Important to have centralised fraud-detection, biometric authentication and live photographs have to be compulsory
Last week, this newspaper reported a fairly significant increase in the number of health insurance claims under Rajasthan’s Bhamashah Swasthya Bima Yojana, on which the centre’s National Health Protection Scheme (NHPS) is being modelled, and speculated on whether this was due to the increasing incidence of fraudulent claims made easy by the fact that hospitals were being allowed to admit a sizeable number of patients without a biometric authentication. As a result of very high claims in the past, insurance premiums shot up from Rs 370 per family in the first phase of two years to Rs 1,263 for the second phase that began in December 2017.
While the Rajasthan government didn’t reply to the queries raised at that point, it has since given a reply. Though it has somewhat different metrics—this newspaper talked of weekly claims rising 2.7 times over 17 weeks—the Rajasthan government also talks of daily claims going up from Rs 1.56 crore per day in December 2017 to Rs 2.74 crore so far in April. The reason for the bump in claims, the government argues, is because while there were 170 private hospitals enrolled when the scheme began in December 2017, this rose to 707 in March 2018, and not because of fraud; while New India Assurance—the insurer—looks at the claims ratio, the government’s success metric is the number of people who use the scheme. While the Rajasthan government is of the view the scheme will stabilise at Rs 2.5-3 crore claims per day, and the insurer will eventually return Rs 100 crore to it from the savings at the end of the year—the contract allows for extra premium to be returned—there is no evidence of stabilisation in claims so far (see graphic).
Though any view on what the levels the scheme will settle at will have to wait till the actual data available at the end of the year, certain clear lessons emerge for NHPS. For one, given the disputes between the insurer and the Rajasthan government on the issue of de-empanelment of hospitals for fraud—the New India Assurance presentation to NITI Aayog suggests it feels the process is too slow and encourages fraud—it is best the fraud management be centrally monitored and not left to 29 states and seven Union territories. Apart from having disputes with each state in case this is not done, a centralised scheme will also mean the parameters for de-empanelment, etc, are the same across the country.
A decision also has to be made at the earliest on what identity parameters are to be used. In the case of Bhamashah, at least two members of the family’s biometrics were on the card; in which case, biometrics for other family members would not always be available for, say, for authenticating identity at a hospital. According to the Rajasthan government, right now, 75% of those availing the scheme are being identified using biometrics. In the rest of the cases, the Medical Officer In-Charge (MOIC) is allowed to bypass the need for Aadhaar authentication—a large proportion of the cases of fraud detected by the insurance company in Bhamashah were those where MOIC clearances had been given. The Rajasthan government believes that Aadhaar authentication can go up to around 80% but not above that since there could be people whose prints have faded away or there could be issues of internet connectivity between the hospital and the Aadhaar repository.
Given the role of Aadhaar-based biometrics in weeding out—but not eliminating—fraud, it is imperative that NHPS not accept such a low authentication ratio. It is interesting to note that while hospitals or ration shops show high fault rates in getting Aadhaar authentication, mobile phone operators seem to have a higher success rate. This suggests that where there is a greater incentive to authenticate customers, as in the case of telcos, the agency seems to be getting it right. While insurance companies will still have to conduct their checks to detect fraud—there is nothing to prevent a hospital from paying someone and getting a live biometric but not actually conducting an operation—biometric checks and live pictures are a big deterrent.
Getting the pricing right is equally important. In the case of Bhamashah, when the price was very low, this led to huge losses—from 90% in the first year, the claims ratio rose to 176% in 2016-17—and, as a result, the premium shot up 3.4 times in the second phase. While insurance companies or reinsurers may be prepared to take on such losses for one state, doing it at an all-India level is quite different. The Pradhan Mantri Jeevan Jyoti Yojana, for instance, is running at losses of 104%, and the Pradhan Mantri Suraksha Bima Yojana at 205%, making it clear that pricing is very important. For NHPS to be successful, fixing these loopholes is critical.