Nationally representative epidemiological study of social anxiety disorder from India

Abstract

Background:

Social anxiety disorder (SAD), also termed as social phobia, is a disabling psychiatric condition with limited epidemiological research on it in India. This study, using data from the National Mental Health Survey (NMHS), 2016, is the first to explore its current prevalence and associated factors in India.

Materials and Methods:

The NMHS in India used a comprehensive population-based study with subjects selected through a multistage stratified random cluster sampling technique across 12 states. The study included 34,802 adults interviewed with the Mini-International Psychiatric Interview 6.0.0. Firth penalized logistic regression (FPLR) was used to estimate covariate odds ratios (ORs), and the treatment gap for SAD and disability measured using Sheehan's disability scale was calculated.

Results:

The study found a 0.47% prevalence of SAD, with an average age of 35.68 years (standard deviation (SD) = 15.23) among those affected. Factors, such as male gender, unemployment, and living in urban areas, were associated with higher odds of SAD, while the elderly had lower odds. A significant proportion of individuals with SAD experienced disability in work (63%), social life (77%), and family life (68%). They spent a median of ₹ 2500 per month on treatment and had a high rate of comorbid psychiatric disorders (58%). The treatment gap was substantial at 82%.

Conclusions:

A considerable portion of India's population (approximately >65 lakhs) is affected by SAD. Surprisingly, the NMHS 2016 report indicates a higher risk of SAD among males compared with females, a trend that warrants further investigation. SAD in India is linked to significant disability and a considerable treatment gap, emphasizing the need for innovative approaches to address this large, affected population, especially in light of the scarcity of mental health professionals.

INTRODUCTION

Anxiety disorders are one of the highly prevalent common psychiatric disorders that contribute substantially to the global burden of disease and years lived with disability.[1] Social anxiety disorder (SAD) (also known as social phobia) is one of the most common anxiety disorders characterized by an intense fear of social scrutiny and negative evaluation by others.[2,3] Over the past few decades, it has evolved as a standalone disorder unique from the other phobias.[4]

In worldwide studies, the prevalence of SAD is as high as 13% in the general population, having an adolescent onset, mostly of chronic course, and is almost invariably associated with higher comorbid rates of other anxiety, depression, or substance use disorders.[4] The US National Comorbidity Survey Replication reported a one-year and lifetime prevalence of SAD to be 7.1% and 12.1%, respectively.[5,6] The US National Epidemiological Survey reported a lesser prevalence of 2.8% and 5% for one year and lifetime, respectively,[7] when more stringent criteria for the diagnosis of SAD were used. Overall, the lifetime prevalence of SAD ranges from 2.8 to 13% reported in published literature from epidemiological studies.[8] These differences are attributed to variation in the data collection methodology from the individual studies and interethnic variation in SAD.[8] The gender differences in published literature indicate that SAD is more common and severe in women; however, men and women have comparable clinical courses and outcomes.[9] Females are at 1.2–1.5 times more risk of SAD than males over their lifetime.[8]

It has high rates of disability, especially related to social and work domains, and is associated with decreased work output and quality of life.[10] Despite lifetime chronic nature of this disorder, only every third person gets treated for SAD.[4] Although it is highly prevalent, it remains under-recognized and undertreated. There has been substantial criticism over the lack of good quality scientific literature on the study of SAD worldwide.[11]

Despite the adolescent nature of the onset of symptoms, SAD is prevalent across age groups and is one of the most common anxiety disorders worldwide.[12] Persons with SAD are more likely to be single, have lower socioeconomic status, and are less educated.[13] It is associated with higher rates of substance use (especially alcohol) and depression. Additionally, the burden of disease associated with SAD is likely to be much higher than the current estimates, given that there is substantial impairment associated with subthreshold social anxiety that usually goes undiagnosed.[14]

Despite the above, there is a scarcity of epidemiological data on SAD in India from the general population. Therefore, there is a need to study the epidemiology and its determinants of SAD in Indians to develop effective public health policies.

MATERIALS AND METHODS

The National Mental Health Survey of India (NMHS) 2016 is the largest and most ambitious survey of mental illnesses in India. This study aimed to examine the current prevalence and correlates (including its disability and treatment gap) and the economic burden of SAD from the NMHS 2016.

A detailed description of the methodology of NMHS is published elsewhere[15] In brief, NMHS 2015–16 was the most extensive population-based study conducted across 12 states of India, using a multistage, stratified, random cluster sampling technique based on probability proportionate to size at each stage. All adults 18 years and older were included. All eligible adult respondents within the identified household were interviewed. A nonresponder was defined as a person who could not be interviewed even after three visits. A rigorous attempt was made to include all eligible people in the study. Qualitative and quantitative methods were employed to analyze the data from 39,532 individuals across 720 clusters. Given that the response rates were 88% at individual levels, the final sample size of the completed population surveyed was 34802. Mini-International Neuropsychiatric Interview 6.0.0 is used to diagnose psychiatric disorders.[16] International Classification of Diseases and Diagnostic Criteria for Research version 10 were used to classify the disorders.[2] The current prevalence of SAD in the last month was considered to estimate the cross-sectional prevalence and its disability.

The sociodemographic data, health treatment, care, treatment gap, monthly income, duration of illness, monthly treatment expenses, distance traveled, and expense per visit from eligible participants were collected from specially designed questionnaires as a part of the NMHS methodology. In particular, the treatment gap was assessed using a single question. The Sheehan Disability Scale (SDS) was used for assessing disability.

Ethical clearance was obtained from the Institute Ethics Committee (IEC) of “National Institute of Mental Health and Neurosciences,” Bengaluru, India, and corresponding IECs of partner institutions. Informed consent was obtained from all the respondents. The weighted prevalence estimates are reported considering the unequal probability of selection and nonresponse rates.

Although in cross-sectional study designs, the prevalence ratios are preferred over the odds ratio (OR), this study reports the odds ratio given that the statistical method chosen was logistic regression that computes ORs[17] and has greater acceptability in epidemiology and public health research. Moreover, ORs converge on prevalence ratios for rare outcomes.[18] Modeling the data for binomial outcome (persons with SAD and without SAD) was essential to generate the OR from logistic regression analysis. To overcome the problem of statistical separation and monotone likelihood, Firth's bias-reduced logistic regression was chosen to estimate the association between covariates of interest and diagnosis of SAD.[19] This statistical procedure reduces first-order bias in maximum-likelihood estimates of beta and small sample size and provides finite, consistent estimates. Descriptive statistics were used to compute disability, and the average duration of illness and the time between onset to consultation in persons who sought treatment are reported. All analysis was performed using the R integrated suite of software facilities for statistical analyses.[20]

RESULTS

The current or cross-sectional (last 1 month) weighted prevalence of SAD was 0.47% (95% confidence interval (CI): 0.46–0.48). The prevalence of SAD among male respondents was 0.54% (95% CI: 0.52–0.56) and among female respondents was 0.41% (95% CI: 0.39–0.42).

Table 1 shows the sociodemographic characteristics of the total NMHS sample and persons with SAD. The table demonstrates the representative nature of the sample in persons with SAD and provides estimates of frequency distribution with percentages in basic sociodemographic variables.

Table 1.

Comparison of sample characteristics between total sample and persons with social anxiety disorder

	Sample n (%)	Social anxiety disorder n (%)
Total	34802	125
Age group^
18 to 29	11848 (34)	51 (40.8)
30 to 39	7062 (20.3)	30 (24.0)
40 to 49	5854 (16.8)	16 (12.8)
50 to 59	4448 (12.8)	17 (13.6)
60 and above	5590 (16.1)	11 (8.8)
Gender^
Male	16585 (47.7)	60 (48)
Female	18217 (52.3)	65 (52)
Education
Illiterate	8409 (24.2)	30 (24.0)
Primary	6160 (17.7)	19 (15.2)
Secondary	5722 (16.4)	21 (16.8)
High school	6493 (18.7)	27 (21.6)
Preuniversity/vocational	3873 (11.1)	12 (9.6)
Graduate/postgraduate/professional	4044 (11.6)	15 (12.0)
Not available, applicable, or missing information	101 (0.3)	1 (0.8)
Occupation
Working	16800 (48.3)	54 (43.2)
Not working	17842 (51.3)	70 (56.0)
Not available, applicable, or missing information	160 (0.5)	1 (0.8)
Marital status
Never married	6512 (18.7)	33 (26.4)
Married	25980 (74.7)	85 (68.0)
Widowed/divorced/separated	2144 (6.2)	7 (5.6)
Not available, applicable, or missing information	166 (0.5)	0
Residence^
Rural	23957 (68.8)	70 (56.0)
Urban non-metro	6601 (19)	23 (18.4)
Urban metro	4244 (12.2)	32 (25.6)

^Indicates no missing information in these categories

The highest prevalence of SAD was found among the 18–29 age group (41%), followed by the 30–39 age group (24%). The mean age of persons with SAD was 35.68 years (SD ± 15.23 years).

Among the persons with SAD, 24% were illiterates; 21% were educated till primary school, and around 21.6% were educated till high school. Among occupation categories, 43.2% were employed, and the rest were unemployed. Among marital status, 68% of individuals with SAD were married, and 26.4% were single. 56% of individuals with SAD resided in rural areas.

For penalized logistic regression analysis purposes, we chose the following categorized variables for prediction—gender, age, education, occupation, marital status, and current residence. Table 2 represents the results of the Firth penalized logistic regression (FPLR) analysis for persons with SAD.

Table 2.

Firth penalized logistic regression analysis for factors associated with a social anxiety disorder (n=120 after exclusion of missing data)

	B Coef	SE (Coef)	Chi-sq	P	Odds ratio Exp (B)	Confidence interval
						Lower 0.95	Upper 0.95
(Intercept)	-5.630	0.370	Inf	<0.001	0.004	0.002	0.007
Gender (female (Ref))
Male	0.474	0.217	4.517	0.033	1.606	1.037	2.500
Age (18–29 (Ref))
30–39 years	0.113	0.258	0.182	0.669	1.120	0.664	1.881
40–49 years	-0.351	0.3147	1.228	0.267	0.703	0.367	1.305
50–59 years	-0.150	0.320	0.214	0.643	0.860	0.445	1.615
60–69 years	-0.950	0.378	6.578	0.010	0.386	0.174	0.804
Education (illiterate (Ref))
Primary school	-0.387	0.293	1.699	0.192	0.679	0.372	1.213
Secondary school	-0.294	0.291	0.991	0.319	0.745	0.412	1.328
High school	-0.233	0.277	0.677	0.410	0.792	0.453	1.381
Preuniversity or vocational training	-0.690	0.359	3.763	0.052	0.501	0.237	1.007
Graduate and above	-0.631	0.343	3.373	0.066	0.532	0.261	1.043
Occupation (working (Ref))
Not working	0.444	0.216	4.018	0.045	1.559	1.010	2.424
Marital status (unmarried (Ref))
Married	-0.241	0.260	0.812	0.367	0.785	0.467	1.333
Widowed/divorced/separated	0.120	0.477	0.059	0.807	1.128	0.407	2.840
Residence (rural (Ref))
Urban non-metro	0.208	0.239	0.699	0.403	1.232	0.746	1.959
Urban metro	1.024	0.214	18.768	<0.001	2.786	1.787	4.251

The FPLR revealed a likelihood ratio test = 36.6312 on 15 df, P = 0.001432227, n = 34385; Wald test = 40.854 on 15 df, P = 0.0003365968.

These results indicate that the male gender had a higher OR (1.606) of having SAD, and elderly had a lower OR of having SAD (0.386). Persons who were not working had a higher OR (OR 1.559) of suffering from SAD. Persons in the urban metro had higher odds of suffering from SAD (OR = 2.786).

Disability

The current disability associated with SAD is summarized in Table 3. In all persons with SAD, approximately two-thirds of patients have disabilities in all domains, that is, work, social life, and family life. As expected, the presence of social disability was more than (76.7%) disability in other categories. Approximately more than 40% of patients suffering from SAD suffer from a moderate, severe, or extreme disability.

Table 3.

Current disability in patients with social anxiety disorder

Self-reported disability	No disability % (n)	Any disability % (n)
		Mild	Moderate	Severe	Extreme
Disability at work	37.5 (45)	30.8 (37)	19.2 (23)	8.3 (10)	4.2 (5)
Disability in social life	23.3 (28)	35 (42)	25.8 (31)	12.5 (15)	3.3 (4)
Disability at family life	31.7 (38)	33.3 (40)	20.8 (25)	9.2 (11)	5 (6)

*Disability is assessed using the Sheehan Disability Scale.[21]. *The percentage of grading of any disability in Table 3 is calculated with a denominator of n=120 (excluding five missing values from 125 persons with social anxiety disorder). ^—Only current prevalence was used for the analysis

In the pure SAD group without any mental comorbidity (i.e. without any comorbidity except nicotine use, n = 52, after excluding one missing value), more than half of all persons with SAD are likely to experience mild-to-moderate disability in all domains of functioning. As expected, the disability is more severe in the social domain (73.1%) than in work (55.8%) and family life (53.8%).

Illness details and impact: Table 4 reveals the illness details, its impact on the patients, and their families. The median household income was ₹10000, and the median monthly expenditure of a person with SAD was 25% of the same at ₹2500 (after excluding other comorbidities except for nicotine use, it was ₹1500). The median duration of illness at the cross-sectional evaluation was 5 years. The patients who availed treatment from a healthcare resource did so after a median of 3 months. A single visit to the healthcare provider and availing the treatment was burdening the patients and their family with ₹1500. On average, persons with SAD had to travel 10 kilometers to avail treatment for SAD.

Table 4.

Illness details and impact of social anxiety disorder

Illness and impact	Median (Q1, Q3)
Entire household’s monthly income	₹10000 (5000, 18000)
Duration of illness	60 months (20, 144)
Monthly expenditure due to social anxiety disorder	₹2500 (1000, 5000)
Average distance traveled to seek care^	10 kilometers (2, 67.5)
Average expenditure per visit^	₹1500 (500, 3000)

^Valid for only those who sought treatment; ₹—rupees

Treatment gap: Only 18.4% (n = 23 of 125 patients) were on treatment. This association needs exploration in future epidemiological studies. The elderly population had lower odds of suffering from SAD than the younger population. This association must be viewed with caution rather than the simplistic assumption of elderly age being regarded as a protective factor, as this association may also be driven by a higher prevalence of SAD in younger age categories. Being predominantly a disorder in the young, SAD is associated with disability and decreased economic productivity. Anxiety disorders in the elderly are associated with increased morbidity irrespective of the increased chance of such disorders in the younger population and therefore need appropriate and prompt treatment.[22]

The finding that SAD associated with higher odds of persons living in urban metros (OR: 2.78) contrasts with the finding from other epidemiological surveys.[7] Anxiety and depression in urban metro residents are attributed to estrangement from society, social dislocation due to migration, and poor belongingness, a unique set of social determinants and cultural challenges. There are complex issues that underlie this seemingly straightforward association.[23] Understanding these challenges is essential to delivering primary prevention and early intervention.

Comorbidity

About 58% (72 of 125) persons suffered from one or more comorbid psychiatric disorders [Table 5]. The most common comorbidity associated with SAD was agoraphobia (36%, n = 45), followed by depression (19.2%, n = 24) and alcohol use disorder (AUD) (13.6%, n = 17). Other common mental disorders, such as obsessive–compulsive disorder, post-traumatic stress disorder, panic disorder, and generalized anxiety disorder, were associated with 6–8% of these patients. Four patients were diagnosed with comorbid psychosis spectrum disorder and three patients with bipolar affective disorder.

Table 5.

Current comorbidities associated with social anxiety disorder

Disorder	Percentage [n]
Agoraphobia	36% [45]
Tobacco use disorder	28% [35]
Depression	19.2% [24]
Alcohol use disorder	12.8% [16]
Obsessive–compulsive disorder	6.4% [8]
Post-traumatic stress disorder	5.6% [7]
Panic disorder	5.6% [7]
Generalized anxiety disorder	4.8% [6]
Psychosis	3.2% [4]
Bipolar affective disorder	2.4% [2]

DISCUSSION

This is the first epidemiological study from India with a large sample size that is representative of the general population to report on SAD. Given the robust methodology of NHMS, these findings have greater ecological validity with important public health implications.

The current prevalence of all anxiety disorders from NMHS was already published as 2.57%.[24] The current prevalence (30 days) of SAD in India was 0.47%, almost matches with prevalence data of SAD among other low- and middle-income countries (LMICs).[25] SAD is more prevalent in the developed world, such as the United States of America and New Zealand, where it is as high as 2.5–3%.[25] The elevated prevalence observed in high-income countries presents an intriguing scenario that invites a deeper examination. It could potentially serve as an indicator of a disparity in reporting and the underestimation of SAD in LMICs. This divergence in prevalence rates between high-income and LMICs could be a reflection of the underdiagnosis or underreporting of SAD in LMICs, which could be due to various factors, including limited access to mental healthcare services, stigma associated with mental health issues, or simply a lack of awareness. However, an alternative perspective arises when considering the relatively lower prevalence in LMICs. It raises the question of whether this discrepancy may be linked to the presence of more resilient social and community networks in LMICs. It is possible that these countries have robust support systems, social structures, or cultural factors that either mitigate the development of SAD or facilitate coping mechanisms. Investigating these social and cultural dynamics in LMICs becomes essential to understand whether they indeed contribute to lower prevalence rates. In essence, this observation underscores the need for comprehensive research to unravel the intricate interplay of various factors, including healthcare infrastructure, cultural norms, awareness, and social networks, in shaping the prevalence of SAD in different regions, ultimately allowing for more targeted interventions and strategies.

Worldwide, there is a consistent pattern of increased risk of SAD in young age, those who are unemployed, in females, and those who are single.[25] This study finding resonates with many of these with a few exceptions that are detailed below. The notable observation in this study reveals that males exhibit a higher likelihood of experiencing SAD. This intriguing finding offers a gateway to a more extensive exploration, as it piques curiosity regarding the underlying causes and contributing factors. One potential avenue for this exploration lies in the cultural context. It is possible that SAD in females may be falsely normalized or mistaken for culturally normative shyness. In many societies, particularly those with traditional gender norms, shyness in females may be perceived as socially acceptable or even desirable. As a result, individuals, especially females, may be less likely to recognize and report their symptoms of SAD, leading to underdiagnosis. This dynamic raises questions about the role of culture and gender norms in shaping the manifestation and perception of SAD. A comprehensive exploration of these cultural and gender-related factors is essential to fully grasp the reasons behind the observed gender-based differences in SAD prevalence. Additionally, it can inform more targeted and effective strategies for diagnosis, treatment, and support, particularly among females who might be experiencing SAD but are not seeking help due to societal expectations.

Educational status was not a significant mediator of SAD. This is interesting because lower education is an expected risk factor for 12-month and lifetime prevalence of SAD from the previous studies.[25] However, it is not a predictor of the current prevalence of SAD as per our analysis.

Unemployment is a significant well-replicated association with SAD.[25] This relationship may be complex with bidirectional causality.

The findings of this study did not reveal an elevated likelihood of SAD among individuals who are single. However, it is essential to interpret this result within the context of the study's limitations. One potential reason for this lack of association could be the relatively small sample size used in the study, which may have limited its ability to detect these subtle associations. Furthermore, the absence of an increased OR among single individuals may also be influenced by sociocultural factors.

The increased OR of SAD in persons from urban metros is interesting, yet not unexpected. This has been reported in earlier epidemiological studies as well.[26] This may be primarily driven by many factors, such as redistributed social organization for persons migrating to urban metros, increased nuclear family prevalence, and cultural divides, that prevent social acceptance. Moreover, social integration in rural communities may be more natural as compared to urban communities where employment is often the only bond shared by persons from different backgrounds.[27]

Comorbidities are the norm rather than the exception with SAD.[25] The prevalence of comorbid psychiatric disorders among SAD sample in our study is around 57.6%. These figures resemble the 12-month estimates of comorbidity reported with SAD in the World Mental Health Survey Initiative (WMHSI).[25] The same study also reported that SAD is most likely to be comorbid with other anxiety disorders. In this study, agoraphobia was the most commonly comorbid psychiatric disorder with SAD [36%]. This is not new, given their common origins,[28] shared genetic vulnerabilities,[29] and has been noted earlier in epidemiological studies as well.[30] This association and concurrent risk of both disorders have been documented earlier in a European cohort where persons with agoraphobia had 10.4 odds of having concurrent SAD.[31] Depression was the second most common comorbid psychiatric disorder. Depression comorbid with SAD is associated with increased dysfunction, disability, and poorer outcomes.[32] Other studies report the prevalence of comorbid depression between 20 and 70% for persons with SAD.[33] Initial SAD is associated with 5.7-fold higher risk of new-onset depression that usually follows 2 years or more after onset of SAD.[34] This association must be highlighted to prevent diagnostic overshadowing and the risk of attenuated anxiety treatment response.[35] Our analysis found that AUD was the third most common psychiatric comorbidity with SAD. The evolutionary theory of high comorbid rates of SAD and AUD proposes that persons with SAD seek out alcohol to artificially alter sustained negative affective states or dysfunction.[36] In our sample, alcohol use was present in 13.6% of patients. Untreated social anxiety can be a precipitating and perpetuating factor for alcohol dependence syndrome and lead to an increased risk of relapse. Therefore, SAD perhaps must be treated adequately and appropriately in the early stages of illness to mitigate the future risk AUD.

Disability, treatment gap, treatment cost, and affordability

SAD is a chronic disabling condition and surpasses social disability seen in chronic medical conditions and depression.[37] The high rates of treatment gap are mediated by dual perils of lack of treatment availability and the lack of perceived treatment due to over-normalization of pathological social anxiety as normal shyness, in addition to other factors. However, appropriate and adequate treatment of SAD is vital, as it predominantly affects the younger generations and therefore is associated with considerable disability-adjusted years (DALY) and dysfunction.

In this study, the treatment costs ₹2500 per month. Higher rates of remaining unemployed and a rural background in a substantial proportion of individuals add to the disorder's economic burden. Moreover, the cost of every visit to healthcare provider was ₹1500 and on average 25% of the median household income per month was being spent toward the treatment of persons with SAD (affordability). As expected, the presence of comorbidities increased the median monthly expenditure. This median expenditure is that of severe mental illnesses, such as schizophrenia (₹1000 per month) and depression (₹1500 per month) published in the NMHS.[38] This highlights the huge burden of treatment that is borne by individuals with SAD and their caregivers. The average travel of 10 kilometers to access care also can lead to substantially increased indirect costs of treatment and have deleterious effect in the longer run. Therefore, treatment for SAD must be made available in the nearest health and wellness center to ensure reduction in the treatment gap and to ensure adherence.

Earlier studies demonstrate that less severity of the anxiety, less comorbid additional psychopathology and stress, and employment are favorable predictors of remission in SAD.[39] Our study demonstrates an increased odds of SAD in persons not employed. The dysfunction associated with social and occupational domains may bidirectionally mediate this association due to SAD.

Limitations and future directions

This national cross-sectional survey was analyzed as a binomial outcome specification based on the presence or absence of SAD. This adaptation was vital to facilitate the computation of odds using appropriate statistical methods. It is important to emphasize that the NMHS epidemiological survey's methodology is well-established and sound. Leveraging this foundation, our current analysis not only contributes critical insights but also serves as a stepping stone for shaping future research endeavors.

Looking ahead, it is prudent to acknowledge that psychiatric disorders with low prevalence warrant specialized attention. Therefore, exploring these disorders through separate case–control prospective designs becomes imperative. This approach enables a more nuanced examination of risk factors, enhancing our understanding and paving the way for more targeted and comprehensive investigations in the realm of mental health research.

CONCLUSIONS

The current prevalence of SAD in India stands at approximately 0.47%. Notably, individuals residing in urban metropolitan areas exhibit a higher susceptibility to this condition. Timely recognition and intervention for SAD are imperative. Equally crucial is the early identification and management of concurrent anxiety, substance use, and depressive disorders, as this can substantially narrow the treatment gap. It is important to underscore that SAD predominantly affects the economically active age group, and it follows a chronic and fluctuating course, resulting in significant disability and financial burdens for individuals and society.

Moreover, the intriguing discovery of an increased likelihood of SAD among males demands more rigorous scientific exploration. Subsequent epidemiological research should delve into various aspects, including the lifetime prevalence of SAD, different subtypes of the condition, its trajectory and outcomes, and its interplay with other internalizing spectrum disorders. Such investigations are essential for a more comprehensive understanding of SAD in the Indian context and can inform targeted strategies for prevention and treatment.

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 the six regions and interviewed 39,532 individuals (http://indianmhs.nimhans.ac.in). The funder had no role in implementation, data acquisition, data analysis, interpretation, and write-up of the paper.

Conflicts of interest

There are no conflicts of interest.