An evaluation of the impact of expanding Rashtriya Swasthya Bima Yojana: a field experiment in Karnataka. Anup Malani 3ie, New Delhi June 14, 2013

Transcription

1 An evaluation of the impact of expanding Rashtriya Swasthya Bima Yojana: a field experiment in Karnataka Anup Malani 3ie, New Delhi June 14, 2013

2 The Challenge India s economic growth has not improved the health of all Indians E.g., Infant mortality rate is 47/1,000 births, maternal mortality rate is 200/100,000 births (WB 2011) Financing is an important part of the problem 25% of untreated ailments are due to financial constraints (NSS Report 2004) > 75% of India s health expenditures are out of pocket (Berman et al. 2010) Medical expenses push 63.2 million Indians into poverty every year (Berman et al. 2010)

3 Shift in strategies Traditionally, India relied on supply side solutions (government hospitals, training) As demand outstripped public supply, India turned to demand side subsidies State schemes (Yeshasvini, Vajpayee Arogayshri) Janani Suraksha Yojana (2005) Rashtriya Swasthya Bima Yojana (2008)

4 Possible Solution: RSBY Eligibility: BPL (central floor) Available in > 2/3 of all districts Enrollment of 30m hhds by 2012! Coverage: Rs. 30,000 per hhd for treatment at empanelled, secondary hospitals Rs. 30 annual registration fee only No deductible, co-pay Cashless through biometric Smart Card Funding: most states 75% central, 25% state Above floor, states pay 100% Prices determined by government charge-list

5 Policy Questions Impact of existing RSBY? RSBY suffers low uptake among eligible Impact of expanding RSBY eligibility? Include APL (perhaps progressive premium subsidies) Impact of expanding RSBY coverage? Physician, diagnostics, medicines, tertiary care

6 Indian Health Insurance Experiment I Objectives What is the impact of expanding RSBY coverage to APL? May also influence uptake among BPL and empanelment of more hospitals How does RSBY impact the wellbeing of Indian households? Randomize 9300 hhds in Karnataka to receive insurance, cash transfer, or neither and observe them for two years Additional 330 hhds asked about willingness to pay for RSBY Arguably the second largest HI experiment ever: ~9000 hhds x 2 years = 18,000 hhd years

7 Collaborative Project Primary investigators: Anup Malani (U. Chicago) and Ramanan Laxminarayan (Public Health Foundation of India) Implementation with and support of RSBY-New Delhi: Anil Swarup, (Labour Ministry), Nishant Jain (GIZ), Henna Dhawan (GIZ) Collaboration with RSBY-KA and insurance companies Sri Harsh Gupta (Labor Commissioner), Narasimha Murthy (CEO) and Shantveer Patil United India, Tata-AIG and Medi-Assist Data collection by Centre for Microfinance: Shahid Vaziralli, Sneha Stephen and Tanay Balantrapu Design and analysis input from US-based researchers: Anuj Shah, Alessandra Voena, Gabriella Conti (U. Chicago), Cynthia Kinnan (Northwestern) Cornertstone funding from DFID, U. Chicago

8 Design of the Field Experiment Enroll APL hhds not currently eligible for RSBY in Karnataka Gulbarga District in North, Mysore District in South 25 and 39 empanelled hospitals in Gulbarga and Mysore, resp. Three arms in study Free RSBY insurance (no R. 30 registration fee) Unconditional cash transfer (R. 500) + RSBY option Nothing RSBY option Randomized allocation Observe outcomes for two years Annual household surveys On-going post-health event surveys Powered to estimate 10% change in hospitalizations + 15% attrition

9 Why Three Arms? RSBY has two components (insurance + premium subsidies) Separate policies with different impacts Offering insurance without subsidies yields lower take-up, so we offer both together To net out effect of subsidy, we must offer the value of a premium subsidy to a control group Value of premium is willingness to pay (WTP) for RSBY We offer two approximations (Rs. 500 and Rs. 0) We elicit WTP and scale impacts using two control groups We can assess impact of (1) unsubsidized RSBY, (2) subsidized RSBY, (3) subsidies alone and (4) a (budget-neutral) cash transfer

10 Outcomes: Health Care & Finances Without HI, hhds pay OOP or forego care OOP means loans, saving asset sales (Kruk, Goldmann, & Galea, 2009) If these are too costly, forego care HI reduces the cost of financing health care through risk pooling (Van de ven & Ellis 2000) Increases utilization or reduces cost of care Indirectly improves income, consumption, assets We measure Income, assets, consumption Health care utilization Key personnel: A. Malani, R. Laxminarayan, C. Kinnan

11 Outcome: Health Behavior Insurance changes behavior Increase risk taking (Cutler & Zeckhauser, 2000) Reduce fatalistic behavior We measure: Self-reported health and health behavior Anthropometric outcomes (from 3 members of the hhd) We hope to add biomarkers 6 months after baseline Key personnel: A. Malani, G. Conti

12 Outcome: Cognitive Capacity Economic shocks compromise cognitive capacity, leading to poorer decision-making (Shah et al. 2012) Possible mechanism behind poverty traps Health shocks may have similar impacts Insurance might buffer these effects We measure cognitive capacity of hhd head with and without shocks, across arms Key personnel: A. Shah

13 Outcome: Female/Child Welfare Men often control finances, thus cash transfers (Braido et al. 2012) Women often are better aware of health needs Might the allocation of benefits from health insurance favor women/children more than cash? We measure who utilizes health care, other resources in households Key personnel: A. Voena

14 Timeline Jun-13: Eligible household census/listing Jul/Aug-13: Baseline, allocation for 9300 hhds WTP measurements on 180 hhds Sept-13/June-15: On-going post-health event surveys Jan/Feb-14: Short follow-up, biomarkers for 3000 hhds Jul/Aug-14: Midline, WTP on 150 hhds Jul/Aug-15: Endline, WTP for 9000 hhds

15 Sister Projects Indian HIE II: Impact of Expanding Coverage Impact of private plan with physician, diagnostic coverage (Jul-14) India HIE III: Health Insurance v. Access to Credit Impact of RSBY, 0% health credit line, 24% health credit line, no treatment (Jul-14 or Jul-15)

16 References Braido, Luis HB, Pedro Olinto, and Helena Perrone. "Gender Bias in Intrahousehold Allocation: Evidence from an Unintentional Experiment." Review of Economics and Statistics 94.2 (2012): Berman, P., Ahuja, R. & Bhandari, L., The impoverishing effect of healthcare payments in India: new methodology and findings. Economic & Political Weekly, 45(16), pp Cutler, D.M. & Zeckhauser, R.J., The anatomy of health insurance. Handbook of Health Economics, 1, pp Kruk, M.E., Goldmann, E. & Galea, S., Borrowing and selling to pay for health care in low-and middle-income countries. Health Affairs, 28(4), p.1056 NSS Report 2004 Shah, A.K., S. Mullainathan, and E. Shafir, Some Consequences of Having Too Little. Science, (6107): p Van de ven, W.P.M.M. & Ellis, R.P., Risk adjustment in competitive health plan markets. Handbook of Health Economics, 1, pp Available at: [Accessed April 24, 2013]. World Bank, Data: India. Available at: