A systematic review and meta-analysis of prevalence of seven psychiatric disorders in India

Abstract

Background:

After the National Mental Health Survey in 2016, multiple individual studies showed inconsistencies in the prevalence rates of psychiatric disorders in India. We performed a meta-analysis to estimate an up-to-date pooled estimate of the prevalence of depression, alcohol use disorder (AUD), anxiety disorder (AD), intellectual disability, suicidal attempt/death, autism, and bipolar disorder (BD) in India.

Materials and Methods:

We performed a systematic bibliographic search in Pub Med, Global Health Data Exchange (GHDx), and Google Scholar, along with a manual search for peer-reviewed epidemiological studies reporting the prevalence of depression, AUD, AD, MR, suicidal attempt/death, autism, and BD in India from January 1980 till March 2022. Adopting a random-effects model, we performed the meta-analysis using “MetaXL” software.

Results:

A total of 79 studies were included: depression (n = 28), AUD (n = 14), AD (n = 12), intellectual disability (n = 8), suicidal attempt/death (n = 7), autism (n = 6) and BD (n = 4). The pooled prevalence of depression and AUD was 12.4% (95% CI 9.4–15.9) (P < 0.001, I² = 100%) and 21.5% (95% CI 14.1–30.0) (P < 0.001, I² = 100%), respectively. AD, intellectual disability and suicidal attempt/death showed a prevalence of 11.6% (95% CI 8.1–15.7) (P < 0.001, I² = 99%), 1% (95% CI 0.5–1.6) (P < 0.001, I² = 98%) and 0.5% (95% CI 0.3–0.8) (P < 0.001, I² = 100%), respectively. The meta-analysis in autism and BD showed pooled prevalence of 0.3% (95% CI 0.1–0.6) (P < 0.001, I² = 96%) and 0.3% (95% CI 0.2–0.4) (P < 0.001, I² = 78%), respectively. Subgroup analysis showed an increased prevalence of AD in the urban [24.3% (95% CI 3.7–52.9)] and younger [16.7% (95% CI 5.1–32.7)] population. The prevalence of depression and AD increased during the last two decades on decadal prevalence analysis.

Discussion:

The findings could be used for appropriate policy measures and guiding subsequent national mental health surveys.

BACKGROUND

Nowadays, governments prioritize mental health in their national programs and health policies.[1] In India, after the National Mental Health Programme (NMHP) in 1982, many epidemiological studies have been conducted to estimate the prevalence of psychiatric disorders. These studies reported that the prevalence of psychiatric disorders in India varies from 1% to 37% of the population.[2,3,4,5] These wide variations in prevalence rates are due to multiple methodological issues, using different screening instruments, case definitions, and systematic under-reporting.[5] Large-scale mental health surveys like the World Mental Health Survey (2003), the Indian Council of Medical Research-Department of Science and Technology (ICMR-DST) study on severe mental morbidity (2005), and the National Mental Health Survey (NMHS) (2016) have been conducted in India.[5] The data from these surveys have helped us understand the epidemiology of mental health and the growth of NMHP in India.[6]

The data from NMHS has shown a high prevalence of depression, alcohol use disorder, anxiety disorder, intellectual disability, and autism in the country.[7] Since NMHS in 2016, multiple population-based prevalence studies have been conducted in different regions in India, which have reported highly variable prevalence rates compared to previous large-scale mental health surveys. For example, the prevalence of depression is reported from 2.4% to 54%,[8,9] and that of autism ranges from 0.05% to 1.1%.[10,11] Given varying prevalence rates reported in different regional epidemiological studies, the generalization of prevalence rates becomes difficult for policy planning and management of mental healthcare in the country. In such a scenario, up-to-date pooled estimates are a need of the hour to estimate the burden of psychiatric illnesses.[12]

Meta-analysis is a quantitative technique to combine the results of selected individual studies to calculate the pooled estimates.[12] There is a limited number of prevalence meta-analysis studies of psychiatric disorders in India. In the present study, we will use meta-analytic methods to estimate the up-to-date prevalence rates of seven psychiatric disorders in India, namely depression, alcohol use disorder, anxiety disorder, intellectual disability, suicidal attempt/death, autism, and bipolar disorder, using prevalence data from all studies including large mental health surveys.

MATERIALS AND METHODS

Protocol and Registration

The study protocol was registered in PROSPERO (CRD#42022306539), an international database of prospectively registered systematic reviews.[13] We conducted this study as per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines[14] [Supplementary Table 1]. The institutional human ethics committee approved to conduct this study.

Supplementary Table 1.

Characteristics of studies of depression (n=28), alcohol use disorder (n=14), anxiety disorder (n=12), intellectual disability (n=8), suicidal attempt/death (n=7), autism (n=6), and bipolar disorders (n=4) included for meta-analysis

ID	Author	Place of study	Study design	Population screened	Urban/Rural
Depression
1.	Ahmad A et al., 2007	Aligarh, Uttar Pradesh	School-based	390 (males)	Both
0.2	Biswas SS et al., 2009	Vellore, Tamil Nadu	Population-based survey	204	Urban
3.	Poongothai S et al., 2009	Chennai, Tamil Nadu	Population-based survey	25455	Urban
4.	Rajkumar AP et al., 2009	Vellore, Tamil Nadu	Population-based survey	1000	Rural
5.	Kessler RC et al., 2010	Puducherry, UT	Stratified multistage clustered area probability sample of household	5007	Both
6.	Sahoo S et al., 2010	Ranchi, Jharkhand	College-based	405	Urban
7.	Bansal PD et al., 2011	Bathinda, Punjab	Cross-sectional, school-based	982	Urban
8.	Seby K et al., 2011	Pune, Maharashtra	Population-based survey	202	Urban
9.	Deswal BS et al., 2012	Pune, Maharashtra	Population-based survey	3023	Urban
10.	Sarkar S et al., 2012	Ranchi, Jharkhand	Cross-sectional, school-based	1851	Both
11.	Patil RN et al., 2013	Mumbai, Maharashtra	Cross-sectional study	257	Urban
12.	Jonas JB et al., 2014	Nagpur, Maharashtra	Cross-sectional study	4698	Rural
13.	Rao TSS et al., 2014	Suttur, Karnataka	Population-based survey	3033	Rural
14.	Mathias K et al., 2015	Dehradun, Uttarakhand	Population-based survey	960	Both
15.	Nakulan A et al., 2015	Thrissur, Kerala	Population-based survey	220	Rural
16.	Sengupta P et al., 2015	Ludhiana, Punjab	Population-based survey	3038	Both
17.	Albers HM et al., 2016	Hyderabad, Telangana	Cross-sectional study on migrant workers	884	Urban
18.	Guerra M et al., 2016	Chennai, Tamil Nadu	Population-based survey	1003	Urban
19.	Guerra M et al., 2016	Vellore, Tamil Nadu	Population-based survey	999	Rural
20.	NMHS, 2016	12 states across India	Multistage, stratified, random cluster study	34802	Both
21.	Behera P et al., 2016	Ballabhgarh, Haryana	Community-based cross-sectional	395	Rural
22.	Shidhaye R et al., 2016	Amravati, Maharashtra	Community-based cross-sectional	1456	Rural
23.	Singhal M et al., 2016	Bengaluru, Karnataka	School-based	800	Urban
24.	Jha KK et al., 2017	Patna, Bihar	School-based	1412	Both
25.	Kallakuri S et al., 2018	West Godavari district, Andhra Pradesh	Population-based survey	22377	Rural
26.	Shaikh BM et al., 2018	Pune and Nanded, Maharashtra	School-based	461	Both
27.	Lotfaliany M et al., 2019	Vadu, Gujarat	Population-based survey	11230	Both
28.	Nayak S et al., 2019	Berhampur, Odisha	Population-based survey	244	Urban
Alcohol use disorder
1.	Chavan BS et al., 2007	Chandigarh, UT	Community-based cross-sectional	2992	Both
2.	Ghosh S et al., 2012	Kolkata, West Bengal	Community-based cross-sectional	228	Urban
3.	Ganesh KS et al., 2013	Villupuram, Tamil Nadu	Community-based cross-sectional	946	Rural
4.	Kim S et al., 2013	Vellore, Tamil Nadu	Community-based cross-sectional	2811	Urban
5.	Pillai A et al., 2013	Northern Goa, Goa	Community-based cross-sectional	1899	Both
6.	Jonas JB et al., 2014	Nagpur, Maharashtra	Cross-sectional study	4698	Rural
7.	Katyal R et al., 2014	Meerut, Uttar Pradesh	Community-based cross-sectional	324 (Males)	Urban
8.	Pillai A et al., 2014	Northern Goa, Goa	Population-based survey	1867 (Males)	Both
9.	Rao TSS et al., 2014	Suttur, Karnataka	Population-based survey	3033	Rural
10.	Rathod SD et al., 2015	Sehore, Madhya Pradesh	Community-based cross-sectional	3220	Both
11.	National Mental Health Survey (NMHS), 2016	12 states across India	Multistage, stratified, random cluster study	34802	Both
12.	Manimunda SP et al., 2017	Andaman & Nicobar Islands, UT	Community-based cross-sectional	18000	Rural
13.	Sau A, 2017	Paschim Medinipur, West Bengal	Community-based cross-sectional	99	Rural
14.	Bhatia U et al., 2019	North Goa, Goa	Community-based cross-sectional	1451	Both
Anxiety Disorder
1.	Anita et al., 2003	Rohtak, Haryana	Cross-sectional survey	800	Both
2.	Pillai A et al., 2008	Margao, Morpila, Barcem, Bali and Fatorpa, Goa	Population-based survey	2048	Both
3.	Sahoo S et al., 2010	Ranchi, Jharkhand	College going students	405	Urban
4.	Seby K et al., 2011	Pune, Maharashtra	Population-based survey	202	Urban
5.	Nair MKC et al., 2013	Allapuzha, Kerala	Cross-sectional survey	500	Rural
6.	Rao TSS et al., 2014	Suttur, Karnataka	Population-based survey	3033	Rural
7.	National Mental Health Survey (NMHS), 2016	12 states across India	Multistage, stratified, random cluster study	34802	Both
8.	Sagar R et al., 2017	Eight districts across India	Community-based cross-sectional study	24,371	Both
9.	Jayashree K et al., 2018	Mangaluru, Karnataka	Cross-sectional school-going children	201	Urban
10.	Madasu S et al., 2019	Ballabhgarh, Haryana	Community-based cross-sectional study	729	Rural
11.	Madasu S et al., 2019	Ballabhgarh, Haryana	Community-based cross-sectional study	678	Rural
12.	Kirubasankar A et al., 2020	Puducherry, UT	Cross-sectional school-going children	462	Both
Intellectual disability
1.	Razdan S et al., 1994	Anantnag, Jammu and Kashmir, UT	Population-based survey	63645	Rural
2.	Anita et al., 2003	Rohtak, Haryana	Cross-sectional survey	800	Both
3.	Srinath S et al., 2005	Bengaluru, Karnataka	Population-based survey	1578	Both
4.	Khairkar P et al., 2013	Wardha, Maharashtra	Observational, hospital-based	3671	Both
5.	Rao TSS et al., 2014	Suttur, Karnataka	Population-based survey	3033	Rural
6.	Sharma S et al., 2015	Kangra, Himachal Pradesh	Population-based survey	2420	Rural
7.	National Mental Health Survey (NMHS), 2016	12 states across India	Multistage, stratified, random cluster study	34802	Both
8.	Sharma S et al., 2016	Kangra, Himachal Pradesh	Population-based survey	5300	Both
Suicidal attempt/death
1.	Chaudhary N et al., 2008	North Goa district, Goa	Community-based survey	Women, 8595	Urban
2.	Sauvaget C et al., 2009	Thiruvananthapuram, Kerala	Household visits	132000	Urban
3.	Nath Y et al., 2011	Ahmedabad, Gujarat	College students	1817	Urban
4.	Patel V et al., 2012	Multiple locations across India	Community-based survey	6671000	Both
5.	Gosavi SV et al., 2014	Kharangana, Maharashtra	Community-based survey	4790	Rural
6.	Joshi R et al., 2015	East and West Godavari districts, Andhra Pradesh	Household visits	185629	Rural
7.	National Mental Health Survey (NMHS), 2016	12 states across India	Multistage, stratified, random cluster study	34802	Both
Autism
1.	Raina SK et al., 2015	Five districts across Himachal Pradesh	Population-based cross-sectional	11000	Both
2.	National Mental Health Survey (NMHS), 2016	12 states across India	Multistage, stratified, random cluster study	1191	Both
3.	Poovathinal SA et al., 2016	Shoranur, Kerala	Community-based survey	18480	Both
4.	Rudra A et al., 2017	Kolkata, West Bengal	School-based	11849	Urban
5.	Raina SK et al., 2017	Five districts across Himachal Pradesh	Population-based cross-sectional	28078	Both
6.	Arora NK et al., 2018	Five districts across India	Population-based cross-sectional	3964	Urban
Bipolar Disorders
1.	Merikangas KR et al., 2011	Puducherry, UT	Population-based survey	5007	Urban
2.	Rao TSS et al., 2014	Suttur, Karnataka	Population-based survey	3033	Rural
3.	National Mental Health Survey (NMHS), 2016	12 states across India	Multistage, stratified, random cluster study	34802	Both
4.	Shaji KS et al., 2017	Five districts across Kerala	Population-based survey	192980	Both
ID	Age range	Screening tool/instrument	Diagnosis by	Cases detected	Prevalence/1000
Depression
1.	10–19 years	The Youth Report of Paediatric Symptom Checklist (Y-PSC)	International Classification of Diseases (ICD-10)	12	3.10%
0.2	>60 years	Two-question screen	Revised Clinical Interview Schedule	65	31.5%
3.	≥20 years	Modified Patient Health Questionnaire (PHQ-9)	NA	3847	15.9%
4.	>65 years	Multiple tools	ICD-10	127	12.70%
5.	≥18 years	NA	WHO Composite International Diagnostic Interview (CIDI) Version 3.0	NA	4.5%
6.	17–22 years	NA	Depression, Anxiety, and Stress Scale-21 (DASS-21) and Mini-International Neuropsychiatric Interview (MINI)	NA	12.1%
7.	10–15 years	Childhood Psychopathology Measurement Schedule (CPMS)	ICD-10	4	2%
8.	>65 years	Geriatric depression scale-15	NA	33	16.3%
9.	≥18 years	NA	WHO-CIDI 3.0	95	3.14%
10.	NA	NA	Schedule for Affective Disorders and Schizophrenia for School-Age Children (K-SADS-PL)	58	3.13%
11.	5–14 years	Interview schedule	Diagnostic and Statistical Manual (DSM-IV) criteria	1	0.4%
12.	>30 years	Center for Epidemiologic Studies Depression Scale (CESD)	NA	613	13%
13.	All age groups	NA	MINI adult and Kid version	NA	6.62%
14.	>18 years	PHQ-9	NA	58	6%
15.	>65 years	Montgomery Asberg Depression Rating Scale (MADRS)	Diagnostic criteria for research for Depression and ICD-10	NA	39.1%
16.	>60 years	Geriatric Depression Scale (GDS-15)	NA	271	8.90%
17.	All age groups	Brief Patient Health Questionnaire (BPHQ)	NA	NA	3%
18.	>65 years	NA	ICD-10 and EURO-D	NA	3.9%
19.	>65 years	NA	ICD-10 and EURO-D	NA	12.6%
20.	>18 years	NA	MINI adult and Kid version	NA	5.20%
21.	>60 years	Geriatric Depression Scale	MINI	NA	11.40%
22.	>18 years	PHQ-9	NA	NA	14.60%
23.	13–18 years	Children's Depression Inventory (CDI)	NA	140	18%
24.	NA	Beck's Depression Inventory II (BDI)	NA	NA	49.20%
25.	≥18 years	PHQ-9	NA	546	2.4%
26.	NA	DASS-21	NA	249	54.1%
27.	>50 years	Self-report	WHO-CIDI	NA	15.2%
28.	>60 years	Geriatric Depression Scale	NA	172	NA
Alcohol use disorder
1.	≥15 years	Instrument by Ray R et al.	ICD-10	201	NA
2.	≥18 years	AUDIT	NA	150	65.8%
3.	≥10 years	AUDIT	NA	NA	9.4%
4.	≥18 years	AUDIT	NA	588	20.9%
5.	18–49 years	Author's questionnaire	NA	732	NA
6.	>30 years	AUDIT	NA	1081	23%
7.	≥15 years	AUDIT	NA	NA	29.6%
8.	20–49 years	AUDIT	NA	1007	NA
9.	All age groups	MINI Adult			3.95%
10.	≥18 years	AUDIT	NA	NA	13.3%
11.	>18 years	MINI Adult	ICD-10	NA	4.7%
12.	≥14 years	AUDIT	NA	NA	20.7%
13.	>18 years	AUDIT	NA	NA	34.50%
14.	18–49 years	AUDIT	NA	184	12.7%
Anxiety Disorder
1.	6–14 years	Childhood Psychopathology Measurement Schedule (CPMS)	Diagnostic Interview Schedule for Children (DISC) and ICD-10	23	2.87%
2.	12–16 years	NA	Development and Well-Being Assessment (DAWBA) AND DSM-IV	20	1%
3.	NA	Depression, Anxiety, and Stress Scale	MINI	NA	19%
4.	>65 years	General health questionnaire-12	NA	17
5.	11–19 years	Screen for Child Anxiety Related Emotional Disorders (SCARED)	Schedule for Affective Disorders and Schizophrenia for School-Age Children/Present and Lifetime Version (K-SADS-PL)	NA	14.40%
6.	All age groups	NA	MINI adult and Kid version	NA	1.91%
7.	>18 years	NA	MINI adult and Kid version	NA	1.9%
8.	≥18 years	NA	World Mental Health-Composite International Diagnostic Interview (WMH-CIDI)	NA	3.41%
9.	15–19 years	SCARED	NA	110	54.7%
10.	10–19 years	SCARED	NA	154	22.7%
11.	10–19 years	SCARED	MINI Kid/adolescent version	112	16.6%
12.	NA	SCARED	NA	167	36%
Intellectual disability
1.	≥15 years	WHO protocol for measuring the prevalence of neurological disorders in developing countries, 1981	Clinical evaluation	133	2.09
2.	6–14 years	Childhood Psychopathology Measurement Schedule (CPMS)	Diagnostic Interview Schedule for Children (DISC) and ICD-10	26	3.25%
3.	4–16 years	Multiple questionnaires	ICD-10-DCR	4	0.3%
4.	4–16 years	Vineland Social Maturity Scale and Malin's IQ test	ICD-10-DCR	62	1.68%
5.	All age groups	NA	Structured interview based on DSM-IV TR and ICD-10 criteria	NA	0.33%
6.	1–10 years	Ten Questions Screen for the disability	Clinical evaluation and Stanford-Binet intelligence test	52	2.51%
7.	13–17 years	NA	Intellectual disability screener	NA	0.6%
8.	1–10 years	Ten Questions Screen for the disability	Clinical evaluation and Stanford-Binet intelligence test	91	1.71%
Suicidal attempt/death
1.	18–50 years	Author's questionnaire	Revised Clinical Interview Schedule (CIS-R)	45	2.57% (suicidal attempts)
2.	≥35 years	Verbal autopsy	NA	385	Suicidal deaths
3.	18–24 years	Author's questionnaire	NA	73	4.04% (suicidal attempts)
4.	≥15 years	NA	ICD-10	2684	3% (suicidal deaths)
5.	All age groups	NA	Pre-tested interview schedule	10	1.1% (suicidal attempts)
6.	All age groups	Verbal autopsy as per Registrar General of India's Sample Registration System	ICD-10	280	4.8% (suicidal deaths)
7.	>18 years	NA	MINI adult	NA	0.9%
Autism
1.	1–10 years	Indian Scale for Assessment of Autism (ISAA)	Clinical psychologist and public health specialist	10	0.9
2.	13–17 years	NA	MINI adult and Kid version	NA	1.6%
3.	1–30 years	Author's questionnaire	DSM-IV-TR	43	23.3/10000
4.	3–8 years	Social communication disorder checklist (SCDC), social communication questionnaire (SCQ), and autism diagnostic observation schedule (ADOS)	Clinical psychologist	6	0.23%
5.	1–10 years	Indian Scale for Assessment of Autism (ISAA)	Clinical psychologist and public health specialist	43	0.15%
6.	2–9 years	Team of physicians and social scientists	INCLEN Diagnostic Tool for Autism Spectrum Disorder (INDT-ASD) and DSM-IV-TR	44	11
Bipolar Disorders
1.	>18 years	NA	WHO Composite International Diagnostic Interview (WMH-CIDI, version 3.0) and DSM-IV	5	0.1%
2.	All age groups	NA	Mini International Neuropsychiatric Inventory (MINI) adult version and the MINI-Kid version	NA	0.52%
3.	>18 years	NA	Mini International Neuropsychiatric Inventory (MINI) adult version and the MINI-Kid version	NA	0.3%
4.	All age groups	Symptoms in Others Questionnaire (SOQ) and General Health Questionnaire (GHQ-12)	ICD-10 and a psychiatrist	NA	0.29%

Search strategy

Two authors (Vikas Dhiman and Geetha R. Menon) worked independently and searched Pub Med, Global Health Data Exchange (GHDx), and Google Scholar databases to retrieve the relevant studies. In Pub Med, the search was performed from January 1980 till March 2022 using search term “disease,” that is, “depression,” “depressive symptoms,” “depressive disorder,” “dysthymia,” “alcohol use disorder,” “alcoholism,” “anxiety disorder,” “stress disorder,” “mental retardation,” “intellectual disability,” “suicidal attempt,” “suicide,” “suicidal death,” “autism,” “autism spectrum disorder” and “bipolar disorder” combined separately with terms “prevalence,” “epidemiology,” “morbidity,” “incidence,” and “mortality.” Each of these phrases combined with the term “India” using the Boolean operator “AND” and search strings were run in the “title/abstract” category in the advanced search tab for retrieving relevant articles. A thorough literature search was done to retrieve the relevant articles in the GHDx data catalog of the Institute of Health Metrics and Evaluation (IHME), which is a database of the Global Burden of Diseases (GBD) project.[15] In addition, eligible articles were also searched in the Google Scholar database using the above-mentioned phrases under the “with the exact phrase” category. The government websites, scientific reports, and Magazine articles were also hand-searched for grey literature.

Study Inclusion Criteria

The following inclusion criteria were considered for the studies: (i) population-based study, (ii) adequate information about the screened population, (iii) details of screening instruments and diagnostic criteria, (iv) positive cases, and (v) published in the English language. All psychiatric prevalence studies conducted in India from January 1980 till March 2022 were included for initial screening. The studies which were not population-based and conducted on specific socio-economic or ethnic groups were excluded. For inclusion, the authors should have given information about the structured screening tool employed for the study participants or used as a two-step survey methodology. The eligible diagnostic criteria for inclusion were as per the International Classification of Diseases (ICD) or Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria or diagnosis by the psychiatrist. We did not specify any case definition for the disorder and adopted the definitions used by the individual studies. We also screened editorials, conference proceedings, and brief communications in the literature to look for any relevant data.

Study Selection and Data Extraction

Two authors (Vikas Dhiman and Geetha R. Menon) worked independently and screened all the retrieved articles for study selection. The initial screening was done by reviewing the title and abstract of the articles. A detailed screening of full texts of articles found suitable on initial screening was performed. All disagreements were resolved through discussion between two authors (Vikas Dhiman and Geetha R. Menon) and with the third reviewer (Rajnarayan R. Tiwari), who was not part of the initial screening. We used Microsoft Excel software to collate and de-duplicate the articles from databases. Two authors (Vikas Dhiman and Geetha R. Menon) independently extracted the data in a pre-piloted Excel sheet. The data regarding the authors and place of the study, study design, population screening, participant age groups, types of screening instruments or diagnostic criteria, positive cases, and prevalence rates were extracted from all eligible studies.

Study Quality Assessment

Two authors (Vikas Dhiman and Geetha R. Menon) worked independently and used Joanna Briggs Institute's (JBI) critical appraisal tool for prevalence studies to assess the methodological quality of the included studies.[16] All nine questions of the tool were scored for each article. For every “yes” and “no” response to the question, a score of 1 and 0 was given, respectively, while a 0.5 score was given when the response was “unclear.” The quality score was calculated from a total score of 9. We set more than or equal to 75%, that is, more than or equal to 6.5 out of 9, as a good quality score. A thorough discussion between two authors (Vikas Dhiman and Geetha R. Menon) and the third reviewer (Rajnarayan R. Tiwari) resolved the disparity in assigning the studies' quality scores.

Statistical analysis

The meta-analysis was performed when three or more eligible studies were available for each disorder to calculate the pooled prevalence rate. We used MetaXL(version 5.3) for statistical analysis, a free add-in software in Microsoft Excel. The pooled prevalence rates with a 95% confidence interval were calculated by using the random effects model.[17] The individual studies and the pooled estimates were represented in the Forest plots. Cochran's Q test and I² statistics were used to assess the heterogeneity across studies. The P value for Cochran's Q test was fixed at less than 0.01, and values less than 25%, between 25 and 50%, and over 50% as low, moderate, and high heterogeneity, respectively, in I² statistics. Funnel and Doi plots were used for visual detection of the publication bias. We used the Luis Furuya–Kanamori (LFK) index to assess the degree of asymmetry in Doi plots, wherein the asymmetry is defined as none (index less than or equal to ± 1), minor (index between ± 1 and ± 2), and major (index more than or equal to ± 2).[18] The LFK test was chosen over Egger's test as it is more sensitive in detecting publication bias when the number of studies for meta-analysis is less than ten.[18] The leave-one-out approach, that is, iteratively removing one study at a time and recalculating the pooled estimate, was used to perform the sensitivity analyses. The subgroup analyses were done to calculate the pooled prevalence rates in urban vs. rural population, old vs. young population, and year of the publications.

RESULTS

Study Selection

A total of 1320 records were retrieved through search in Pub Med, Google Scholar, and GBD databases. After the removal of duplicates (n = 745), non-relevant articles (n = 397), and review articles (n = 12), 166 articles were evaluated for full-text screening. The relevant data were not available from 87 articles on full-text screening. The remaining 79 studies were ultimately eligible for inclusion in the meta-analysis. The total number of studies included in each disorder was: depression (n = 28), alcohol use disorder (n = 14), anxiety disorder (n = 12), intellectual disability (n = 8), suicidal attempt/death (n = 7), autism (n = 6) and bipolar disorder (n = 4). Figure 1 shows the PRISMA chart for screening and selection of eligible studies through searches from various databases using the search strategy.

Figure 1.

PRISMA chart showing the screening and selection of studies to estimate the pooled prevalence of depression, alcohol use disorder, anxiety disorder, intellectual disability, suicidal attempts/deaths, autism, and bipolar disorder. (Original)

Study Characteristics

Study area and population

The prevalence of depression was reported in 28 studies, covering a population of 126,788. The prevalence of alcohol use disorder and anxiety disorder were reported in 14 and 12 studies that included 76,370 and 68,231 populations, respectively. A total of 115,249 and 7,038,633 were screened for intellectual disability and suicidal attempt/death from eight and seven studies, respectively. The prevalence rates of autism and bipolar disorders were available from six and four studies, covering a total population of 74,562 and 235,822, respectively. The maximum number of studies on depression (n = 12), alcohol use disorder (n = 5), anxiety disorder (n = 8), suicidal attempt/death (n = 4), and bipolar disorders (n = 4) were conducted in south India, while maximum studies on intellectual disability (n = 5) and autism (n = 5) were conducted in North India. In all disorders, most studies were population-based surveys, while seven and three studies were school/college-based on depression and anxiety disorder, respectively. Among all disorders, maximum studies were conducted in combined urban and rural populations, while only the urban population was studied more in depression (n = 11) and suicidal attempt/death (n = 3). The majority of the studies on depression (n = 19), alcohol use disorder (n = 13), and suicidal attempt/death (n = 5) were conducted in the adult population (≥18 years), while studies on anxiety disorder (n = 6), intellectual disability (n = 6) and autism (n = 6) were conducted commonly in pediatric (≤18 years) population.

Screening instruments and diagnostic criteria

The most common instruments used for screening in each disorder were depression [(Geriatric Depression Scale and Patient Health Questionnaire (PHQ)-9, (n = 4)], alcohol use disorder [Alcohol Use Disorders Identification Test (AUDIT) (n = 10)], anxiety disorder [Screen for Child Anxiety Related Disorders (SCARED), (n = 5)], intellectual disability [(10 questions screen for disability, (n = 2)], suicidal attempt/death [author's own questionnaire (n = 2)], and autism [Indian Scale for Assessment in Autism (ISAA), (n = 2)]. ICD, Tenth Revision (ICD-10) was the most common diagnostic criteria used in depression (n = 3), alcohol use disorder (n = 2), intellectual disability (n = 4), and suicidal attempt/death (n = 2), while Mini-International Neuropsychiatric Interview (MINI) was used in anxiety disorder (n = 4) and bipolar disorder (n = 2). The details of study characteristics are presented in Supplementary Table 2.

Supplementary Table 2.

PRISMA checklist

Section and Topic	Item#	Checklist item	Location where item is reported
TITLE
Title	1	Identify the report as a systematic review.	3
ABSTRACT
Abstract	2	See the PRISMA 2020 for Abstracts checklist.	1
INTRODUCTION
Rationale	3	Describe the rationale for the review in the context of existing knowledge.	2
Objectives	4	Provide an explicit statement of the objective (s) or question (s) the review addresses.	3
METHODS
Eligibility criteria	5	Specify the inclusion and exclusion criteria for the review and how studies were grouped for the syntheses.	4
Information sources	6	Specify all databases, registers, websites, organizations, reference lists, and other sources searched or consulted to identify studies. Specify the date when each source was last searched or consulted.	3,4
Search strategy	7	Present the full search strategies for all databases, registers, and websites, including any filters and limits used.	3
Selection process	8	Specify the methods used to decide whether a study met the inclusion criteria of the review, including how many reviewers screened each record and each report retrieved, whether they worked independently, and, if applicable, details of automation tools used in the process.	4
Data collection process	9	Specify the methods used to collect data from reports, including how many reviewers collected data from each report, whether they worked independently, any processes for obtaining or confirming data from study investigators, and, if applicable, details of automation tools used in the process.	4
Data items	10a	List and define all outcomes for which data were sought. Specify whether all results that were compatible with each outcome domain in each study were sought (e.g., for all measures, time points, analyses), and if not, the methods used to decide which results to collect.	4
	10b	List and define all other variables for which data were sought (e.g., participant and intervention characteristics, funding sources). Describe any assumptions made about any missing or unclear information.	4
Study risk of bias assessment	11	Specify the methods used to assess the risk of bias in the included studies, including details of the tool (s) used, how many reviewers assessed each study and, whether they worked independently, and if applicable, details of automation tools used in the process.	4
Effect measures	12	Specify for each outcome the effect measure (s) (e.g., risk ratio, mean difference) used in the synthesis or presentation of results.	5
Synthesis methods	13a	Describe the processes used to decide which studies were eligible for each synthesis (e.g., tabulating the study intervention characteristics and comparing against the planned groups for each synthesis (item #5)).	5
	13b	Describe any methods required to prepare the data for presentation or synthesis, such as handling of missing summary statistics, or data conversions.	5
	13c	Describe any methods used to tabulate or visually display the results of individual studies and syntheses.	5
	13d	Describe any methods used to synthesize results and provide a rationale for the choice (s). If meta-analysis was performed, describe the model (s), method (s) to identify the presence and extent of statistical heterogeneity, and software package (s) used.	5
	13e	Describe any methods used to explore possible causes of heterogeneity among study results (e.g., subgroup analysis, meta-regression).	5
	13f	Describe any sensitivity analyses conducted to assess the robustness of the synthesized results.	5
Reporting bias assessment	14	Describe any methods used to assess the risk of bias due to missing results in a synthesis (arising from reporting biases).	5
Certainty assessment	15	Describe any methods used to assess certainty (or confidence) in the body of evidence for an outcome.	5
RESULTS
Study selection	16a	Describe the results of the search and selection process, from the number of records identified in the search to the number of studies included in the review, ideally using a flow diagram.	5
	16b	Cite studies that might appear to meet the inclusion criteria, but which were excluded, and explain why they were excluded.	5
Study characteristics	17	Cite each included study and present its characteristics.	6
Risk of bias in studies	18	Present assessments of risk of bias for each included study.	7
Results of individual studies	19	For all outcomes, present, for each study: (a) summary statistics for each group (where appropriate) and (b) an effect estimate and its precision (e.g., confidence/credible interval), ideally using structured tables or plots.	7
Results of syntheses	20a	For each synthesis, briefly summarise the characteristics and risk of bias among contributing studies.	7
	20b	Present results of all statistical syntheses conducted. If meta-analysis was done, present for each the summary estimate and its precision (e.g., confidence/credible interval) and measures of statistical heterogeneity. If comparing groups, describe the direction of the effect.	7
	20c	Present results of all investigations of possible causes of heterogeneity among study results.	7,8
	20d	Present results of all sensitivity analyses conducted to assess the robustness of the synthesized results.	7,8
Reporting biases	21	Present assessments of risk of bias due to missing results (arising from reporting biases) for each synthesis assessed.	7,8
Certainty of evidence	22	Present assessments of certainty (or confidence) in the body of evidence for each outcome assessed.	7,8
DISCUSSION
Discussion	23a	Provide a general interpretation of the results in the context of other evidence.	9
	23b	Discuss any limitations of the evidence included in the review.	10
	23c	Discuss any limitations of the review processes used.	10
	23d	Discuss the implications of the results for practice, policy, and future research.	10
OTHER INFORMATION
Registration and protocol	24a	Provide registration information for the review, including the register name and registration number, or state that the review was not registered.	3
	24b	Indicate where the review protocol can be accessed or state that a protocol was not prepared.	3
	24c	Describe and explain any amendments to information provided at registration or in the protocol.	NA
Support	25	Describe sources of financial or non-financial support for the review and the role of the funders or sponsors in the review.	10
Competing interests	26	Declare any competing interests of review authors.	10
Availability of data, code, and other materials	27	Report which of the following are publicly available and where they can be found: template data collection forms; data extracted from included studies; data used for all analyses; analytic code; any other materials used in the review.	NA

From: Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71. doi: 10.1136/bmj.n71. For more information, visit: http://www.prisma-statement.org/

Quality Assessment

All included studies scored more than 6.5 out of 9. The highest score, that is, 9/9, was scored by most studies. For example, in depression and alcohol use disorder, 10 and 9 studies scored 9. Four studies in anxiety disorder and five studies in intellectual disability scored 9. In suicidal attempt/death, four studies scored 9, while three scored 8. In autism and bipolar disorder, three and two studies scored 9, respectively.

Estimation of Pooled Prevalence Rates

In meta-analysis of 28 studies, the pooled prevalence of depression was found to be 12.4% (95% CI 9.4–15.9) (Q = 7444.8, P < 0.001, I² = 100%) [Figure 2]. On performing meta-analysis of 14 and 12 studies, the pooled prevalence of alcohol use disorder and anxiety disorder was 21.5% (95% CI 14.1–30.0) (Q = 7339.0, P < 0.001, I² = 100%) [Figure 3] and 11.6% (95% CI 8.1–15.7) (Q = 1704.5, P < 0.001, I² = 99%) [Figure 4], respectively. The pooled prevalence of intellectual disability and suicidal attempt/death in meta-analysis of eight and seven studies, were 1% (95% CI 0.5–1.6) (Q = 386.2, P < 0.001, I² = 98%) and 0.5% (95% CI 0.3–0.8) (Q = 1932.5, P < 0.001, I² = 100%), respectively. The meta-analysis of six studies in autism and four studies in bipolar disorder showed pooled prevalence as 0.3% (95% CI 0.1–0.6) (Q = 126.2, P < 0.001, I² = 96%) and 0.3% (95% CI 0.2–0.4) (Q = 13.6, P < 0.001, I² = 78%), respectively.

Figure 2.

Pooled prevalence (proportion) of depression (a) and alcohol use disorder (b) in India (1980–2022). Error bars indicate 95% confidence intervals. Diamond shows the pooled prevalence rate with 95% confidence intervals based on the random effects (RE) model (Original)

Figure 3.

Pooled prevalence (proportion) of anxiety disorder (a) and intellectual disability (b) in India (1980–2022). Error bars indicate 95% confidence intervals. Diamond shows the pooled prevalence rate with 95% confidence intervals based on the random effects (RE) model (Original)

Figure 4.

Pooled prevalence (proportion) of suicidal attempts/deaths (a), autism (b), and bipolar disorder (c) in India (1980–2022). Error bars indicate 95% confidence intervals. Diamond shows the pooled prevalence rate with 95% confidence intervals based on the random effects (RE) model (Original)

Publication Bias

The Funnel and Doi plots were asymmetrical in all disorders except autism and bipolar disorder, indicating the presence of publication bias [Supplementary Figures 1 ^{(316.6KB, tif)} -7 ^{(342.1KB, tif)} ]. The LFK index of asymmetry from Doi plots was 2.67 (depression), 5.75 (alcohol use disorder), 6.57 (anxiety disorder), 5.58 (intellectual disability), and 8.10 (suicidal attempt/death). The LFK index for autism showed minor asymmetry (1.39), while no publication bias was seen in bipolar disorder (LFK index = 0.85).

Sensitivity Analysis

On performing sensitivity analysis by leave-one-out analysis by iteratively removing one study at a time and recalculating the pooled prevalence, it was found that I² values did not change significantly in all disorders except bipolar disorders. In bipolar disorder, I² dropped from 78% to 58%, by removing a study by Merikangas K.R. et al.[19] [Supplementary Figure 8 ^{(201.5KB, tif)} ].

Subgroup Analysis

Urban vs. rural

The pooled prevalence of depression in urban and rural populations was almost the same, that is, 11.5% (95% CI 5.6–19.1) and 12.3% (95% CI 6.7–19.3), respectively. The prevalence of alcohol use disorder was more in the urban population [37.2% (95% CI 13.9–63.8)] than in the rural population [16.0% (95% CI 8.3–25.5)] [Supplementary Figure 9a ^{(570KB, tif)} and b ^{(570KB, tif)} ]. Similarly, the prevalence of anxiety disorder was found more in the urban population [24.3% (95% CI 3.7–52.9)] than in the rural population [11.1% (95% CI 1.2–27.0)] [Supplementary Figure 9c ^{(570KB, tif)} and d ^{(570KB, tif)} ]. The subgroup analysis in the rest of the disorders could not be performed due to less than three studies.

Younger vs. older population

On performing a subgroup analysis of pooled prevalence in younger vs. older populations, the pooled prevalence of depression was slightly higher in the population more than 18 years [13.4% (95% CI 9.9–17.3)] than in the population less than 18 years of age [12.2% (95% CI 0–34.4)]. The prevalence of anxiety disorder was much higher in the younger population [16.7% (95% CI 5.1–32.7)] than in the older population [6.2% (95% CI 4.0–9.0)] [Supplementary Figures 10a ^{(640KB, tif)} and 10b ^{(640KB, tif)} ]. Due to less than three studies on other disorders, subgroup analysis could not be performed.

Decadal pooled prevalence (1991–2021)

On performing the decadal analysis of the pooled prevalence of each disorder from 1991 to 2021, it was found that the prevalence of depression increased from 11.4% (95% CI 5.8–18.4) during 2001–2010 to 12.6% (95% CI 9.0–16.7) during 2011–2021. Similarly, the pooled prevalence of anxiety disorder increased from 5.7% (95% CI 0–15.4) during 2001–2010 to 14.1% (95% CI 9.7–19.2) during 2011–2021 [Supplementary Figure 10c ^{(640KB, tif)} and d ^{(640KB, tif)} ]. The pooled prevalence of intellectual disability remains the same during 1991–2010 and 2011–2021, that is, 0.8% (95% CI 0.00–2.2) and 1.1% (95% CI 0.5–1.9), respectively. The decadal analysis of alcohol use disorder, suicidal attempt/death attempts, autism, and bipolar disorder could not be performed due to less than three studies in each decade.

DISCUSSION

The present meta-analysis was performed to estimate up-to-date pooled prevalence rates of seven psychiatric disorders, viz. depression, alcohol use disorder, anxiety disorder, intellectual disability, suicidal attempt/death, autism, and bipolar disorder, in India. The total study population was very heterogeneous across different studies due to varied age groups, geographical areas, and community settings used in the studies. Most psychiatric epidemiological studies were conducted in South and North India. Few studies were reported from the Northeast and Central India,[20,21] which may be attributable to inadequate mental health infrastructure, difficult-to-reach areas, tribal population, and lack of awareness in these regions.[22] Although school/college-based psychiatric studies have been reported on depression and anxiety disorder, such studies are required in autism, bipolar disorder, and alcohol use disorder, especially given the recent emergence of these disorders in adolescents.[5] The present analysis showed that different screening instruments and diagnostic tools were employed across studies, contributing to high heterogeneity among the studies. AUDIT was the only tool used uniformly in studies on alcohol use disorders.

In the present analysis, the pooled prevalence of depression and alcohol use disorder was found to be similar to previous studies.[20,23,24,25,26] However, the pooled prevalence of depression, alcohol use disorder, and anxiety disorder reported in the present meta-analysis were higher compared to NMHS. This is because NMHS was an epidemiological survey which used multi-stage, stratified, random cluster sampling technique and both quantitative and qualitative methods were employed. The present study is a meta-analysis which pooled the results of all previously conducted prevalence studies (including NMHS) of all age groups and geographical locations of the country. The prevalence of intellectual disability, suicidal attempt/death, autism, and bipolar disorder were similar to the prevalence reported in NMHS,[5,20] albeit the number of studies reporting prevalence was few.

The meta-analysis showed a high prevalence of alcohol use disorder and anxiety disorder in the urban population vs. rural population. These disorders are closely linked to non-communicable diseases, violence, and social problems, commonly seen in urban lifestyles.[27] The prevalence of depression was higher in the older population in India, similar to multiple studies reporting a higher prevalence of depression among the elderly population from other low and middle-income countries.[23,28] The present study showed a high prevalence of anxiety disorder in the younger population, similar to previous studies.[29] Both depression and anxiety disorders showed an increase in prevalence from 2001 to 2020 on decadal analysis, which coincides with the findings from NMHS.[20]

A meta-analysis of prevalence of overall child and adolescent psychiatric disorders was previously done by Malhotra et al.[30] which showed the prevalence rate of child and adolescent psychiatric disorders to be 6.5%, which is much higher than the present meta-analysis as we have estimated the pooled prevalence rate of only autism and intellectual disability. Another meta-analysis[31] of prevalence of autism showed pooled prevalence of 0.1% which is similar to pooled prevalence of autism in the present meta-analysis. Another meta-analysis[32] showed a prevalence of depression in elderly population of age more than 60 years only to be 34.4%, which is much higher than the present study because the prevalence of depression is higher among elderly population due to multiple co-morbidities.

There are several limitations to be considered while interpreting the results of this study. The age groups studied in different studies were variable, hence pooling of results as per age groups was not possible due to lesser number of studies in each group. Hence pooling of data from across ages was done for meta-analysis. The information regarding screening tools used as not mentioned in few studies. Different case definitions, screening tools, and diagnostic instruments used to ascertain the cases across different studies led to high heterogeneity among the pooled estimates. Due to this, the pooled prevalence of depression, alcohol use disorder, and anxiety disorder were reported were much higher in the present study as compared to NMHS. We found only a few studies on autism, bipolar disorders, suicides, and intellectual disability, so subgroup analysis could not be possible. All published studies were included, but there are chances of missing out on studies from the grey literature.

CONCLUSION

The present study is the most up-to-date prevalence of depression, alcohol use disorder, anxiety disorder, intellectual disability, suicidal attempt/death, autism, and bipolar disorder in India, which can guide policy decisions. Other than varied methodological approaches and screening/diagnostic tools used in the studies, the present study identified that further psychiatric epidemiological studies are required in Northeast and Central India. Also, more epidemiological studies need to be conducted on autism, bipolar disorder, suicidal attempt/death, and intellectual disability. To avoid such methodological problems in the future, the prevalence studies should ideally be conducted as a multi-centre study, where uniform criteria for age of inclusion, the case definitions, and screening/diagnostic tools (DSM/ICD or others). We suggest that large epidemiological exercises like NMHS should be conducted more frequently (once in two years) so that methodological issues like high heterogeneity due to usage of different case definitions, screening tools, and diagnostic instruments may be avoided.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Supplementary Figure 1

Funnel Plot and Doi Plot for meta-analysis of studies of prevalence of depression (a and b). In the Funnel plot, the standard error of effect size is plotted on the y-axis and the log-transformed prevalence is on the x-axis. The black circles represent the result estimates for each included study. The diagonal lines represent 95% confidence intervals whereas the central vertical line represents the natural logarithm of the overall pooled prevalence (proportion). In the Doi plot, log transformed prevalence is plotted on the x-axis and z-score, which is a measure of precision derived from the standard error of each study's effect size, is plotted on the y-axis. Empty circles indicate the result estimates of the included studies

Supplementary Figure 2

Funnel Plot and Doi Plot for meta-analysis of studies of prevalence of alcohol use disorder (a and b). In the Funnel plot, the standard error of effect size is plotted on the y-axis and the log-transformed prevalence is on the x-axis. The black circles represent the result estimates for each included study. The diagonal lines represent 95% confidence intervals whereas the central vertical line represents the natural logarithm of the overall pooled prevalence (proportion). In the Doi plot, log transformed prevalence is plotted on the x-axis and z-score, which is a measure of precision derived from the standard error of each study's effect size, is plotted on the y-axis. Empty circles indicate the result estimates of the included studies

Supplementary Figure 3

Funnel Plot and Doi Plot for meta-analysis of studies of prevalence of anxiety disorder (a and b). In the Funnel plot, the standard error of effect size is plotted on the y-axis and the log-transformed prevalence is on the x-axis. The black circles represent the result estimates for each included study. The diagonal lines represent 95% confidence intervals whereas the central vertical line represents the natural logarithm of the overall pooled prevalence (proportion). In the Doi plot, log transformed prevalence is plotted on the x-axis and z-score, which is a measure of precision derived from the standard error of each study's effect size, is plotted on the y-axis. Empty circles indicate the result estimates of the included studies

Supplementary Figure 4

Funnel Plot and Doi Plot for meta-analysis of studies of prevalence of intellectual disability (a and b). In the Funnel plot, the standard error of effect size is plotted on the y-axis and the log-transformed prevalence is on the x-axis. The black circles represent the result estimates for each included study. The diagonal lines represent 95% confidence intervals whereas the central vertical line represents the natural logarithm of the overall pooled prevalence (proportion). In the Doi plot, log transformed prevalence is plotted on the x-axis and z-score, which is a measure of precision derived from the standard error of each study's effect size, is plotted on the y-axis. Empty circles indicate the result estimates of the included studies

Supplementary Figure 5

Funnel Plot and Doi Plot for meta-analysis of studies of prevalence of suicidal attempts/deaths (a and b). In the Funnel plot, the standard error of effect size is plotted on the y-axis and the log-transformed prevalence is on the x-axis. The black circles represent the result estimates for each included study. The diagonal lines represent 95% confidence intervals whereas the central vertical line represents the natural logarithm of the overall pooled prevalence (proportion). In the Doi plot, log transformed prevalence is plotted on the x-axis and z-score, which is a measure of precision derived from the standard error of each study's effect size, is plotted on the y-axis. Empty circles indicate the result estimates of the included studies

Supplementary Figure 6

Funnel Plot and Doi Plot for meta-analysis of studies of prevalence of autism (a and b). In the Funnel plot, the standard error of effect size is plotted on the y-axis and the log-transformed prevalence is on the x-axis. The black circles represent the result estimates for each included study. The diagonal lines represent 95% confidence intervals whereas the central vertical line represents the natural logarithm of the overall pooled prevalence (proportion). In the Doi plot, log transformed prevalence is plotted on the x-axis and z-score, which is a measure of precision derived from the standard error of each study's effect size, is plotted on the y-axis. Empty circles indicate the result estimates of the included studies

Supplementary Figure 7

Funnel Plot and Doi Plot for meta-analysis of studies of prevalence of bipolar disorder (a and b). In the Funnel plot, the standard error of effect size is plotted on the y-axis and the log-transformed prevalence is on the x-axis. The black circles represent the result estimates for each included study. The diagonal lines represent 95% confidence intervals whereas the central vertical line represents the natural logarithm of the overall pooled prevalence (proportion). In the Doi plot, log transformed prevalence is plotted on the x-axis and z-score, which is a measure of precision derived from the standard error of each study's effect size, is plotted on the y-axis. Empty circles indicate the result estimates of the included studies. LFK index: Luis Furuya-Kanamori index

Supplementary Figure 8

Leaving-one-out analysis showing the proportion of prevalence of bipolar disorder after removing the study by Merikangas KR et al.

Supplementary Figure 9

Subgroup analysis showing the pooled prevalence of alcohol use disorder (a and b) and anxiety disorder (c and d) in urban vs. rural population

Supplementary Figure 10

Subgroup analysis showing pooled prevalence of anxiety disorder in young vs. older population (a and b) and during 2001-2010 and 2011-2021 (c and d) decadal analysis