of life, and substantial financial and caregiving burdens on families and communities. Addressing this burden requires an understanding of the epidemiology of schizophrenia and its correlates.

Understanding the sociodemographic correlates of an illness is essential, as these factors can aid in early diagnosis, shed light on the role of social and cultural influences in the development of multifactorial illnesses, identify vulnerable populations, underscore cross-cultural differences, and inform targeted preventive and therapeutic strategies, including public health policy formulation. Although schizophrenia is widely recognized as a highly heritable disorder, with an estimated heritability of 79%, a growing body of evidence suggests that a range of environmental and sociodemographic factors are associated with its development.4,5 These factors include, but are not limited to, earlier age of onset in males, urban birth and residence, a history of obstetric complications, maternal infections, maternal malnutrition (eg, anemia, folate deficiency), low prenatal and early life vitamin D levels, winter and spring births, migration, cannabis use, childhood physical and psychological trauma, and tobacco smoking. In Bangladesh, the prevalence of several of these risk factors is notably high. Recent data indicate that among women of reproductive age (15-49 years), 7.5% have vitamin A deficiency, 69.9% are deficient in vitamin D, 20.2% lack adequate folate, 43.4% are zinc deficient, 29.6% have iodine deficiency, and 28.9% suffer from anemia.6 Furthermore, the country continues to grapple with a high burden of undernutrition: among children under 5, 33.4% in urban areas and 40.2% in rural areas are undernourished—one of the highest rates globally.7 Although the burden of communicable diseases in Bangladesh has significantly decreased between 1990 and 2019, infectious diseases remain a considerable public health concern.8 These nutritional deficiencies and infectious disease burdens are disproportionately concentrated in rural areas, among lower-income families, and among individuals with limited educational attainment, reflecting socioeconomic disparities.7 Given the suggested links between these factors and schizophrenia, it is imperative to examine how the unique sociodemographic and health-related conditions in Bangladesh may influence the occurrence of this disorder.

Despite extensive research in high-income countries, national data on the prevalence and correlates of schizophrenia remain limited in LMICs, including Bangladesh. The country’s unique sociodemographic context—marked by rapid urbanization, persistent poverty, widespread nutritional deficiencies, and inequitable access to health services may shape the epidemiology of schizophrenia in ways that diverge from patterns observed in more developed settings. Moreover, the treatment gap for mental disorders in LMICs is alarmingly wide, with an estimated 85% of affected individuals receiving no formal care.9 Understanding the factors associated not only with the occurrence of schizophrenia but also with treatment-seeking behavior is therefore critical for developing effective, locally tailored interventions. This study aimed to address critical knowledge gaps in the epidemiology of schizophrenia in Bangladesh through the following objectives: firstly, to estimate the prevalence of schizophrenia among adults using data collected through a multi-stage, nationwide household (HH) survey; secondly, to identify key sociodemographic and familial correlates associated with the disorder; and thirdly, to assess correlates of treatment-seeking behavior among individuals diagnosed with schizophrenia. Findings from this study are expected to support evidence-based mental health planning and guide resource allocation and policy formulation in comparable LMIC settings.

Methods

Design, Settings, and Participants

This study is part of the Bangladesh National Mental Health Survey 2019, a comprehensive nationwide initiative designed to assess the mental health status of the adult population. Figure 1 presents an overview of the study design and sampling strategy.

Figure 1.

Figure 1

Diagram of Overall Steps of the Study

This multicentric, stratified, HH-based survey was conducted across Bangladesh in 2019. A 3-stage random sampling approach was employed to ensure national representativeness. First, primary sampling units (PSUs) were selected using stratified cluster random sampling from ~300 000 PSUs defined in the national population census. In the second stage, using systematic random sampling, 9 male and 9 female HHs were chosen from an average of 100-220 HHs within each PSU. Finally, in the third stage, 1 adult respondent was randomly selected from each HH using simple random sampling.

Eligible participants were adult Bangladeshi residents aged 18 years or older who had been living in the selected HH for at least 6 months. A HH was defined as a group of people living together in the same residence and sharing meals, with a designated head of HH making decisions for the family. In accordance with this definition, individuals residing in institutional settings (eg, dormitories, messes, hostels, long-term care institutions) and temporary residents such as tourists were excluded from the sampling frame. A team consisting of 3 data enumerators and a psychiatrist surveyed 18 HHs per PSU over a 3-day period. To cover all 496 PSUs, 20 of these teams worked in the field for ~4 months, with each team accompanied by a psychiatrist for their 3-day stay in each location. The psychiatrists were recruited from nearby medical college hospitals, and the data enumerators all held at least a graduate degree and had previous experience in community surveys. Prior to the study, 70 data enumerators and 30 psychiatrists completed an 8-day training program covering data collection, monitoring, and diagnostic procedures. This training was followed by a pilot study conducted in 2 urban and 2 rural PSUs. Written informed consent was obtained from all selected participants. Data were collected through face-to-face interviews conducted by trained enumerators using a structured questionnaire programmed on tablet devices. The questionnaire included a semi-structured sociodemographic module and the 24-item Self-Reporting Questionnaire (SRQ-24). For SRQ administration, able participants completed the tool themselves, while for illiterate participants, enumerators read out each question verbatim and recorded the responses. If participants had any difficulty understanding an item, enumerators provided clarification strictly according to a predefined protocol. Participants who screened positive on the SRQ-24 were subsequently assessed by trained research psychiatrists using the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria to establish clinical diagnoses. All data were uploaded to a secure cloud server in real time.

The study protocol received ethical approval from the National Institute of Mental Health, Bangladesh (reference: NIMH/2018/1984) and the Bangladesh Medical Research Council (reference: BMRC/NREC/2016-2019/152). Further details regarding the survey methodology and tools are available in the Bangladesh National Mental Health Survey Report 2019.10

Measures

To screen for mental disorders, the Bangla version of the 24-item Self-Reporting Questionnaire (SRQ-24) was administered. The tool has shown sensitivity between 62.9% and 90%, and specificity ranging from 44% to 95.2% across cultures.11 The first 20 items of the SRQ-24 assess symptoms associated with neurotic disorders, while the remaining 4 items screen for psychotic symptoms. The Bangla-validated version of the SRQ has shown high sensitivity, with 97.9% for a cutoff score of 4 in West Bengal and 79.5% for a cutoff score of 6 in Bangladesh.12,13 For the psychotic items, 1 study reported sensitivity of 47.6% and specificity of 98.9% at a cutoff score of 1 or more.13

In the present study, a participant was considered to screen positive if they responded positively to 4 or more of the first 20 items or to any 1 of the psychosis-related items (items 21-24). All individuals who screened positive were subsequently interviewed by a trained research psychiatrist, who was part of the field data collection team and recruited from a nearby medical college hospital. A clinical diagnostic interview was conducted, and diagnoses were made based on the criteria outlined in the DSM-5. The DSM-5, published by the American Psychiatric Association in 2013, is the standard classification system used globally for the diagnosis and categorization of mental disorders.14

Statistical Analyses

All analyses accounted for the complex survey design, including stratification, clustering, and sampling weights. Unweighted frequencies are reported for absolute numbers, while percentages, means, and standard deviations are presented as weighted estimates. Sampling weights were derived from the probability of selection at each stage and subsequently adjusted for non-response. These were then calibrated to match national population distributions based on the 2011 Bangladesh Population and Housing Census.

All statistical analyses were conducted using SPSS version 23.0. Descriptive statistics, including means, standard deviations, and proportions, were used to summarize the sociodemographic characteristics of the study population. Household economic status was estimated using principal component analysis of HH assets. The resulting wealth quintiles were consolidated into 3 categories: upper (highest and high), middle, and lower (low and lowest). For the purposes of prevalence estimation and association analyses, schizophrenia was defined as a composite category comprising schizophrenia (n = 45), schizophreniform disorder (n = 2), and schizoaffective disorder (n = 2), while brief psychotic disorder and delusional disorder of schizophrenia spectrum disorders were excluded from this category. Prevalence rates were calculated across key subgroups such as sex, urban versus rural residence, and age categories and are reported as percentages with corresponding 95% confidence intervals (CIs). All prevalence estimates and logistic regression analyses were conducted using survey-weighted procedures to account for the complex stratified cluster sampling design. Confidence intervals were adjusted for clustering, stratification, and weighting, incorporating the design effect of the survey.

To identify factors associated with schizophrenia, univariate logistic regression analyses were conducted for key sociodemographic and familial variables. Variables demonstrating statistical significance in univariate analyses were subsequently included in multivariate logistic regression models to adjust for potential confounders. To examine factors related to treatment-seeking behavior among adults diagnosed with schizophrenia, we first assessed bivariate associations between sociodemographic variables (age, sex, residence, educational level, economic status, and employment status) and treatment uptake using the Rao–Scott chi-square test, which accounts for complex survey design effects. Subsequently, these same variables were included in a survey-weighted logistic regression model to identify independent predictors of treatment-seeking. Given the small number of cases, backward elimination was employed to reduce the risk of overfitting, with variables sequentially removed based on the highest P-values until only those with statistically significant associations remained in the final model. All chi-square and regression analyses incorporated survey weights to account for the complex sampling design.

Results

Data enumerators visited 8928 HHs, resulting in a final sample of 7270 adults who completed all study measures, yielding an overall response rate of 90.4%. The mean age of participants was 38.3 years (SD = 14.5), and the average number of completed years of education was 5.9 (SD = 4.4). The sample comprised slightly more female participants (51.9%) and was predominantly married (83.5%), with 3.5% reporting a family history of mental illness. Table 1 summarizes the sociodemographic and familial characteristics of the study population including participants diagnosed with schizophrenia.

Table 1.

Sociodemographic and Familial Characteristics of Adult Participants in the Study

Category Full sample (n = 7270) Schizophrenia cases (n = 49)
Unweighted number (n) Unweighted percentage (%) Weighted percentage (%) Unweighted number (n) Weighted percentage (%)
Age (year) 18-39 3777 52.0 58.8 22 50.7
40-49 2530 34.8 29.1 19 34.0
≥ 60 963 13.2 12.1 8 15.3
Sex Male 3465 47.7 48.1 18 42.0
Female 3805 52.3 51.9 31 58.0
Residence Urban 3521 48.4 24.9 27 32.8
Rural 3749 51.6 75.1 22 67.2
Family type Nuclear 5790 79.6 69.7 41 81.6
Joint 1480 20.4 30.3 8 18.4
Educational level (year) >10 1092 15.0 14.8 9 23.4
6-10 2242 30.8 31.4 12 37.8
0-5 3936 54.1 53.8 28 38.9
Marital status Married 6506 89.5 83.5 40 66.9
Never married 449 6.2 11.8 3 11.2
Divorced/separated 45 0.6 0.8 2 17.0
Widow 270 3.7 3.9 4 4.9
Economic status Upper 2908 40.0 40.0 18 56.1
Middle 1454 20.0 21.5 10 12.1
Lower 2908 40.0 38.5 21 31.8
Mental illness in the family Present 292 4.0 3.5 8 16.0
Suicidal behavior among family members Present 86 1.2 1.3 2 7.5
Smoking among family members Present 2950 40.6 40.3 23 56.1

A total of 79 individuals were diagnosed with schizophrenia spectrum disorders. Among them, 6 cases were identified as brief psychotic disorder, 24 as delusional disorder, 2 as schizophreniform disorder, 45 as schizophrenia, and 2 as schizoaffective disorder. For the purposes of prevalence estimation and further analysis of associated sociodemographic factors, individuals diagnosed with schizophreniform disorder, schizophrenia, and schizoaffective disorder were grouped under a composite category labeled as “schizophrenia.” Table 2 presents the prevalence estimates across key subgroups. The overall prevalence of schizophrenia in the sample was 0.6% (95% CI: 0.4%-0.9%) with a design effect of 2.26, with no significant differences across age groups, sexes and place of residence.

Table 2.

Weighted Prevalence of Schizophrenia Among Adults by Age, Sex, and Residence

Age Male % (95% CI) Female % (95% CI) Urban % (95% CI) Rural % (95% CI) Overall % (95% CI)
18-39 0.5 (0.1-1.8) 0.5 (0.2-1.2) 0.5 (0.2-1.4) 0.5 (0.2-1.3) 0.5 (0.2-1.1)
40-59 0.3 (0.1-1.1) 1.0 (0.6-1.9) 0.9 (0.3-2.3) 0.6 (0.3-1.2) 0.7 (0.4-1.2)
≥60 0.9 (0.3-2.8) 0.5 (0.1-2.0) 2.3 (0.6-8.0) 0.4 (0.1-1.2) 0.7 (0.3-1.7)
18-99 0.5 (0.2-1.1) 0.6 (0.4-1.1) 0.8 (0.4-1.4) 0.5 (0.3-0.9) 0.6 (0.4-0.9)

Table 3 presents the results of logistic regression analyses examining sociodemographic and familial correlates of schizophrenia among adults in Bangladesh. Schizophrenia was significantly associated with being unemployed (aOR = 4.63, 95% CI: 1.57%-13.62%, P = .005), divorced or separated (aOR = 31.61, 95% CI: 4.33%-230.5%, P = .005), having a family history of mental illness (aOR = 3.30, 95% CI: 1.03%-10.56%, P = .044), and having a family history of suicidal behavior (aOR = 5.50, 95% CI: 1.07%-28.35, P = .041), while no significant associations were found with age, sex, residence, education, economic status, family structure, smoking in the family, or season of birth.

Table 3.

Univariate and Multivariate Regression Analyses of Sociodemographic and Familial Correlates of Schizophrenia Among Adults

Univariate regression Multivariate regression
OR (95% CI) P-value aOR (95% CI) P-value
Age (year)
 18-39 1 - -
 40-59 1.35 (0.52-3.53) 0.523 - -
 ≥60 1.46 (0.45-4.77) 0.884 - -
Sex
 Male 1 - -
 Female 1.28 (0.48-3.39) 0.613 - -
Residence
 Urban 1 - -
 Rural 0.67 (0.28-1.64) 0.390 - -
Family type
 Nuclear 1 - -
 Joint 0.51 (0.15-1.73) 0.286 - -
Educational level (year)
 >10 1 - -
 6-10 0.90 (0.22-3.69) 0.886 - -
 0-5 0.50 (0.16-1.60) 0.248 - -
Marital status
 Married 1 1
 Never married 1.18 (0.26-5.23) 0.751 0.83 (0.25-2.73) 0.831
 Divorce/separated 28.08 (4.24-185.6) 0.011a 31.61 (4.33-230.5) 0.005a
 Widowed 1.60 (0.43-5.88) 0.476 1.03 (0.23-4.69) 0.960
Economic status
 Upper 1
 Middle 0.38 (0.12-1.15) 0.089 0.35 (0.10-1.22) 0.101
 Lower 0.54 (0.21-1.35) 0.191 0.53 (0.20-1.44) 0.220
Employment status
 Employed 1 1
 Unemployed 3.98 (1.09-14.58) 0.036a 4.63 (1.57-13.62) 0.005a
Mental illness in the family
 Absent 1 1
 Present 5.26 (1.88-14.76) 0.002a 3.30 (1.03-10.558) 0.044a
Suicidal behavior in the family
 Absent 1 1
 Present 6.56 (1.45-29.57) 0.014a 5.50 (1.07-28.35) 0.041a
Smoking among family members
 Absent 1 - -
 Present 1.84 (0.77-4.39) 0.169 - -
Birth time
 Other seasons 1 - -
 Winter or spring birth 1.27 (0.46-3.47) 0.640 - -

Dashes (-) indicate variables excluded from multivariate analysis.

a

Significant test results (P < .05). Regression analyses were conducted using the full weighted sample (n = 7270) and account for clustering, stratification, and survey weights.

A total of 39.3% of patients with schizophrenia had received treatment. Table 4 presents the associations between sociodemographic characteristics and treatment-seeking behavior among adults with schizophrenia. In bivariate analysis, treatment uptake was significantly higher among younger adults (18-39 years) compared with those ≥40 years (60.4% vs. 17.5%; OR = 7.19, P = .030), and among rural residents compared with urban residents (52.6% vs. 12.0%; OR = 0.12 for urban, P = .010). In the multivariable logistic regression, both age and residence remained significant predictors: younger adults were more likely to seek treatment (OR = 8.09, 95% CI: 1.28%-50.89%, P = .027), while urban residents had markedly lower odds of treatment uptake (OR = 0.10, 95% CI: 0.01%-0.80%, P = .031). No other sociodemographic factors were significant.

Table 4.

Associations between Sociodemographic Factors and Treatment-Seeking Behavior Among Adults with Schizophrenia in Bangladesh

Category Unweighted n Treatment taken (weighted %) Treatment not taken (weighted %) Odds ratio Adjusted F (df1, df2) P-value
Age group (years) 18-39 22 60.4 39.6 7.19 4.97 (1, 48) 0.030
≥40 27 17.5 82.5
Sex Male 18 51.8 48.2 2.49 0.76 (1, 48) 0.387
Female 31 30.2 69.8
Residence Urban 27 12.0 88.0 0.12 7.24 (1, 48) 0.010
Rural 22 52.6 47.4
Educational level (years) >5 21 46.4 53.6 2.41 0.98 (1, 48) 0.325
≤5 28 26.4 73.6
Economic status Upper 18 50.3 49.7 3.01 1.42 (1, 48) 0.238
Lower and middle 31 25.1 74.9
Employment status Employed 43 39.7 60.3 1.13 0.01 (1, 48) 0.923
Unemployed 6 36.8 63.2

Weighted percentages are based on complex survey weights. Odds ratios are unadjusted estimates. P-values are derived from the Rao–Scott adjusted chi-square test (reported as adjusted F-statistic).

Discussion

This study is among the few conducted in LMICs that report the prevalence and correlates of schizophrenia using data from a nationally representative HH survey. We observed a prevalence of 0.6%, which is consistent with the commonly cited lifetime prevalence of ~1%, as well as with the range of 0.3%-0.7% reported in the DSM-5.14,15 Similarly, the Global Burden of Disease Study 2019 estimated the prevalence of schizophrenia in South Asia at 0.3%.16 Our findings are also in line with a meta-analysis of 29 studies conducted across diverse geographic regions, which reported 10th to 90th percentile prevalence estimates ranging from 0.18% to 1.16%.1 Importantly, that review emphasized an almost 8-fold variation in prevalence across cultures, highlighting the need for context-specific epidemiological evidence. In the regional context, the Indian National Mental Health Survey reported a current prevalence of 0.42% and a lifetime prevalence of 1.42%,17 further supporting the comparability of our results with those from neighboring countries.

In our analysis, schizophrenia was significantly associated with being divorced or separated, having a family history of mental illness or suicidal behavior, and being unemployed. Although the absolute numbers of divorced/separated individuals and those reporting family history of mental illness were small, the observed associations are consistent with prior research highlighting the adverse impact of schizophrenia on marital stability and the elevated risk among families with psychiatric vulnerability.18,19 Similarly, while we examined smoking behavior among family members, the lack of association in our sample should be interpreted cautiously given the absence of individual-level substance use data. As the number of participants in these categories was relatively small in our sample, these findings should be interpreted with caution and considered exploratory.

A well-established finding in schizophrenia research is the ~10% increased risk of the disorder among individuals born during winter and spring months (December/January to March/April) compared to those born at other times of the year.20,21 Several hypotheses have been proposed to explain this seasonal pattern, including increased maternal infections, reduced prenatal vitamin D levels, and second-trimester maternal-fetal chronobiological disruptions. However, our findings did not reveal any significant association between season of birth and schizophrenia. This aligns with research suggesting that increased schizophrenia risk due to seasonality is more pronounced at higher latitudes farther from the equator, which may not be as relevant in the context of Bangladesh.21

We did not find any significant association between schizophrenia and age, sex, or place of residence. This aligns with findings from a systematic review of 132 studies across 46 countries.1 Although not statistically significant, we observed a higher prevalence in females, particularly in the 40-59 age group, which coincides with the second peak of schizophrenia onset in women during the perimenopausal period.22 Higher female prevalence has also been reported in other studies and is sometimes attributed to higher suicide rates among males.23 Similarly, a non-significant trend of higher prevalence was noted among urban residents across all age groups. While some systematic reviews have linked urban living to increased schizophrenia risk due to factors such as migration, deprivation, pollution, infection, and social isolation,24 others have not found such associations.1

We initially hypothesized that schizophrenia would be more prevalent among rural residents, individuals from lower-income families, and those with lower educational attainment, given the association of these factors with maternal malnutrition, infections, obstetric complications, and prior evidence linking schizophrenia to socioeconomic disadvantage. Previous studies have shown that both parental low education and income, as well as community-level socioeconomic deprivation, are associated with increased schizophrenia risk.25 However, contrasting findings have also been reported, including increased risk among children from the highest social classes.26 Our findings align more closely with the social drift theory, which posits that schizophrenia and other mental illnesses lead to a downward shift in socioeconomic status rather than being caused by it.27 This is further supported by our observation that individuals with schizophrenia in the sample were more likely to be unemployed. Additionally, the mean (SD) years of education in our sample was 5.92 (4.43), consistent with the national average of Bangladesh. In a population where educational attainment is uniformly low, limited variability may hinder detection of a meaningful association between education level and schizophrenia, as it does not distinguish affected individuals from the general population.

The low treatment uptake observed in this study aligns with existing evidence from LMICs, where substantial gaps in mental health service utilization persist. The World Health Organization (WHO) has estimated that the treatment gap for mental disorders in LMICs may reach up to 85%,28 while the Indian National Mental Health Survey reported a treatment gap of 72% for schizophrenia.17 Poor help-seeking behavior, limited awareness of the need for treatment, and stigma are widely recognized barriers to accessing care. In addition, resource constraints and their unequal distribution in LMICs further widen the gap between those in need of treatment and those receiving it.17 In our analysis, the 2 factors associated with treatment uptake were younger age and rural residence; however, the wide confidence intervals indicate some instability in these estimates, likely due to the smaller sub-sample size. Younger patients may be more likely to seek treatment due to generational effects, higher health literacy, and being in their productive years. The higher treatment uptake observed in rural areas may reflect the influence of stigma and the difficulty of maintaining anonymity in closely knit communities. In contrast, lower treatment rates in urban areas may partly result from reduced uptake in disadvantaged neighborhoods, such as slums, as documented in previous studies from this region.17 However, some contrasting evidence exists, for example, studies from India show no significant differences in the duration of untreated psychosis between richer, more educated urban residents and poorer, less educated rural residents,29 suggesting that socioeconomic or geographic status alone may not determine care-seeking behavior. Cross-national data from 24 countries further suggest that perceived need and positive attitudes are more important determinants of treatment uptake than structural barriers.30

The strengths of this study include its nationwide HH coverage of a nationally representative population, face-to-face interviews conducted by psychiatrists, and the use of DSM-5 diagnostic criteria, ensuring high validity and reliability of diagnoses. This study contributes to the limited high-quality data available from LMICs. Additionally, the high response rate strengthens the representativeness of our findings. While some authors have raised concerns about the reliability of the SRQ-24, particularly its psychotic items in identifying psychosis, we note that schizophrenia often co-occurs with neurotic disorders, which can lead to positive SRQ responses that are then further validated through detailed psychiatric interviews.

Several limitations should be noted. First, individuals living in institutional settings and the homeless population groups known to have a higher prevalence of schizophrenia were excluded. As a result, the true prevalence may be underestimated. Second, the psychosis subscale includes only 4 items and has relatively low sensitivity, which may have led to under-detection of schizophrenia cases. Although diagnostic confirmation was conducted by psychiatrists, this limitation could still have contributed to underestimation of prevalence and influenced the observed associations. Third, because schizophrenia is a low-prevalence disorder, only 49 cases were identified in the survey. This small sample size reduced the precision of prevalence estimates and limited the robustness of observed associations, resulting in wide confidence intervals in regression analyses, especially for subgroups with very few participants (eg, divorced or separated individuals, families with mental illness, or families with suicidal behavior). Finally, the wide confidence intervals for treatment-seeking behavior reflect the small sample size and limited statistical power, and these findings should therefore be interpreted with caution.

Conclusion

This study provides robust evidence on the prevalence, correlates, and treatment patterns of schizophrenia in Bangladesh using data from a nationwide HH survey. While associations with marital status, unemployment, and family history of mental illness or suicide were observed, these findings should be interpreted with caution. Younger age and rural residence were associated with higher treatment-seeking behavior, contrasting with some existing evidence and highlighting the need for further research to explore the determinants of treatment uptake. By addressing the limited epidemiological evidence from LMICs, this study highlights the burden of schizophrenia and underscores the need to improve mental health literacy as well as to expand the accessibility and availability of mental health services.

Acknowledgments

Dr M. Mostafa Zaman, Dr Rizwanul Karim, Dr Ferdous Hakim, Rijwan Bhuiyan, Mst Khaleda Akter, Dr Mekhala Sarkar, Dr Farzana Rahman Dina, Dr Md Delwar Hossain, Dr Ramendra Kumar Singha, Dr Avra Das Bhowmik, Dr Mahbub-E-Kibria, Dr Masud Rana Sarker, Dr Muntasir Maruf, Dr Abdullah Al Mamun, Dr Mahbub Hasan, Dr Nadia Afroz, Dr Mamun Al Mujahid, Dr Shohedul Alam, Dr Md Anwar Hossain, Dr Taiyeb Ibna Zahangir, Dr Jhowhar Datta, Dr Shammi Akhtar, Dr Afroza Rahman Lopa, Dr Muntasir Khan, Kamruzzaman Mukul, Golam Rosul, Anwar Hossain, and the Office Staffs of NIMH.

Contributor Information

Ahsan Aziz Sarkar, Department of Psychiatry, Pabna Mental Hospital, Pabna 6602, Bangladesh.

Md Faruq Alam, Department of Psychiatry, Popular Medical College Hospital, Dhaka 1209, Bangladesh.

Mohammad Tariqul Alam, Department of Psychiatry, Sher-E-Bangla Medical College Hospital, Barisal 8200, Bangladesh.

Helal Uddin Ahmed, Department of Psychiatry, Faridpur Medical College, Faridpur 7800, Bangladesh.

Niaz Mohammad Khan, Department of Child, Adolescent & Family Psychiatry, National Institute of Mental Health, Dhaka 1207, Bangladesh.

Munshi Mohammad Jalal Uddin, Department of Psychiatry, National Institute of Neurosciences and Hospital, Dhaka 1207, Bangladesh.

Funding

This survey was supported by financial assistance from the Non-Communicable Disease Control division of the Directorate General of Health Services, operating under Bangladesh’s Ministry of Health & Family Welfare.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Ethical Approval

Ethical approval was obtained from the National Institute of Mental Health, Bangladesh and Bangladesh Medical Research Council (BMRC).

Data Availability

Supporting data may be obtained from the corresponding author upon justified request, subject to reasonable conditions.

References

  • 1. Saha  S, Chant  D, Welham  J, McGrath  J. A systematic review of the prevalence of schizophrenia. PLoS Med. 2005;2:e141. 10.1371/journal.pmed.0020141 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Hjorthøj  C, Stürup  AE, McGrath  JJ, Nordentoft  M. Years of potential life lost and life expectancy in schizophrenia: a systematic review and meta-analysis. Lancet Psychiatry. 2017;4:295–301. 10.1016/S2215-0366(17)30078-0 [DOI] [PubMed] [Google Scholar]
  • 3. Charlson  FJ, Ferrari  AJ, Santomauro  DF, et al.  Global epidemiology and burden of schizophrenia: findings from the global burden of disease study 2016. Schizophr Bull. 2018;44:1195–1203. 10.1093/schbul/sby058 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Velligan  DI, Rao  S. The epidemiology and global burden of schizophrenia. J Clin Psychiatry. 2023;84:MS21078COM5. 10.4088/JCP.MS21078COM5 [DOI] [PubMed] [Google Scholar]
  • 5. Davis  J, Eyre  H, Jacka  FN, et al.  A review of vulnerability and risks for schizophrenia: beyond the two hit hypothesis. Neurosci Biobehav Rev. 2016;65:185–194. 10.1016/j.neubiorev.2016.03.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Global Alliance for Improved Nutrition (GAIN), Institute of Public Health Nutrition (IPHN), International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B), United Nations Children’s Fund (UNICEF) . National Micronutrient Survey, Bangladesh 2019-2020. National Nutrition Service; 2022. Accessed April 5, 2025. https://nnsop.gov.bd/storage/files/file-2023-08-08-64d1dbdd5fb30.pdf [Google Scholar]
  • 7. Nahar  MZ, Zahangir  MS. The role of parental education and occupation on undernutrition among children under five in Bangladesh: a rural-urban comparison. PLoS One. 2024;19:e0307257. 10.1371/journal.pone.0307257 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. GBD 2019 Bangladesh Burden of Disease Collaborators . The burden of diseases and risk factors in Bangladesh, 1990-2019: a systematic analysis for the global burden of disease study 2019. Lancet Glob Health. 2023;11:e1931–e1942. 10.1016/S2214-109X(23)00432-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Ndetei  DM, Mutiso  V, Osborn  T. Moving away from the scarcity fallacy: three strategies to reduce the mental health treatment gap in LMICs. World Psychiatry. 2023;22:163–164. 10.1002/wps.21054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Alam  MF, Ahmed  HU, Alam  MT, et al.  National Mental Health Survey 2019. National Institute of Mental Health, 2021. [Google Scholar]
  • 11. Beusenberg  M, Orley  JH; World Health Organization, Division of Mental Health . A User’s Guide to the Self Reporting Questionnaire (SRQ) .  World Health Organization; 1994. Accessed August 31, 2025. https://iris.who.int/handle/10665/61113 [Google Scholar]
  • 12. Firoz  AHM, Karim  ME, Alam  MF, Rahman  AHM, Zaman  M. Prevalence, medical care, awareness and attitudes towards mental illness in Bangladesh. Bangladesh J Psychiatry. 2006;20:9–36. [Google Scholar]
  • 13. Chowdhury  AN, Brahma  A, Sanyal  D. The validation of the Bengali version of the self-rating questionnaire (SRQ). Indian J Clin Psychol. 2003;30:56–61. [Google Scholar]
  • 14. American Psychiatric Association , Ed. Diagnostic and Statistical Manual of Mental Disorders. 5th edn.  American Psychiatric Association, 2013. 10.1176/appi.books.9780890425596. [DOI] [Google Scholar]
  • 15. Marder  SR, Cannon  TD. Schizophrenia. N Engl J Med. 2019;381:1753–1761. 10.1056/NEJMra1808803 [DOI] [PubMed] [Google Scholar]
  • 16. Solmi  M, Seitidis  G, Mavridis  D, et al.  Incidence, prevalence, and global burden of schizophrenia - data, with critical appraisal, from the global burden of disease (GBD) 2019. Mol Psychiatry. 2023;28:5319–5327. 10.1038/s41380-023-02138-4 [DOI] [PubMed] [Google Scholar]
  • 17. Hegde  PR, Nirisha  LP, Basavarajappa  C, et al.  Schizophrenia spectrum disorders in India: a population-based study. Indian J Psychiatry. 2023;65:1223–1229. 10.4103/indianjpsychiatry.indianjpsychiatry_836_23 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Thara  R, Srinivasan  TN. Outcome of marriage in schizophrenia. Soc Psychiatry Psychiatr Epidemiol. 1997;32:416–420. 10.1007/BF00788182 [DOI] [PubMed] [Google Scholar]
  • 19. Chou  IJ, Kuo  CF, Huang  YS, et al.  Familial aggregation and heritability of schizophrenia and co-aggregation of psychiatric illnesses in affected families. Schizophr Bull. 2017;43:1070–1078. 10.1093/schbul/sbw159 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Torrey  EF, Miller  J, Rawlings  R, Yolken  RH. Seasonality of births in schizophrenia and bipolar disorder: a review of the literature. Schizophr Res. 1997;28:1–38. 10.1016/s0920-9964(97)00092-3 [DOI] [PubMed] [Google Scholar]
  • 21. Schwartz  PJ. Season of birth in schizophrenia: a maternal-fetal chronobiological hypothesis. Med Hypotheses. 2011;76:785–793. 10.1016/j.mehy.2011.02.019 [DOI] [PubMed] [Google Scholar]
  • 22. Li  X, Zhou  W, Yi  Z. A glimpse of gender differences in schizophrenia. Gen Psychiatr. 2022;35:e100823. 10.1136/gpsych-2022-100823 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Phillips  MR, Yang  G, Li  S, Li  Y. Suicide and the unique prevalence pattern of schizophrenia in mainland China: a retrospective observational study. Lancet.  2004;364:1062–1068. 10.1016/S0140-6736(04)17061-X [DOI] [PubMed] [Google Scholar]
  • 24. Vassos  E, Pedersen  CB, Murray  RM, Collier  DA, Lewis  CM. Meta-analysis of the association of urbanicity with schizophrenia. Schizophr Bull. 2012;38:1118–1123. 10.1093/schbul/sbs096 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Werner  S, Malaspina  D, Rabinowitz  J. Socioeconomic status at birth is associated with risk of schizophrenia: population-based multilevel study. Schizophr Bull. 2007;33:1373–1378. 10.1093/schbul/sbm032 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Mäkikyrö  T, Isohanni  M, Moring  J, et al.  Is a child's risk of early onset schizophrenia increased in the highest social class?  Schizophr Res. 1997;23:245–252. 10.1016/s0920-9964(96)00119-3 [DOI] [PubMed] [Google Scholar]
  • 27. Dunham  HW. Community and Schizophrenia: An Epidemiological Analysis. Wayne State University Press, 1965. [Google Scholar]
  • 28. World Health Organization . mhGAP Intervention Guide for Mental, Neurological and Substance Use Disorders in Non-Specialized Health Settings: Mental Health Gap Action Programme (mhGAP). Version 2.0. World Health Organization, 2016. [PubMed] [Google Scholar]
  • 29. Thirthalli  J, Reddy  KS, Kare  M, Das  S, Gangadhar  BN. Rural-urban differences in accessing mental health treatment in patients with psychosis. Int J Soc Psychiatry. 2017;63:694–698. 10.1177/0020764017730849 [DOI] [PubMed] [Google Scholar]
  • 30. Andrade  LH, Alonso  J, Mneimneh  Z, et al.  Barriers to mental health treatment: results from the WHO World Mental Health surveys. Psychol Med. 2014;44:1303–1317. 10.1017/S0033291713001943 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Supporting data may be obtained from the corresponding author upon justified request, subject to reasonable conditions.


Articles from Schizophrenia Bulletin are provided here courtesy of Oxford University Press

RESOURCES