A close critical look of India's National Mental Health Survey 2016 Authors: Rajeshkrishna P. Bhandary, Soyuz John, Anil Kumar M. Nagaraj, Samir K. Praharaj, Chythra R. Rao, Muralidhar M. Kulkarni, Sheena K. Agarwal Published: 2023 ABSTRACT The National Mental Health Survey 2016 (NMHS 2016) was a large epidemiological study, one of its kind, conducted by the National Institute of Mental Health and Neuro Sciences (NIMHANS), Bengaluru to overcome the shortcomings of the previous surveys. The detailed report of the study is available in two parts- ‘mental health systems’ and ‘prevalence, pattern and outcomes’. Though done comprehensively, there are some inevitable limitations. The private sector, a substantial health care provider in the country was not a participant in the survey. Though MINI version 6.0 is a standard and structured instrument, it does not cover many commonly encountered mental illnesses like somatoform disorders. Further, the methodology of the survey makes it difficult for an accurate calculation of the prevalence of individual major psychiatric disorders. The survey has been appraised using a standard checklist for prevalence studies. The detailed qualitative data has not been shared in the report. The contribution of the traditional indigenous systems of healthcare and accessibility of services in rural areas have not been elaborated. Thus, the need for a comprehensive and culturally sensitive assessment tool, involvement of the private sector, and enhancing funding provision to improve the infrastructure are emphasized as future directions for the subsequent phases of the survey. INTRODUCTION OF NMHS Psychiatric disorders are increasingly recognized in India as major public health problems, contributing to morbidity and disability. From the earliest survey of the prevalence of psychiatric disorders in India in 1960[1] to the subsequent ones,[2345] there are differences in the reported prevalence rates and the rates of rural versus urban psychiatric morbidity. To overcome the limitations such as methodological issues, case definition, case identification, and systematic under-reporting in those surveys, and to further strengthen the mental health programs and develop data-driven programs, the Ministry of Health and Family Welfare, Government of India commissioned the National Institute of Mental Health and Neuro Sciences, Bengaluru, to conduct the National Mental Health Survey (NMHS), which was done during 2015-2016. The detailed report of NMHS 2016 is available in two parts.[67] Although done comprehensively, there are some limitations. The objective of this article is the critical analysis of the NMHS 2016. Critical analysis of methodology A master protocol was prepared after a rigorous process as elaborated in the survey report. However, while introducing the existing mental healthcare in India, due importance has been given to the Government institutions and Government-run programs like National Mental Health Program. The importance of the private sector has been acknowledged in the recommendations of the report (page xxviii) that “more than 85% of medical care occurs in the private nongovernmental sphere”. However, the private sector was not given an opportunity to be a part of the survey. In fact, the National Mental Health Policy of India 2014 has recognized the importance of intersectoral collaboration between Government and non-Governmental sectors to meet the needs of the mentally ill.[8] The country has nearly 300 private medical colleges and several private psychiatric hospitals/nursing homes, NGOs, and rehabilitation centers. Thus, not incorporating the contribution of the private sector in the mental health system assessment is an important shortcoming. The total sample size from 12 states was 34,802 adults and 1,191 adolescents. One of the reviews on NMHS states that, although this appears to be an adequate sample for the general mental health morbidity and burden, the disease-specific total number of cases is smaller.[9] Thus, the survey methodology was not geared to determine the prevalence of severe mental illness to a reasonable level of certainty. Furthermore, although this initial phase of NMHS has not included metro cities with >10 million population, nevertheless, new-age disorders like internet addiction and gaming disorders are not unique to metro cities only. They remain an area of concern and have been duly recognized by the American Psychiatric Association and the World Health Organization. An assessment to identify such novel but not-so-uncommon problems of the contemporary world could have been a part of the survey. The Mini International Neuropsychiatric Interview (MINI) version 6.0 was the main tool used in the study.[10] It is considered to be simple, clear, and easy to administer with compatibility with both ICD 10 and DSM IV, and available in Indian languages. The survey report states that it can be administered with limited training. However, the kappa value of 0.54 on re-interview of 6% of study subjects shows that the degree of agreement is only moderate, reflecting the difference between the assessment by a professional and a trained person. If it was feasible to re-interview a larger sample, the degree of agreement could have been appreciated in a better way. Moreover, MINI 6.0 has excluded somatoform disorders, which is a common mental disorder. Also, it can diagnose 17 different psychiatric disorders only. More importantly, the unspecified/Not Otherwise Specified categories are absent. In the real world, the Not Otherwise Specified category of anxiety disorders and somatoform disorders are commonly encountered. Furthermore, many personality disorders are not included in MINI. These are the key shortcomings of MINI, which could influence the results of the study. However, it is understandable that personality disorder assessment takes a longer time and may not be feasible for epidemiological studies. Nevertheless, by not assessing, we are missing the real morbidity status. Thus, using a simper alternative screening tool for personality disorders could have given some estimate of the problem. MINI remains a structured and standard instrument. However, a comprehensive and culturally sensitive tool, specifically designed and validated for our study population, could have been used considering the massive data that were collected during NMHS which is likely to influence policy decisions. Indian Psychiatric Survey Schedule,[11] a comprehensive Indian instrument for psychiatric diagnosis was indigenously developed and used by RL Kapur in 1974 but was never revised thereafter. Furthermore, owing to the limitations of Field Data Collectors and cultural sensitivity, some important and not-so-uncommon disorders like Antisocial Personality disorder, eating disorders, and organic/drug-induced causes of psychiatric illnesses have been excluded, which can further underestimate the burden of morbidity. Epilepsy, a neurological disorder, has been reported to be included under the NMHS, for operational and technical reasons. However, only generalized seizure disorder was screened, as it is said to be the most common and can be identified easily by Field Data Collectors. However, the analysis and reporting of the comorbidity of seizure disorders such as mood disorders, anxiety disorders, substance use disorders, etc., (except inter-ictal psychosis) has not been examined. These comorbidities are common in epilepsy and can add up to the overall psychiatric burden. Also, complex partial seizure presents with psychiatric comorbidities as commonly as generalized tonic clonic seizure[1213] and was not considered for identification in the survey. The study sample included participants interviewed in households only. Subjects in old age homes, rehabilitation homes, orphanages, and homes for the homeless have a high level of mental morbidity. They are a double-edged sword. Including subjects from all such facilities existing in the clusters might possibly skew the data and excluding them means that we are missing on the actual morbidity data of the community. However, for a more accurate representativeness of the population, samples from those facilities can be considered in the upcoming phases of National Mental Health Program. Critical analysis of results Although it is a large survey of one of its kind, the overall weighted prevalence for any mental morbidity identified was 13.7% lifetime and 10.6% current. This is however in contrast to other surveys including WMHS[14] that recorded higher prevalence in Western and other Asian countries. Also, the outcome of substance and tobacco use is in contrast with previously reported literature from India. The Global Adult Tobacco Survey[15] has shown that tobacco use is more than 40% across states. In the National Family Health survey[16] too, tobacco use was reported to be more than 40%. However, NMHS reports a prevalence of Tobacco use disorder to be 20%, and not just ‘Tobacco use’, unlike other studies. The new tobacco use module used in the survey is an expansion and adaptation of the original Fagerstrom Nicotine Dependence Scale,[17] to include both smoking and smokeless tobacco, making it more stringent. Rather than asking ‘yes’ or ‘no’ screening questions, it has picked up only the problematic pattern of use. However, the survey could have also assessed just the nicotine use, to add more value to the findings. The gender differences mentioned in some disorders like bipolar disorders (0.6% in males vs. 0.4% in females) appear small and probably insignificant. Also, detailed qualitative data have not been shared in the report, although it has a mention in the methodology. Finally, Table 1 represents a critical appraisal of NMHS as per the standard checklist for prevalence studies.[18] However, the checklist being prepared for prevalence studies falls short of an analysis of mental health systems, another major part of NMHS. Table 1Critical analysis of NMHS using prevalence study checklist[18]CriteriaYes/no/unclear/NARemarks1. Was the sample representative of the target population?NoIt was a weighted analysis of the sample from 12 representative states and from select cities and villages only, excluding metro cities with a population >10 million. Representativeness is debatable.2. Were study participants recruited appropriately?YesMultistage, stratified, random cluster sampling technique was employed. However only the household population was recruited and inmates of old age homes, homes for the homeless, etc., were excluded.3. Was the sample size adequate?YesThe sample size was computed with an absolute precision of 2% at a 95% confidence level using a specified formula. The total sample size was 34,802.4. Were the study subjects and the setting described in detail?YesAn adequate explanation of who are the study subjects and the setting (households) is explained in the report.5. Was the data analysis conducted with sufficient coverage of the identified sample?UnclearThis information has not been clearly mentioned in the report6. Were objective, standard criteria used for the measurement of the condition?YesAlthough standard tools were used, MINI has its limitations as several common and important diagnostic entities are omitted, as highlighted in the text and hence the same are not identified in the project.7. Was the condition measured reliably?NoRe-interview of 6% of study subjects using MINI version 6.0 shows that there was a moderate degree of agreement only, reflecting the difference between the assessment by a professional and a trained person. Better agreement would have possibly become evident if more subjects were re-interviewed.8. Was there appropriate statistical analysis?YesStatistical analysis has been elaborated in the report.9. Are all important confounding factors/subgroups/differences identified and accounted for?NoPrevalence rates of some disorders are different compared to other major surveys as mentioned in the text. However, the same has not been discussed in the report.10. Were subpopulations identified using objective criteria?YesDifferent disorders from the sample were diagnosed using MINI Plus and MINI Kid. Further classification based on gender, age, and rural-urban living is done. Critical analysis of NMHS using prevalence study checklist[18] Critical evaluation of Mental Health Systems Assessment The domains covered for assessing the mental health system were comprehensive. While the recommendations in the survey report mention that 85% of the medical care is through the private sector, it also mentions that Governmental agencies are providing the majority of mental healthcare (page 43 of the report on mental health systems). These appear to be two different inferences on the mental health infrastructure and manpower of the country. Furthermore, Indians find herbal medicines and faith healers useful. Thus, information on the services provided by other indigenous systems of healthcare does not seem to be captured in this survey. Healthcare distribution across rural and urban areas may not be the same, even in the Government sector. Most care facilities are likely concentrated in urban areas. The future versions of the survey can share these data comparing urban versus rural infrastructure, in terms of gaps in mental health manpower, mental health beds, rehabilitation, and deaddiction facilities, etc., so that it gives insight into the magnitude of difference if any, and measures that can be implemented in policies. Also, the information shared is on the quantitative availability of services. The qualitative estimation of services available in the setups will make it more comprehensive. Furthermore, while costs and treatment utilization were assessed, accessibility in terms of time taken to reach the healthcare facility, supports available and patients' affordability for different services have no mention. They also form the key information on the utilization of mental health systems. Critical evaluation of the recommendations of the report The mental health burden in India calls for the immediate attention of political leaders, policymakers, health professionals, opinion-makers, and society at large. The NMHS report recommends the integration of the Mental Health Program with other National health programs; however, there is a need to ensure appropriate mechanisms so that it does not remain as guidelines but translates into practice. There is a need to deliberate on the means of integrating mental health services with other programs such as maternal and child health, adolescent health, etc., as it is resource-intensive. If the survey data are handy in identifying the high-risk groups, it would be useful in prioritizing the proposed screening and health promotion initiatives. Substance use disorders are identified as the most common psychiatric disorders in India in the survey. This should facilitate the treatment options for these problems to be available at the primary care level instead of concentrating at a few centers in the district headquarters. The report emphasizes on m-Health and e-Health initiatives to strengthen mental healthcare services which is the need of the hour. The key challenge here is the poor network connectivity in rural and remote areas. This can also be addressed by separate earmarked funding rather than dependence on the NCD flexi pool budget. However, in the absence of basic infrastructure and resources, m-health might not be useful in providing quality healthcare. Besides, the lack of clarity in the utilization of available funds needs to be addressed. Also, mental disorders need to be covered without biases, under health insurance schemes. Conclusions and future directions NMHS was very meticulously planned, well-coordinated, and executed on a national level. When planning such a large-scale survey, challenges are likely. Limitations due to the MINI version 6.0, weighted analysis and noninclusion of samples from all states in generalizing results, and noninclusion of the infrastructure and manpower from the private sector are briefed. Limited information on the qualitative aspects of the services provided, issues with the funding, and governance with respect to the implementation of the outcome as policies are the other key areas that we have addressed in this critical review. All in all, designing a tool akin to MINI version 6.0 specifically for this study to address some of its limitations and cover other common as well as new-age disorders (behavioral addiction), providing an opportunity for the private sector to participate and assist in covering the enormous population of the country, enhancing funding provision to improve the infrastructure in the areas like m-health and e-health initiatives, expanding the mental health systems assessment to include the contribution of indigenous systems and services (yoga, faith healing) are some of the key future directions that we wish to suggest. We look forward to the upcoming phases of the survey and hope the abovementioned suggestions are implemented as feasible. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.