Schizophrenia spectrum disorders in India: A population-based study

Abstract

Introduction:

Schizophrenia spectrum disorders represent a significant global health concern, contributing significantly to the global burden of disease. The National Mental Health Survey (NMHS) of India, conducted between 2015 and 2016, investigated the prevalence and epidemiological correlates of schizophrenia spectrum disorders in India.

Materials and Methods:

The NMHS conducted a population-based cross-sectional study in 12 Indian states from 2015 to 2016, encompassing 34,802 adults. The overall study design of the NMHS was a multistage, stratified, random cluster sampling technique, incorporating random selection based on probability proportion to size at each stage. The Mini-International Neuropsychiatric Interview 6.0 was used for psychiatric diagnoses, disability was assessed using Sheehan's disability scale, and the illness-related socioeconomic impact was assessed using a questionnaire based on the World Health Organization Disability Assessment Schedule 2.0. Firth penalized logistic regression was employed to understand the correlates of current schizophrenia spectrum disorder.

Results:

The study found a lifetime prevalence of schizophrenia spectrum disorders at 1.41%, with a current prevalence of 0.42%. A substantial treatment gap of 72% existed for current cases, rising to 83.3% in urban non-metro areas. The penalized logistic regression revealed that the age group category of 30–49 years, unemployed status, and lower education level had higher odds of association with schizophrenia spectrum disorders.

Conclusion:

The primary finding of this study is a lifetime prevalence of 1.41%, a current prevalence of 0.42%, and a substantial treatment gap of 72%. Addressing this treatment gap and holistic intervention is crucial for reducing the socioeconomic impact of this disorder. Strengthening the National Mental Health Program and implementing community-based rehabilitation are essential first steps in this direction.

INTRODUCTION

Schizophrenia spectrum disorders are considered severe mental illnesses and are one of the leading contributors to the global burden of disease.[1] It affects individuals across various sociodemographic strata worldwide. Globally, the pooled one-year and lifetime prevalence of schizophrenia are approximately 0.34% and 0.55%, respectively,[2] with wide variations reported in published literature.[3] Studies that have included all psychotic spectrum disorders have reported higher prevalence rates, varying between 0.5% and 2.5%.[4,5,6,7]

In India, the prevalence of schizophrenia is estimated to be 0.3% as per the Global Burden of Diseases (GBD) study of 1990–2017. Schizophrenia accounts for 10% of the total disability-adjusted life years (DALYs) due to mental morbidity in India.[8] In a household study done in the Kerala state of India, the lifetime prevalence of psychosis and schizophrenia was 0.71% and 0.23%, respectively.[9] An earlier epidemiological study in urban Chennai, a city in India, reported the prevalence of schizophrenia as 0.25%.[10] Prior studies have not adequately addressed the representativeness of the selected sample in relation to the general population. Wide variations in Indian epidemiological studies are attributed to factors such as the nature of psychiatric disorders, diagnostic methods, and case definitions. Systematic underreporting, recall bias, and reliance on single informants further contribute, along with variations in treatment, screening instruments, clinical interviews, and sampling procedures leading to potential bias. National Mental Health Survey (NMHS) of India aimed to overcome these methodological issues, ensuring a more comprehensive and representative assessment of the prevalence of schizophrenia spectrum disorders in India.[11]

International Classification of Diseases-11 (ICD-11) and Diagnostic and Statistical Manual of Mental Disorder-5 (DSM-5) have adopted the concept of schizophrenia spectrum disorders,[12] including schizophrenia and other primary psychotic disorders while excluding secondary causes of psychotic disorders.

Data from a nationally representative sample are essential for policy-making concerning longstanding disabling illnesses like schizophrenia spectrum disorder. It helps strengthen the National Mental Health Programme (NMHP) of India and aims to reduce the burden of mental morbidity, including schizophrenia spectrum disorder. The findings will also help in forming a framework for planning, programming, financing, and delivering mental health programs in India. The paper presents an estimate of the prevalence and epidemiological correlates of schizophrenia spectrum disorders, analyzing a nationally representative survey to identify the treatment gap, disability, and socioeconomic impact of schizophrenia spectrum disorders.

MATERIALS AND METHODS

The NMHS of India was a multisite population-based cross-sectional study between 2015 and 2016 across 12 states of India. It employed a standardized and reliable methodology described in detail elsewhere.[13] The measurement scales used in the NMHS study demonstrated good validity and reliability, with detailed explanations provided in a prior publication.[13] Only a brief explanation of the NMHS methodology has been presented in this paper. The study included adults (age >18 years) who were selected by approaching the selected household in the nationally representative population. All available adults in each household were approached, and those consenting were included as respondents. A total of 39,532 adults were contacted, and 34,802 were interviewed (88% response rate).

The interview included a standardized study instrument, with Mini-International Neuropsychiatric Interview (MINI; version 6.0) used to diagnose psychiatric disorders. MINI is available in multiple Indian languages, allows for single-stage interviews for diagnosis, and is compatible with the International Classification of Diseases. After translating the MINI to local languages by using a standard procedure, systematic and robust training was conducted at multiple levels to ensure reliable data collection. The interview was conducted by trained field data collectors. Disability was assessed using Sheehan's disability scale across work, social, and family functioning, and information regarding health treatment, care-seeking patterns, and socioeconomic impacts was collected using a validated questionnaire. Assessment of illness-related socioeconomic impact, adapted from World Health Organization Disability Assessment Schedule version 2.0 (WHO-DAS), encompassed seven questions covering self-reported challenges, their 30-day duration, effects on daily activities, financial implications, and any disruptions to family, social, or leisure engagements due to the illness.[14] The study was approved by the National Institute of Mental Health and Neurosciences (NIMHANS) ethics committee and relevant authorities in the participating regions.

Statistical analysis

Schizophrenia spectrum disorder was assessed as a group rather than as individual categories of psychotic illness, as per the categorization of MINI. Both current and lifetime prevalence were computed for schizophrenia spectrum disorder. Weighted prevalence was calculated to estimate the actual burden in the representative population for different sociodemographic factors. The sociodemographic characteristics included gender, age groups, education level, marital status, occupational status, residence, and income quintiles.

In this study, the utilization of a binomial outcome, distinguishing between individuals with schizophrenia spectrum disorder and those without, was imperative for deriving odds ratios through logistic regression analysis. To address concerns related to statistical separation and monotone likelihood, the research employed Firth bias-reduced logistic regression. The rationale for its use and other details are mentioned elsewhere.[15] This statistical technique effectively mitigates the initial bias in maximum likelihood estimates of beta, particularly when dealing with smaller sample sizes, ensuring that the estimates remain stable and consistent.

Furthermore, the study employed descriptive statistics to assess disability, average illness duration, and the time gap between symptom onset and the initiation of treatment. All statistical procedures were conducted using the integrated R software suite, which offers a comprehensive set of tools and capabilities for analysis.[16]

The estimates were reported using a 95% confidence interval (CI) with significance defined for a P value of <0.05.

RESULTS

Overall, 429 persons out of 34,802 screened positive for lifetime schizophrenia spectrum disorders, and 149 persons screened positive for cross-sectional prevalence of schizophrenia spectrum disorder. Out of these 149, a total of 94 persons had schizophrenia spectrum disorder after exclusion of psychiatric comorbidities. Out of these 94 persons, the disability was calculated based on 74 responses after the exclusion of missing data (1 additional data point was missing for data on work life).

In the NMHS, individuals aged 18–29 constituted the largest group (34%) compared to other age categories. In addition, 68.8% of respondents hailed from rural areas, with females comprising 52.3% of the respondents, and 74.7% were married. Respondents who were illiterate accounted for 24.3% of the sample. The predominant occupation among respondents was household duties, comprising 30% of the study sample. Other sociodemographic characteristics of the sample are depicted in Table 1.

Table 1.

Prevalence of schizophrenia spectrum disorders for different sociodemographic characteristics among adults in India: NMHS 2015-16

	Prevalence of schizophrenia spectrum disorder
	Lifetime (95% CI)	Current (95% CI)
Total	1.41 (1.39–1.43)	0.42 (0.41–0.44)
Gender
Male	1.49 (1.46–1.52)	0.46 (0.45–0.48)
Female	1.33 (1.31–1.36)	0.39 (0.37–0.40)
Age group
18–29	1.39 (1.36–1.43)	0.38 (0.37–0.40)
30–39	1.33 (1.28–1.37)	0.47 (0.44–0.50)
40–49	1.61 (1.56–1.66)	0.59 (0.56–0.63)
50–59	1.31 (1.26–1.77)	0.31 (0.28–0.33)
60 and above	1.43 (1.37–1.48)	0.36 (0.33–0.39)
Education
Illiterate	1.09 (1.06–1.13)	0.38 (0.36–0.41)
Primary	2.27 (2.12–2.33)	0.67 (0.64–0.70)
Secondary	1.65 (1.60–1.71)	0.45 (0.42–0.48)
High School	0.93 (0.89–0.98)	0.27 (0.25–0.29)
Preuniversity/Vocational	0.82 (0.77–0.86)	0.20 (0.17–0.22)
Graduate/Post Graduate/Professional	1.54 (1.47–1.60)	0.50 (0.46–0.54)
Marital Status
Never Married	1.40 (1.35–1.44)	0.45 (0.42–0.48)
Married	1.43 (1.41–1.46)	0.42 (0.40–0.43)
Widowed/Divorced/Separated	1.28 91.20–1.37)	0.47 (0.43–0.53)
Occupational Status
Working	1.27 (1.25–1.30)	0.39 (0.37–0.40)
Not working	1.50 (1.47–1.53)	0.45 (0.44–0.47)
Residence
Rural	1.05 (1.03–1.08)	0.37 (0.36–0.39)
Urban Non-Metro	0.92 (0.88–0.97)	0.36 (0.33–0.39)
Urban Metro	3.05 (2.98–3.12)	0.66 (0.62–0.69)
Income quintile
Lowest	1.55 (1.50–1.60)	0.70 (0.67–0.74)
Second	1.24 (1.20–1.29)	0.51 (0.48–0.54)
Middle	1.36 (1.31–1.40)	0.36 (0.34–0.39)
Fourth	1.31 (1.27–1.36)	0.28 (0.26–0.30)
Highest	1.58 (1.54–1.63)	0.31 (0.29–0.33)

NMHS=National Mental Health Survey

The overall weighted prevalence of lifetime schizophrenia spectrum disorder was 1.41% (95% CI: 1.39–1.43). There was a slight variation across the sociodemographic groups. Lifetime prevalence of schizophrenia was relatively higher in the 40–49 years age group (1.61%), among those with primary education (2.27%), secondary education (1.65%), and for those residing in cities with a population of >1 million (3.05%). The current weighted prevalence of schizophrenia spectrum disorder was 0.42% (95% CI: 0.41–0.44), with a higher prevalence in the 40–49 years age group (0.59%), among those with primary education (0.7%), residing in cities with a population of >1 million (0.66%), and belonging to the lowest income quintile (0.7%) [Table 1].

Region-wise, the prevalence of schizophrenia spectrum disorder was higher in states belonging to the Eastern region (Jharkhand and West Bengal), both for a lifetime (2.2%) and current (0.5%) prevalence estimates. Conversely, states in the Northern region (Punjab and Uttar Pradesh) had a lower lifetime (0.3%) and current (0.2%) prevalence estimates of schizophrenia spectrum disorder.

Firth penalized logistic regression was used to estimate the association of sociodemographic variables with schizophrenia spectrum disorder. The results are depicted in Table 2. In comparison to the 18–29 years age group, the population aged 30–39 years and 40–49 years had 72% and 90% higher likelihood of having schizophrenia spectrum disorder, respectively. Those who were not employed had 71% higher odds of having schizophrenia spectrum disorder than those who were working. In addition, individuals with education up to primary school had 1.7 times higher odds of having schizophrenia spectrum disorder in comparison to professionals. Although not statistically significant, it is important to highlight that the male gender exhibited a trend toward a higher prevalence of schizophrenia.

Table 2.

Firth penalized logistic regression analysis for factors associated with current schizophrenia spectrum disorders among adults in India: NMHS 2015-16

	Coefficients	Std. Error (Coe)	Chi-Square	Adjusted OR	95%CI for OR	P
Female(ref)
Male	0.36	0.20	3.01	1.43	0.95–2.15	0.08
18–29 [ref]
30–39	0.54	0.26	4.20	1.72	1.02–2.92	0.04
40–49	0.64	0.28	5.21	1.89	1.09–3.30	0.02
50–59	0.30	0.32	8.25	1.34	0.70–2.52	0.36
>60	0.15	0.31	2.22	1.16	0.62–2.18	0.64
Graduation and above(ref)
Illiterate	0.54	0.23	5.27	1.71	1.08–2.72	0.02
Primary school	0.10	0.27	1.30	1.10	0.64–1.87	0.72
Middle school	−0.30	0.29	1.05	0.74	0.41–1.31	0.31
High school	−0.69	0.40	3.22	0.50	0.21–1.06	0.07
Preuniversity/vocational courses	−0.22	0.33	4.29	0.80	0.40–1.53	0.51
Working(ref)
Not working	0.54	0.20	6.87	1.71	1.14–2.57	0.01
Never married(ref)
Married	−0.27	0.27	8.89	0.77	0.45–1.34	0.35
Widower/divorced/separated	−0.001	0.43	9.59	1.00	0.41–2.32	1.00
Rural(ref)
Cities with a population of <1 million	−0.10	0.22	1.81	0.91	0.57–1.40	0.67
Cities with a population of >1 million	0.36	0.23	2.29	1.43	0.89–2.22	0.13

NMHS=National Mental Health Survey

Out of 149 individuals with current schizophrenia, 55 had screened positive for a comorbid mental disorder (including substance use disorders). Among these individuals, 9.4% had depression, and 8.1% had anxiety. In addition, alcohol use disorder and tobacco use disorder were present in 10.7% and 27.5% of the individuals with current schizophrenia, respectively. Among the neurotic disorders, phobic anxiety disorder was present in 8.1%, panic disorder in 3.4%, and PTSD in 1.3% of those with a schizophrenia spectrum disorder. A lifetime diagnosis of OCD was present in 6.7% of those with a current schizophrenia spectrum disorder.

Assessing disability in the last 1 month for those with current schizophrenia disorder without mental comorbidity, it was observed that in persons with schizophrenia spectrum disorder, disability was present in more than half of the individuals in the domains of work (53.4%), social life (59.5%), and family life (59.5%) across the range of mild-to-extreme severity. The details of the disability are given in Table 3, and the socioeconomic impact of schizophrenia spectrum disorders is summarized in Table 4.

Table 3.

Disability associated with current schizophrenia spectrum disorders among adults in India: NMHS 2015-16

Severity of disability	n*	No disability (%)	Any disability (%)	Mild (%)	Moderate (%)	Marked (%)	Extreme (%)
Work life	73	34 (46.60)	39 (53.40)	12 (16.60)	8 (11)	8 (11)	11 (15.10)
Social life	74	30 (40.50)	44 (59.50)	12 (16.20)	12 (16.20)	11 (14.90)	9 (12.20)
Family life	74	30 (40.50)	44 (59.50)	12 (16.20)	11 (14.90)	11 (14.90)	10 (13.50)

Disability was assessed using Sheehan’s disability scale.[36] *Missing data (missing n=20 for social and family life; missing n=21 for work life) has been excluded from the cohort of persons with current schizophrenia without other psychiatric comorbidities, i.e., 94. NMHS=National Mental Health Survey

Table 4.

Socio-economic impact of current Schizophrenia spectrum disorders among adults in India: NMHS 2015-16

	n (%); or median value
1. Difficulties with activities of daily life (n=73)
1.1. Could do activities as usual	41 (56.2%)
1.2. Could do activities but not everything	11 (15.10%)
1.3. Could do only some activities	9 (12.3%)
1.4. Extreme difficulties or inability to do most activities	12 (16.4%)
2. Median number of days with difficulty to carry out usual activities in the past 30 days (n=37)	15
3. Median number of days with total inability to carry out usual activities in the past 30 days (n=25)	15
4. Median number of days with reduction of usual activities in the past 30 days (n=19)	10
5. Median number of days family members were not able to go to work in the past 3 months for care of patient (n=16)	10
6. Median number of days family leisure or social activities was missed (n=18)	17.5
7. Median monthly expenditure in INR (n=17)	1000

INR=Indian Rupees, NMHS=National Mental Health Survey

These disabilities persisted for a median period of 15 days in the previous month, during which individuals were entirely unable to carry out any usual activities. Family members caring for persons with schizophrenia were absent from work due to the caregiving requirements for a median of 10 days in the last 3 months. Approximately INR 1000 per month was spent by family members for the care of the persons with current schizophrenia spectrum disorder.

The treatment gap was 72% for the current schizophrenia spectrum disorder, with urban non-metro having a higher treatment gap of 83.3%. The median duration of the illness was 48 months, with a median time interval between onset and consultation of 3.5 months. Among those who were currently on treatment, the median duration of being on treatment was 29 months.

DISCUSSION

The current and lifetime prevalence of schizophrenia spectrum disorder in India were 0.41% and 1.41%, respectively. The finding is comparable to the global estimates of schizophrenia spectrum disorder.[4,5,6,7] A systematic review of schizophrenia-related epidemiological studies has reported the best-estimated rates for 1 year and lifetime prevalence of schizophrenia spectrum disorder to be 0.6 and 1.45%, respectively.[2] The 1-year prevalence was slightly higher in the systematic review. This could be partly explained by including studies that sampled both households and institutions in the systematic review and because the assessment period was 1 year. The pooled 1-year rate from studies that assessed only household population reported was 0.43%, similar to our study findings. Studies that have looked at only schizophrenia, excluding other schizophrenia spectrum disorders, have reported lower rates. The systematic review said one-year and lifetime prevalence of schizophrenia to be 0.34% and 0.55%, respectively. The GBD estimate reported a prevalence of 0.28% globally for schizophrenia.[1] The prevalence varied across sociodemographic characteristics, with middle-aged adults, persons with lower education, and belonging to urban cities having a higher prevalence.

In India, the GBD estimate showed a point prevalence of 0.25% for schizophrenia.[8] Epidemiological studies from the southern states reported the prevalence of schizophrenia and schizophrenia spectrum disorder to be 0.29% and 0.71%, respectively.[9] Another study conducted in a large city reported the prevalence of current schizophrenia and schizophrenia spectrum disorder to be 0.39% and 0.56%, respectively.[10] The higher prevalence of schizophrenia spectrum disorder in these studies might be because of the varied definitions of schizophrenia spectrum disorder, the age group studied, diagnostic tools used, and the study setting, in addition to other regional variations.

Variation has been documented in the prevalence and across sociodemographic characteristics. Our finding of a significant association of schizophrenia with the age group around 40 years is consistent with reports from GBD, which also reported increased prevalence at around 40 years, with a decline in older age groups.[1] Consistent with previous research,[17] this study also observed no statistically significant difference between males and females in schizophrenia prevalence, although there was a trend toward higher prevalence among males. These findings underscore the need for further investigation into potential gender-related disparities in schizophrenia prevalence.

Our analysis also showed that schizophrenia spectrum disorder is associated with high unemployment, which is consistent with findings worldwide.[18] It has been established that occupational functioning is associated with a better quality of life.[19] Therefore, the state needs to evolve policies and provide increased resources to this marginalized group. Reserving jobs for individuals with mental illness is an initiative taken by the government of India in recent years. This step is in the right direction to help persons with severe mental illness. There is a well-established association between urbanization and the increased likelihood of schizophrenia in published literature.[20] Our study also highlights similar findings in that schizophrenia spectrum disorder is more prevalent in residents in the urban city, although it was not statistically significant.

Schizophrenia is associated with significant disability affecting all domains of life.[21] Persons with schizophrenia tend to have difficulty in relationships and work performance. DALY is used in the GBD study to assess the burden of diseases. DALY is a measure that combines data on mortality (years of life lost due to premature death = YLL) and disability (years lived with disability = YLD). As per GBD 2016, schizophrenia is the 12^th most disabling disorder among 310 diseases and injuries globally in 2016. It contributes 13.4 (95% uncertainty interval: 9.9–16.7) million YLDs to the burden of disease globally, equivalent to 1.7% of total YLDs in 2016.[1] In GBD 2017, schizophrenia accounted for 0.51% of all-cause DALYs. In our study, nearly one-third of the persons with current schizophrenia spectrum disorder reported significant difficulty in activities of daily living. This is of specific interest given that the treatment gap is as high as 72% for schizophrenia spectrum disorder in India, consistent with findings from low income countries.[22] Poor help-seeking behavior, low perceived need, and stigma were the major barriers to mental health treatment globally.[23] In addition, in low-income countries like India, lack of access and unequal distribution of resources contribute to the high treatment gap.[24,25] Community-based mental health becomes important to overcome the high treatment gap. The NMHP, launched in 1982, has its role in decentralizing mental health care through District Mental Health Programme (DMHP).[26] The program's effectiveness has not been consistent across the states because of the shortage of human resources, limited funding, and poor motivation among healthcare workers to provide service at all levels.[27] Consistent efforts at the primary level have shown its ability to cater to the majority.[28] Strengthening primary care to include mental health and utilizing community health workers to provide mental health care is the way forward to reduce the treatment gap.[29]

Schizophrenia spectrum disorder is associated with a significant economic burden because of the cost of care and loss due to disability.[17] Literature from published studies indicated that the financial costs associated with schizophrenia are greater than those of other physical and psychiatric illnesses.[30] Earlier Indian studies done in 2015 noted the costs incurred for the treatment of schizophrenia to be around INR 4005 per month, including direct and indirect costs.[31] In the current study, the monthly expenditure for care of a person with schizophrenia was INR 1000 per month, which is a lower figure. However, in the present study, indirect costs such as loss of wage for patients and families were not included. In addition, the cost of care was calculated only for those with an exclusive diagnosis of schizophrenia spectrum disorder without mental comorbidities in our analysis. With household income averaging INR 9000 per month in the NMHS-surveyed households, the cost of care for schizophrenia spectrum disorder is substantial, adding a significant economic burden to the families. States should, in addition to providing decentralizing care, focus on financially enabling the patient and family. Although disability benefit schemes are available, they are not fully utilized by the beneficiaries.[32] A more proactive role needs to be taken by the state to ensure that the financial aid reaches the deserved. With the support of the public health system, models such as Community Based Rehabilitation (CBR) can be set up. CBR has been found to be cost-effective, significantly reducing the financial burden in families with persons with mental illness.[33] Initiatives involving telepsychiatry have also found their usefulness in delivering a cost-effective, feasible, and well-accepted model of service, which can be explored in future research.[34,35]

Strengths and limitations

The general limitation of the NMHS methodology has been discussed earlier.[13] The study included all the patients with schizophrenia spectrum disorder (ICD 10, F20–F29) and not just schizophrenia. The course and disability associated with schizophrenia spectrum disorder vary considerably. In addition, over half of the patients with schizophrenia spectrum disorder had a comorbid psychiatric illness. To address it, we studied the disability, treatment gap, and impact on those without comorbid psychiatric illness. The strength of the study includes the robust methodology, which included training field teams and ensuring inter-rater reliability.

CONCLUSION

The findings of the NMHS offer compelling evidence to underscore the disabling nature of schizophrenia spectrum disorders, aligning with prevalence rates reported in existing studies and systematic reviews. Notably, this condition is more prevalent among individuals in the working-age group, and the resulting disability impacts not only those directly affected but also extends to their dependent family members.

These findings underscore the critical importance of reinforcing the NMHP to ensure that individuals with schizophrenia and their caregivers have access to essential services and actively utilize them. It is clear that addressing the needs of this vulnerable population and their families requires concerted efforts from the government and other relevant stakeholders. This collaborative approach is essential to provide the necessary support and resources, ultimately improving the quality of life for those affected by schizophrenia spectrum disorders and their families.

Financial support and sponsorship

The National Mental Health Survey (NMHS) was funded by the Ministry of Health and Family Welfare, Government of India, and was implemented and coordinated by the National Institute of Mental Health and Neurosciences (NIMHANS), Bengaluru, INDIA in collaboration with state partners. NMHS phase 1 (2015-16) was undertaken in 12 states of India across six regions and interviewed 39,532 individuals (http://indianmhs.nimhans.ac.in). The funder had no role in the implementation, data acquisition, data analysis and interpretation, and write-up of the paper.

Conflicts of interest

There are no conflicts of interest.