An Audit of Indian Health Insurance Claims for Mental Illness from Pooled Insurance Information Bureau's Macroindicator Data.

INTRODUCTION: Information on the social and voluntary insurance coverage of mental illness in India is scarce. We attempted to address this lacuna, utilizing a secondary macrodata approach for 3 years. Mental illness per se is not covered by most of existing Indian health insurance policies.
MATERIALS AND METHODS: Publicly available de-identified claim macrodata for all health (nonlife) insurance for Indian financial year from 2011-2012 to 2013-2014 were collected. The age group, gender, amount of claims, proportion of claims, and details of number of days of hospitalization were collected and analyzed. Descriptive statistics, Chi-square test, and Wilcoxon tests were used appropriately. P ≤ 0.05 was considered statistically significant.
RESULTS: In 2011-2012, there were 2864 claims from the registered 2,591,781 members citing mental illness (0.11%) which decreased to 0.03% in 2012-2013 and marginally rose to 0.07% of all claims. The total amount of claims paid for mental illness was Rs. 51.7 millions in 2011-2012, Rs. 97.2 million in 2012-2013, and Rs. 150 million in 2013-2014. Statistically significant difference emerged in terms of age group, gender, amount and proportion of claim, and number of days of hospitalization.
CONCLUSION: The penetration of health insurance is low and claim for mental illness remains low. The difference in patterns of age, gender, amount of claims, and number of days for mental illness provides detailed relevant information to formulate future policies.

INTRODUCTION

It is reported that in India, raising healthcare costs push people into poverty. Owing to accommodating unexpected health expenditure, about 3.5% of Indian population slide into below poverty line category and about 5% of all Indian households suffer “Catastrophic health expenditure due to unaffordable health costs.”[1] The result of the first ever large-scale, structured, epidemiological Mental Health Indian Survey indicates that 10.6% of Indian adults (≥18 years of age) suffer from mental morbidity (excluding tobacco use disorders) and lifetime prevalence of mental disease who require active intervention is 13.7% of Indian population. This translates to about 150 million Indians.[2] The same survey indicates that the continuous cost incurred with long-term care that is often required for psychiatric illnesses, place undue stress on the patient as well as the family and often the economically deprived population are relatively more vulnerable to this disorder.[2]

To provide a substantial relief from unexpected health expenditure, coverage by insurance companies (both as a social and voluntary [private] health – nonlife insurances) became more popular during and after the World Wars. It is in the aftermath period, many Western insurance companies begun catering to mental illness. In the Indian context, Medical insurance was introduced in the early 21st century and it took nearly 10 years to penetrate and reach about 3% of all population.[345] Health insurance or health cover, in Indian context, is defined by the registration of Indian insurance companies regulations 2000 “as the effecting of contracts which provide sickness benefits or medical, surgical or hospital expense benefits, whether inpatient or outpatient, on an indemnity, reimbursement, services, prepaid hospital, or other plan basis, including assured benefits and long-term care.”[6] Indian insurance companies were active since the 1970s and offered no health insurances. It is in the last decade of the 20th century, such specific policies were designed targeting specific populations and later their scopes enlarged.[34]

Most of the Indian health insurance providers explicitly avoid mental illness with a clause “any mental illness, psychosomatic dysfunction, or problems connected to psychiatric conditions.”[5] Organic psychosis and a small number of plans (group-plans) targeting specific populations are said to be the only exclusion for this widely declared mental illness shunning policy of the Indian insurance bodies. However, these exceptions largely remain anecdotal and we did not meet any published or presented literature in this regard. It has also been pointed that the draft of the Indian Mental Health Care bill, 2013/2016, which when implemented proposes widespread changes in the policy to provide coverage for the mental illness too.[5]

With such ongoing reforms in mental health climate in India, private sectors, governments, and policy makers will largely benefit from better information on the existing claim pattern of mental illness which will help them for making decisions on resource allocation, financing, and further capacity building of mental health systems as well as revamp insurance policies toward mental health coverage. Through this work, we present an insight into the reimbursement/claim pattern of Indian patients with mental disorders who were covered with health (nonlife) insurance.

MATERIALS AND METHODS

Basic data for this study were collected from publicly available, delimited health insurance (nonlife) data analysis reports of years 2011–2012, 2012–2013, and 2013–2014 published by Insurance Information Bureau (IIB) of India (www.iib.gov.in). This yearly reports are collated from mandatory submissions by the various entities that offer health insurance in India. From these reports, macroindicators of number of persons registered for health insurance; total number of claims; total claims paid (in 10 million) - including below Rs. 1000 and above Rs. 2,000,000; total number of claims above Rs. 2000000; total claims paid (in 10 million) - excluding below Rs. 1000 and above Rs. 2,000,000; total claim paid above Rs. 2,000,000 Lakhs (In 10 million); average claim paid (in rupees); number of claims for mental disorders - excluding above Rs. 2,000,000; total claims paid for mental disorders (in 10 million) and average claim paid for mental disorders were noted down and presented in Table 1. In all instances, payments/claims below Rs. 1000 and above Rs. 2,000,000 were not considered.

Table 1

The macro-data collected from the insurance information bureau

The submissions by insurers in all years are plagued by errors. For example, in report of 2013–2014, (a) some insurers have not disclosed information for about 4814.2 million, (b) large number of small claims (< Rs. 1000) accounting to 15% of number of claims paid but only 0.1% of total amount of claims paid, (c) small number of large claims (more than Rs. 2,000,000) accounting to 0.09% of number of claims paid but accounted for 17% of total amount of claims paid, (d) only 92% of claims have gender, (e) Only 37% of submitted data have valid diagnostic code. The missing data have been excluded from analysis in reports. Computations were performed using formulae provided by the report.

All such collected data were entered and analyzed using Statistical Package for Social Services (Version 20, IBM, IL, USA) Descriptive analysis including frequency analysis, Chi-square tests, and Wilcoxon sign rank test was appropriately used. Appropriate weighing of data was performed for outcome variables, when required. A P < 0.05 was considered as statistically significant.

RESULTS

From the reports, the details of Claims Filed Citing Mental Illness (CFCMI) in all 3 years were collected. For the years 2011–2012 (financial year–April 1, 2011–March 31, 2012) and 2012–2013, the numbers were estimated from the tables provided in the report. In several instances of claims (even up to 25% of all claims), the age, gender, or type of illness (ICD Classification) were missing. Furthermore, claims below INR 1000.00 or above 2000000.00 were not included in the tabulation. The estimated CFCMI for the years 2011–2012, 2012–2013, and 2013–2014 were 1958, 1138, and 3290, respectively, while observed data for further analysis were available for 2009, 938, and 2863 patients, respectively. This difference stems from the variations in reporting pattern. For this work, only available data were taken into consideration.

The age-gender with year-wise distribution of CFCMI is listed in Table 2. The difference in the distribution was statistically significant. It is observed that among males, in 2011–2012, the age group of 6–35 years was commonly involved which shifted to 26–55 years group in 2012–2013 while in 2013–2014, it was above 70 years age group that was involved in CFCMI. In females, in 2011–2012, the age groups of 6–45 years were commonly involved which remained static in 2012–2013 and 2013–2014. The Graphs 1 and 2 show the distribution in genders in terms of age group during the study period. Males vastly outnumber females in terms of CFCMI, reflecting the innate male preponderance in subscription of health insurance policies underlined in all IIB reports.

Table 2

Age group, gender wise distribution of claims in the study population (percentage in brackets)

Graph 1

The distribution of male study population age group during study period

Graph 2

The distribution of female study population age group during study period

All amount of monies were considered only in Indian Rupees (at the time of writing of this manuscript, 1 US$ was about 67 Indian Rupees). In 2011–2012, 88% of the CFCMI were filed in the band of 1000–25,000, especially in 1000–5000 accounting for 57% of all claims in the financial year. In 2012–2013, the claim in the 1000–5000 band reduced to 19.7% of all claims, still with 68.1% accounted by 1000–25,000 claim band. In 2013–2014, the 1000–5000 reduced to 12% and only 44.3% of claims falling in the band of 1000–25,000. Interestingly, the band of 25,001–50,000 drastically increased to 22.7% of all claims [Table 3 and Graph 3].

Table 3

Distribution of claims based on year and amount of claims among the study population

Graph 3

Graph with dual axis showing number of claims and the percentage of claims during the study period

During the study period, the mean number of claims fell as the claim band increased, with statistical significance as reflected by Wilcoxon test [Table 4]. The amount of claims disbursed during the study period is more described in Table 5. Largest mean amount of disbursement was seen in 500,001–1,000,000 for a 25.1 million followed by 1,000,001–1,500,000 while the lowest was in 1000–5000 categories. The difference was statistically significant (P = 0.000). The Graph 4 describes the split in terms number of claims by year. Table 6 shows the number of days of hospitalization and amount reimbursed for the same. Closer scrutiny reveals that for <1 day hospitalization for mental illness, on an average, in 2011–2012, had a reimbursement of Rs. 19,099.38 that nearly doubled to Rs. 403,030.3 in 2012–2013 and subsequently reduced to Rs. 318,021.2 in 2013–2014.

Table 4

Mean number of claims based on amount of claims among the study population

Table 5

Mean amount of claims (in 10 million) disbursed in each claim band among the study population in the entire study period

Graph 4

The number of claims during the study period on basis of number of days of hospitalization

Table 6

Number of days of hospitalization for Claims Filed Citing Mental Illness in each year and amount disbursed

DISCUSSION

In Indian health insurance industry, till date, mental illness has been shunned and often denied claims. Exceptionally, in conditions of very few group policies and conditions such as organic psychosis and reimbursement may be provided. Hence, the CFCMI is relatively miniscule of all the claims. In addition, the operational difficulties such as collecting all details from all providers about details of claims in stipulated period becomes increasingly difficult. Such submissions in proper desired format also increase the operational costs. Large amount of CFCMI data as admitted by IIB without age, gender, hospitalization details, and/or International code of diseases would exist clouding the exact number of mental illness claims. The limited data available presently in open domain renders further analytical and inferential statistics application difficult. However, in the absence of other related data sources, this would help us to get a robust idea about claim pattern. Till date, to the best of our knowledge, there is no pertinent literature that has studied the relation of mental illness and health insurance, in terms of pattern of claims and reimbursement. Hence, comparison of the results of this study with related literature is rendered nearly impossible.

The Table 1 clearly shows the differential treatment toward mental illness by the sector. We could observe in table that proportion of CFCMI in 2011–2012 was 2864 from 2,591,781 (0.11%) which decreased to 0.03% in 2012–2013 and marginally rose to 0.07%. This is because only certain restricted policies and organic psychosis were reimbursed while others were denied. The last estimate of prevalence rate of organic psychosis in Indian population was at 0.4 in 1998.[7] Probably, this low prevalence rate of organic psychosis has been reflected proportionately in the CFCMI too. Considering the total mentally ill population requiring active treatment being pegged at 150 million, awarding claims for few thousands shows the lacunae in the health insurance sector as well as serves as a grim reminder of the differential attitude of service providers. This probably also emanates from the stigma of mental illness as well as lack of understanding about illness and to the fact that the treatment is long-term and considerably attracts high costs.[5] This is seen in the distribution of average claim paid for CFCMI. In 2011–2012, it was Rs. 18,058 that rose more than five times to Rs. 103,654 and dropped by a fourth to Rs. 75,501 in 2013–2014. This huge fluctuating difference cannot be explained and thus needs more in-depth analysis.

The male preponderance is reported in accessing the health insurance schemes. Majority of the members enrolled are males. Hence, the male preponderance reflected in Table 2 and Graph 2, 3 is a reflection of enrollment process rather than the actual gender difference. Such a bias needs to be addressed. More females need to be enrolled and encouraged to subscribe to health insurance schemes. Among males, those below 35 years sought more claim while among females, it was up to 55 years. There is a minor variation with every year. This could be either an inherent difference in female enrollment or due to proportionally increased prevalence of mental illness among females, especially in fourth to sixth decade of life. This finding warrants more detailed study.

In 2011–2012, among males, 37.8% of all claimants were below 15 years of age while in 2012–2013, it was 10.8% and in 2013–2014, it was 8.9%. Among females, it was 24.9%, 15.2%, and 10.2%, respectively. The numbers indicate two possibilities: Considering gross under/overstating/errors, some significant portion of entrants would have mistakenly entered age code as 0–5 or 6–15 years alternatively the development of mental illness in this young age group. Both these entities are a cause of concern and need to be probed further.

The exceptionally large claim (57%) in claim band of Rs. 1000–5000 in 2011–2012 is a cause of concern given the high degree of errors. The reduction in subsequent years (to 19.7% then further to 12%) indicates that lower claims are being rejected or amount of claim per submission is increasing [Graph 3]. Either of this phenomenon is a cause of concern. Most of the CFCMI were filed below Rs. 25,000 till 2012–2013 while it was till Rs. 50,000 in 2013–2014. This strengthens the finding in Table 1 that the average claim per submission is increasing rapidly from Rs. 18058 in 2011–2012 to Rs. 75,501 in 2013–2014. This could be also partially due to the increasing cost of mental illness treatment.

The number of claims is highest for Rs. 1000–5000 that reduces till Rs. 100,000. In Rs. 100,001–300,000, there is an increase after which it decreases to the minimum. This could be due to the fact that most of the health insurance are in the range of Rs. 100,001–300,000. The mean amount of claim slowly increases with the claim band and is seen highest disbursed in Rs. 500,001–1,000,000. The reason behind this phenomenon also needs to be studied further.

Graph 4 and Table 6 reveal more of the claims for duration of hospitalization. In 2011–2012, large numbers of CFCMI were for <1-day treatment and in subsequent years, this is drastically reduced by manifolds. On a closer observation, the reduction in number of claims for less than a day of hospitalization is accompanied by a nearly threefold increase in amount of claim disbursed. In 2013–2014, for 283 claims, 90 millions of rupees has been disbursed and occupies a major chunk. The reason for several columns being empty in 2013–2014 for amount is due to the fact that the reports give only 2 digits of claims and in higher denomination of 10 million. Owing to this, further analysis at Rs. 100,000 levels is not possible. This discrepancy strengthens the anecdotal suggestions that once a mental illness is diagnosed/established, a single, one-time payment is provided as insurance settlement and patient is encouraged to quit or surrender the policy.

Further closer observation reveals that in 2013–2012, more CFCMI were applied for <1 day while in 2013–2014, more CFCMI were for 3 days of hospitalization. It is interesting to note that in the higher claim bands, fewer numbers of claims (<20) for >15 days, had more than a million rupees reimbursement. The reason for the shift of the number of days of hospitalization needs to be studied in-depth.

Mental illness affects about 150 million Indians. A substantial number of them would require financial coverage to continue to adhere to their treatment. In the absence of viable help from insurance sectors, most of them either seek support from government mental health-care system. This places undue stress on the already overburdened system as well as may create a feeling of inability and helplessness, loss of self-esteem that could worsen the psychiatric condition. For example, in Kerala, the state-sponsored mental health facility has been shown to lack in sufficient funds.[8] The recent series of articles in Lancet highlight the mental health facility and care delivery situation in India.[910] Taken together with the recent Indian national mental health survey findings, it can be safely assumed that a vast majority of mentally ill Indians are forced to discontinue their treatment regime for financial consideration.[2]

Over the 3-year study period [Table 1], only about 5811 (excluding exclusions listed in material and method section) people were helped with 298.9 million rupees– about Rs. 51,437 only per person. This amount is not sufficient to provide lifelong support for mental health illness. The strongest support for inclusion of mental illness under health insurance comes as an opinion in Mental Health Care bill, 2013 followed by the 2016/17 versions and the recent editorial in an Indian Journal of Mental Health and Behavior.[5] This editorial argues with reasons for inclusion of mental illness under health insurances. With rapid, forward strides in diagnosis of mental illness, the patient of psychiatry is currently explained in terms of abnormalities in the brain. At present, use of radio imaging and multiple biomarkers has heightened the accuracy of diagnosis in psychiatry, removing the ambiguity of diagnosis being subjective in nature.[11]

Mental illness and its disability in any perspective need to be viewed as:

A disorder that often begin very early in life

Most of the individuals require continuous, specialized care but only minority of them receive it

Proper treatment is delayed for many years for various reasons, including financial constraints and stigma attached.[12]

To minimize the mental disability, a bridging of the “treatment gap” must occur. The treatment gap here is the percentage of people who require treatment but not receive it. Health insurance, when extended to large number of Indian population, will help reduce this “treatment gap.” Early institution of treatment minimizes the colossal expenses in terms of finance, lost working hours, and emotional turmoil imposed by the disease. Health insurance, if implemented as per the draft Indian Mental Health Care Bill 2013 and its latest version, 2016/17, will help minimize the treatment gap in mental health segment.[12] Currently, though each health insurance service provider have their own list of exclusions, the Insurance regulatory and Development authority of India issued educational handbook[13] does not explicitly list the mental health issues, though about 10% of Indian population require immediate mental health care. India also suffers from “systematic discrimination of exclusion of mental disorders from some social and private insurance schemes for health care.”[14] Emergence of large-scale social insurance schemes can be seen as a game changer in mental health finance, but the effect of which has not been still documented widely. In India, 70% of health expenses are catered by “out-of-pocket” model. It is reported that this type of payments, when exceeds 10% of average household income, retards treatment-seeking behavior.[15] The annual market worth of neuropsychiatry drugs alone, in 2015 is worth of 56,460 million rupees, an indicator of the mental illness drug usage.[16] Hence, the average Indian with mental illness needs financial support in form of social insurance and voluntary health insurance scheme to carter for his illness. The success of such schemes in surgical specialties has been reported from India[17] but none regarding mental illness. This could be due to the fact that mental illness is being kept out of the ambit of such schemes.

With the changing lifestyle pattern and disorders associated with it in India and the mounting evidence of influence of certain lifestyle disorder drugs influencing mental health,[18] huge burden of mental illness is expected soon. In such a given situation, the stakeholders including government, insurance providers, and collective bodies of mental health professionals need to pitch into include mental illness under the ambit of social and private (voluntary) health insurance scheme. Such a move also would contribute to destigmatization of mental illness but also prevents affected citizens and their families from unexpected, long-term, catastrophic health expenditures.

CONCLUSION

The claim pattern, claim band distribution, age, gender, and number of days of hospitalization pattern of Indians claiming for health insurance citing mental health insurance have been explored. The study would give a robust estimate of contribution of the health insurance (voluntary and social) toward the reimbursement of mental health expenditure in India. Only a very handful of cases are being supported by these societal structures. Till these structures are strengthened, the burden on existing government mental healthcare infrastructure as well as “out-of-pocket” expenses will continue to rise. This will in turn place more strain on the mental healthcare-seeking behavior of Indians and may continue to add to the stigma attached to this disease. Furthermore, the present study highlights the need for improving the data collection and submitting policies by insurance agencies to present a valid wholesome data to draw meaningful conclusions.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.