Prevalence and its correlates of anxiety disorders from India’s National Mental Health Survey 2016

Abstract

Introduction:

Anxiety disorders (ADs) impact the quality of life and productivity at an individual level and result in substantial loss of national income. Representative epidemiological studies estimating the burden of ADs are limited in India. National Mental Health Survey (NMHS) 2016 of India aimed to strengthen mental health services across India assessed the prevalence and pattern of public health priority mental disorders for mental health-care policy and implementation. This article focuses on the current prevalence, sociodemographic correlates, disability, and treatment gap in ADs in the adult population of NMHS 2016.

Materials and Methods:

NMHS 2016 was a nationally representative, multicentered study across 12 Indian states during 2014–2016. Diagnosis of ADs (generalized AD, panic disorder, agoraphobia, and social AD) was based on Mini-International Neuropsychiatric Interview 6.0.0. Disability was by Sheehan’s Disability Scale.

Results:

The current weighted prevalence of ADs was 2.57% (95% confidence interval: 2.54–2.60). Risk factors identified were female gender, 40–59 age group, and urban metro dwellers. Around 60% suffered from the disability of varying severity. The overall treatment gap for ADs was 82.9%.

Conclusions:

The burden of AD is similar to Depressive disorders, and this article calls for the immediate attention of policymakers to institute effective management plans in existing public health programs.

INTRODUCTION

Anxiety disorders (ADs) are among the most common psychiatric illnesses that impact the quality of life and functioning and productivity of an individual and thereby loss of national income. As per the World Mental Health Survey (WMHS), the lifetime prevalence of ADs across different countries ranges between 3% and 19%.[1] The Global Burden of Disease 2015[2] ranked ADs as the sixth-largest contributor of years lived with disability. ADs are primarily underdetected and undertreated despite the huge disease burden and lead to substantial health impairment and economic loss.

It is essential to understand the epidemiology of psychiatric disorders to plan and implement effective mental health policy and programs. Epidemiological surveys estimating the prevalence of ADs in developing countries like India are limited. Most of them are confined to certain geographical regions, and cannot be generalized because of their small sample sizes, lower response rates, and methodology limitations.[3] Indian data from WMHS 2005 estimated the 12-month prevalence of ADs to be 3.41%.[4] This Indian data of WMHS conducted a decade earlier at eight sites, and reported that the 12-month prevalence is estimated only for three ADs (generalized AD [GAD], panic disorder (PD), and specific phobia). A more extensive nationally representative survey to estimate the current prevalence of ADs was a felt need.

Toward strengthening mental health services across India, National Mental Health Survey (NMHS) 2016 was conducted to understand the burden of mental disorders. It focused on current prevalence estimates, treatment gaps, and the disability of mental disorders in a nationally representative population.

This article focuses on the current prevalence estimate of AD, its correlates, disability, and treatment gap from the nationally representative adult population in India’s NMHS 2016.

MATERIALS AND METHODS

The detailed methodology of NMHS 2016 is published.[5,6,7] NMHS 2016 was a large-scale, multicentered, nationally representative study conducted among the adult population aged 18 years and above across 12 Indian states (two states from six regions) during 2014–2016. In brief, a multistage, stratified, random cluster sampling technique with random selection based on probability proportional to size at each stage was adopted. The Mini-International Neuropsychiatric Interview (MINI) version 6.0.0[8] was used to diagnose psychiatric disorders, including ADs as per the International Statistical Classification of Diseases, Tenth Revision Diagnostic Criteria for Research.[9] It overcame the two-stage interview required for diagnosis in field surveys. It has other advantages such as the relative ease of training field staff, shorter administration time, and validation in multiple Indian languages. Disability and socioeconomic impact were assessed using Sheehan Disability Scale.[10] A specifically designed questionnaire was used to assess treatment, care-seeking patterns, and socioeconomic impact. Ethical clearance was obtained from the Institute Ethics Committees of National Institute of Mental Health and Neurosciences, Bengaluru, India, and individual state coordinating centers.

In the present study, the four ADs such as GAD, PD, social AD (SAD), and agoraphobia have been included. Current prevalence in this study corresponds to the presence of PD, SAD, or agoraphobia in the past 1 month or the presence of GAD in the past 6 months of the survey (as per MINI 6.0.0).

The weighted prevalence was estimated for ADs to factor unequal probabilities of sampling and nonresponse rates. Estimates are with a 95% confidence interval (CI). Logistic regression was undertaken considering the presence of ADs as the dependent variable and sociodemographic characteristics (i.e., gender, age, education, occupation, marital status, and place of residence) as independent variables for identifying factors associated with ADs. The adjusted odds ratio was calculated using the model reflecting the risk of having ADs for each selected group of participants against the risk in the reference group. IBM SPSS (Statistical Package for Social Sciences) version 27.0 from International Business Machines Corporation, New York, USA was used for all analyses.[11]

RESULTS

Among adults >18 years across 12 states with an individual response rate of 88%, 34,802 adults were interviewed. Table 1 shows the sociodemographic characteristics of the total NMHS sample and individuals with ADs.

Table 1.

Comparison of sample characteristics between total National Mental Health Survey data and anxiety disorders

	Total, n (%)	Anxiety disorders, n (%)
Total	34,802	852
Age group
18-29	11,848 (34)	239 (28.1)
30-39	7062 (20.3)	169 (19.8)
40-49	5854 (16.8)	162 (19)
50-59	4448 (12.8)	132 (15.5)
60 and above	5590 (16.1)	150 (17.6)
Gender
Male	16,585 (47.7)	302 (35.4)
Female	18,217 (52.3)	550 (64.6)
Education
Illiterate	8409 (24.2)	243 (28.5)
Primary	6160 (17.7)	179 (21)
Secondary	5722 (16.4)	121 (14.2)
High school	6493 (18.7)	147 (17.3)
Preuniversity/vocational	3873 (11.1)	87 (10.2)
Graduate/postgraduate/professional	4044 (11.6)	70 (8.2)
Others*	101 (0.3)	5 (0.6)
Occupation
Working	16,800 (48.3)	367 (43.1)
Not working	17,842 (51.3)	483 (56.7)
Not known	160 (0.5)	2 (0.2)
Marital status
Never married	6512 (18.7)	118 (13.8)
Married	25,980 (74.7)	658 (77.2)
Widowed/divorced/separated	2144 (6.2)	72 (8.5)
Others*	166 (0.5)	4 (0.5)
Residence
Rural	23,957 (68.8)	526 (61.7)
Urban nonmetro	6601 (19)	123 (14.4)
Urban metro	4244 (12.2)	203 (23.8)

*Others category refers to not known/not applicable

Among the age groups, the highest ADs distribution was among the 18–29 age group (28.1%) followed by the 30–39 age group (19.8%). Concerning education, illiterates (28.5%), and those with primary schooling (21%) were highest, followed by high school (17.3%) and secondary (14.2%) education. While 43.1% were employed, 77.2% of individuals with ADs were married, and 61.7% of individuals with ADs resided in rural areas.

The overall weighted prevalence of current ADs in the adult population was found to be 2.57% (95% CI: 2.54–2.60). The prevalence among females was 3.01% (95% CI: 2.78–3.27) and 1.8% (95% CI: 1.62–2.03) among males. Within the individual ADs, the prevalence ranged from 1.6% (agoraphobia) to 0.57% (GAD). Risk factors identified were female gender, 40–59 age group, and urban metro cities [Table 2].

Table 2.

Multiple logistic regression of sociodemographic correlates of anxiety disorders

	AOR	95% CI
Gender
Male (reference)	1.0
Female	1.6a	1.35-1.9
Age
18-29	0.85	0.66-1.09
30-39	0.95	0.74-1.21
40-49	1.04	0.81-1.33
50-59	1.13	0.88-1.46
>60 (reference)	1.0
Education
Illiterate	1.38a	1.02-1.86
Primary	1.55a	1.15-2.08
Secondary	1.2	0.88-1.63
High school	1.26	0.93-1.69
Preuniversity	1.41a	1.02-1.94
Graduate (reference)	1.0
Occupation
Employed (reference)	1.0
Unemployed	1.02	0.86-1.2
Marital status
Single (reference)	1.0
Married	1.13	0.89-1.43
Separated/widowed	1.14	0.79-1.65
Residence
Rural (reference)	1.0
Urban nonmetro	0.89	0.73-1.1
Urban metro	2.6a	2.2-3.1
Household income
Lowest	1.88a	1.49-2.37
Second	1.58a	1.24-2.01
Middle	1.49a	1.17-1.89
Fourth	1.24	0.98-1.58
Highest (reference)	1.0

^aShows the significant odds (P<0.05). AOR – Adjusted odds ratio; CI – Confidence interval

The overall treatment gap was estimated to be 82.9%. The treatment gap for females and males was 80.5% and 84.2%, respectively. Among residence, urban nonmetros had a higher treatment gap (87.8%) compared to rural (83.5%) and urban metros (78.3%).

To calculate disability in ADs, respondents with both severe mental disorders and ADs have been removed as it can be a potential confounding factor [Table 3].

Table 3.

Disability associated with anxiety disorders

Self-reported disability (n=798)	Any disability, (%*)				No disability, (%)
	Mild	Moderate	Severe	Extreme
Disability at work	58.90	24.80	11.80	4.30	44.40
Disability at social life	54.10	26.80	14.60	4.30	42.60
Disability at family life	51.80	31.70	10.30	6.24	43.90

*Of any disability, percentages were calculated for mild, moderate, severe, and extreme disability, missing data (n=7)

DISCUSSION

The NMHS 2016 included a representative national population, thus overcoming several limitations of previous epidemiological studies in India through its rigorous methodology. ADs being one of the most common psychiatric disorders worldwide,[12] each country needs to estimate ADs and their impact on planning and implementing effective mental health services.

Current prevalence and distribution of anxiety disorders

The current weighted prevalence of total ADs was 2.57% among the adult general population in the present survey. Among ADs, agoraphobia was most common (1.6%), followed by GAD (0.57%).

Previous Indian epidemiological studies have shown a wide range of prevalence, and two meta-analyses of select Indian studies reported prevalence estimates to be 5.6% and 16.5%.[13,14] However, most of the included studies in the meta-analysis were single-site studies and less rigorous, making comparisons and generalization difficult. According to Indian WMHS 2005 (published in 2017), the 12-month prevalence of ADs was estimated to be 3.41%.[4] The differences could be due to the methodology per se, or the types of ADs included for the WMHS (which assessed only three ADs: Specific phobia, PD and GAD). Prevalence of specific phobia was highest at 2.47%, followed by PD 0.52% and GAD 0.34%). The present survey estimates of current prevalence essentially reduce recall bias and are thus more accurate. The present survey has also estimated the prevalence of agoraphobia and social phobia, which were not included in the WMHS survey.

In comparison to other countries, the 1-month prevalence rate of ADs (equivalent to current prevalence) was estimated to be in the range of 2.3% (in Mexico) to 10.3% (in the USA).[15,16,17] A systematic review and meta-analyses of the global prevalence of ADs show that about one in 14 people (7.3%) has ADs at any point in time.[12]

Females are 1.67 times more affected compared to men, which is consistent with other studies.[18] The factors for female preponderance in ADs could be poor social support, differential access to care, and possibly an increased tendency of “worrying.”[19,20] In addition, women are also likely to relapse more frequently compared to men.[21]

The role of urbanicity in the development of ADs is well-documented. This survey also found that people living in urban metros are 2.23 times more likely to be diagnosed with ADs compared to rural areas. A meta-analysis of studies on urban–rural differences in mental disorders showed a 21% increase in ADs in urban areas.[22] Postulated reasons are overcrowding, poor social support, social isolation, migration, and poor quality of life.[23,24]

The lower rates of prevalence observed in the present survey could also be an underestimate: exclusion of specific phobias and unspecified ADs is a key issue apart from the noninclusion of culture-specific anxiety syndromes. The current prevalence time frame could have resulted in the under-reporting of the recurrent illness. Although there has been increasing awareness of psychiatric disorders in recent decades, stigma and other social factors continue to contribute to systematic under-reporting.[25]

Prevalence of anxiety disorders in comparison with depressive disorders

Interestingly, the prevalence of ADs (2.57%) is roughly equal with depressive disorders in the NMHS (2.68%).[26] This should draw the attention of the public and policymakers toward giving equal weightage to both depression and ADs in the health programs.

The treatment gap in anxiety disorders

The present survey shows the treatment gap of ADs to be 82.9%. Consistent with common mental disorders, the treatment gap was higher in urban than rural and metro cities. Urban nonmetros have not received due attention concerning mental health. While rural areas are better served by primary health care, metro cities have easy accessibility to tertiary care.[27]

Worldwide, the treatment gap in mental disorders is a challenging conundrum and is not solely mediated by a deficiency in the availability or delivery of mental health services. Even in developed countries such as the United States, around two-third of mental disorders are not treated.[28] It could be attributed to several barriers such as low perceived need, health-care costs, stigma, and inadequate resources.[29] Given the pressing need to take appropriate remedial measures, WHO’s mental health gap action program[30] is a step forward. Evidence suggests that these measures could be feasible and valuable in low and middle-income countries.[31,32]

Among the efforts to manage the COVID pandemic, access to mental health care becomes even more complicated and demands newer ways of delivering health-care services such as telemedicine,[33] which is the need of the hour. Another way forward is targeting the functional treatment gap (bridging missed opportunity in diagnosis at primary care) using a digitally driven primary care psychiatry strategy is pragmatic.[34]

Disability and socioeconomic impact of anxiety disorders

Our survey found that around 50%–55% of people with ADs have different grades of disability and one in two persons with AD report moderate-to-severe grades of disability. Studies reporting increased work absenteeism among people with anxiety and depressive disorders have been consistent.[35] Apart from disability at work, people with ADs report a significant disability in family and social life. Thus, ADs impact one’s functionality which significantly needs no repetition. Since ADs most commonly affect those in the working-age group and their functionality, they impact individual and national economic burden and could be substantive.

Limitations

This study is not ruling out the possibilities of underestimation of ADs given the social acceptance of shyness of SAD, data on specific phobia not collected, prevailing higher stigma of mental illness among the general population, etc. There is also the possibility of missing more disabled subthreshold or subsyndromal ADs. Few important clinical variables, such as the age of onset and duration of illness, are not collected in NMHS. This study does not include obsessive–compulsive and posttraumatic stress disorders because of the contemporary understanding of ADs to exclude them.

CONCLUSIONS

This nationally representative study draws attention to an urgent need to plan, develop, and implement a service strategy for managing ADs that pose a similar or possibly a greater burden than depressive disorders. The wider treatment gap further reiterates the call for urgent action. The results from NMHS 2016 also draw our attention to planning and undertaking health impact studies, especially for psychiatric disorders which continue to be underdetected and undertreated despite the huge disease burden.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.