The effect of health insurance on hospitalization: Identification of adverse selection, moral hazard and the vulnerable population in the Indian healthcare market (original) (raw)

Abstract

The Indian healthcare sector is growing at a rapid pace; nevertheless, inequality in healthcare consumption and catastrophic healthcare expenditure is also increasing at an alarming rate. In addition to socioeconomic differences, poor healthcare infrastructure, and inadequate risk-pooling mechanisms; asymmetric information in the healthcare market is also a potential contributor to this inequity and increasing costs. The consequences of information asymmetry are adverse selection (AS) and moral hazard (MH). AS occurs if people with health risks (high-risk individuals) are more prone to buying health insurance as compared to low-risk individuals. MH occurs when insured individuals are more likely to use healthcare than the uninsured individuals, inflating insurance premiums and medical care costs. Empirically, AS and MH lead to endogeneity due to unobserved heterogeneity. In practice, endogeneity is often addressed by using the instrumental variable estimation technique; however, this approach suffers from identification problems. Therefore, in this paper, we use an instrument-free semi-parametric copula regression technique to examine how health insurance status affects hospitalization using a sample of individuals from a large nationally representative survey for India. Our results suggest the presence of AS and potential MH in the Indian healthcare market. We observe that chronically ill individuals are probable sources of AS, which leads to possible MH. A spline regression analysis suggests nonlinearity in health insurance choice and healthcare utilization across age, education, family size, and household consumption expenditure. We find chronically ill women in India exhibit less insurance coverage and lower hospital care usage. We also identify the vulnerable groups, such as older adults and rural residents, who have low insurance participation and high healthcare consumption. Our results indicate toward the need for evidence-based health care policy to manage the healthcare system and support the disadvantaged population of India.

Figures (17)

Source: Authors’ own definitions & calculations; Observations (n): 191,044. * Remaining constitutes other health insurance schemes. Variable definition and descriptive statistics.

Source: Authors’ own definitions & calculations; Observations (n): 191,044. * Remaining constitutes other health insurance schemes. Variable definition and descriptive statistics.

Fig. 1. Status of health insurance by chronic conditions. The left panel of the figure shows that only 14 percent individuals without any chronic health problem were insured. Whereas, the right panel of the figure shows that almost 27 percent individuals with some chronic condition were insured. The figure suggests that high-risk individuals (chronically ill) are more likely to be insured than low-risk individuals (not chronically ill). Source: Authors’ calculations.

Fig. 1. Status of health insurance by chronic conditions. The left panel of the figure shows that only 14 percent individuals without any chronic health problem were insured. Whereas, the right panel of the figure shows that almost 27 percent individuals with some chronic condition were insured. The figure suggests that high-risk individuals (chronically ill) are more likely to be insured than low-risk individuals (not chronically ill). Source: Authors’ calculations.

Fig. 2. Status of hospitalization by health insurance. The left panel of the figure shows that only 14 percent individuals without any health insurance were hospitalized Whereas, the right panel of the figure shows that almost 18 percent individuals with some health insurance were hospitalized. The figure suggests that insured individuals ar more likely to be hospitalized than uninsured individuals. Source: Authors’ calculations.

Fig. 2. Status of hospitalization by health insurance. The left panel of the figure shows that only 14 percent individuals without any health insurance were hospitalized Whereas, the right panel of the figure shows that almost 18 percent individuals with some health insurance were hospitalized. The figure suggests that insured individuals ar more likely to be hospitalized than uninsured individuals. Source: Authors’ calculations.

The estimated SATE (in %) and confidence interval (CI) for all fitted copula models. Source: Authors’ own calculations. Table 2

The estimated SATE (in %) and confidence interval (CI) for all fitted copula models. Source: Authors’ own calculations. Table 2

Source: Authors’ own calculations. "p< 0.001; “p< 0.01; “p<0.05; ‘p<0.10 Empirical results from both treatment and outcome equations.

Source: Authors’ own calculations. "p< 0.001; “p< 0.01; “p<0.05; ‘p<0.10 Empirical results from both treatment and outcome equations.

Fig. 3. Non-linear estimation results from the aggregate model (HINS). This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on health insurance (HINS). Smooth function estimates and associated 95% point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are <0.000, <0.000, <0.000 and <0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations. Having identified the AS and MH from our joint analysis, we now focus on the sources of variations in the enrollment and We observe scant evidence of gender bias in health insurance status or hospitalization against women in general. However, to further examine this problem and identify the possible cohort of women for whom gender bias in health insurance and healthcare use may exist in India, we add two interaction terms: (FEMALE X - CHRONIC) and (FEMALE X MARRIED). The likelihood of health insurance coverage (all schemes) and hospitalization significantly decreases for women with any chronic health problem. This find- ing is interesting, as chronic health in general increases the proba- bility of getting insurance but females with chronic health problems are less likely to be insured. This finding indicates a degree of underlying bias against women with chronic health

Fig. 3. Non-linear estimation results from the aggregate model (HINS). This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on health insurance (HINS). Smooth function estimates and associated 95% point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are <0.000, <0.000, <0.000 and <0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations. Having identified the AS and MH from our joint analysis, we now focus on the sources of variations in the enrollment and We observe scant evidence of gender bias in health insurance status or hospitalization against women in general. However, to further examine this problem and identify the possible cohort of women for whom gender bias in health insurance and healthcare use may exist in India, we add two interaction terms: (FEMALE X - CHRONIC) and (FEMALE X MARRIED). The likelihood of health insurance coverage (all schemes) and hospitalization significantly decreases for women with any chronic health problem. This find- ing is interesting, as chronic health in general increases the proba- bility of getting insurance but females with chronic health problems are less likely to be insured. This finding indicates a degree of underlying bias against women with chronic health

Figs. 3-8 plot the smooth function estimates for the treatment and outcome equations (and associated Bayesian intervals), for the aggregate model, the private- and the public-health insurance Fig. 4. Non-linear estimation results from the aggregate model (HOSP). This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on hospitalization (HOSP). Smooth function estimates and associated 95% point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, shows the covariate values. The p- values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are <0.000, <0.000, <0.000 and <0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations. Variables on household type show that households belonging to wage-earning and casual labor categories are more likely to be insured by all kinds of insurance schemes than any other group of households. Our results are consistent with Ghosh and Gupta (2017) for public health insurance and Dutta and Husain (2013) for private health insurance. Overall, this result may be due to the availability of employer-sponsored health schemes (both pri- vate and public) for wage-earning individuals, publicly funded

Figs. 3-8 plot the smooth function estimates for the treatment and outcome equations (and associated Bayesian intervals), for the aggregate model, the private- and the public-health insurance Fig. 4. Non-linear estimation results from the aggregate model (HOSP). This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on hospitalization (HOSP). Smooth function estimates and associated 95% point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, shows the covariate values. The p- values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are <0.000, <0.000, <0.000 and <0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations. Variables on household type show that households belonging to wage-earning and casual labor categories are more likely to be insured by all kinds of insurance schemes than any other group of households. Our results are consistent with Ghosh and Gupta (2017) for public health insurance and Dutta and Husain (2013) for private health insurance. Overall, this result may be due to the availability of employer-sponsored health schemes (both pri- vate and public) for wage-earning individuals, publicly funded

Fig. 5. Non-linear estimation results from the private health insurance model (PVTHINS). This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on private health insurance (PVTHINS). Smooth function estimates and associated 95% point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are <0.000, <0.000, <0.000 and <0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations.

Fig. 5. Non-linear estimation results from the private health insurance model (PVTHINS). This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on private health insurance (PVTHINS). Smooth function estimates and associated 95% point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are <0.000, <0.000, <0.000 and <0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations.

Fig. 6. Non-linear estimation results from the private health insurance model (HOSP). This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on hospitalization (HOSP). Smooth function estimates and associated 95% point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are < 0.000, <0.000, <0.000 and < 0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations.

Fig. 6. Non-linear estimation results from the private health insurance model (HOSP). This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on hospitalization (HOSP). Smooth function estimates and associated 95% point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are < 0.000, <0.000, <0.000 and < 0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations.

Fig. 7. Non-linear estimation results from the public health insurance model (PUBHINS). This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on public health insurance (PUBHINS). Smooth function estimates and associated 959 point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, show: the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are <0.000, <0.000, <0.000 and <0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations.

Fig. 7. Non-linear estimation results from the public health insurance model (PUBHINS). This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on public health insurance (PUBHINS). Smooth function estimates and associated 959 point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, show: the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are <0.000, <0.000, <0.000 and <0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations.

Fig. 8. Non-linear estimation results from the public health insurance model (HOSP): This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on hospitalization (HOSP). Smooth function estimates and associated 95% point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are <0.000, <0.000, <0.000 and <0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations.

Fig. 8. Non-linear estimation results from the public health insurance model (HOSP): This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on hospitalization (HOSP). Smooth function estimates and associated 95% point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are <0.000, <0.000, <0.000 and <0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance. Source: Authors’ calculations.

Fig. A1. The non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on health insurance (HINS) with exclusion restriction using the aggregate model. Smooth function estimates and associated 95% point-wise confidence intervals in the health insurance equation is obtained by applying Joe90 copula regression spline model. The rug-plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are all statistically significant at less than one percent (<0.000) level of significance. Source: Authors’ calculations.

Fig. A1. The non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on health insurance (HINS) with exclusion restriction using the aggregate model. Smooth function estimates and associated 95% point-wise confidence intervals in the health insurance equation is obtained by applying Joe90 copula regression spline model. The rug-plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are all statistically significant at less than one percent (<0.000) level of significance. Source: Authors’ calculations.

Descriptive statistics and two-sample mean-test of factors for insured and uninsured individuals. Table Al

Descriptive statistics and two-sample mean-test of factors for insured and uninsured individuals. Table Al

Fig. A2. The non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on hospitalization (HOSP) with exclusion restriction using the aggregate model. Smooth function estimates and associated 95% point-wise confidence intervals in the health insurance equation is obtained by applying Joe90 copula regression spline model. The rug-plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are all statistically significant at less than one percent (<0.000) level of significance. Source: Graph by Authors.

Fig. A2. The non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on hospitalization (HOSP) with exclusion restriction using the aggregate model. Smooth function estimates and associated 95% point-wise confidence intervals in the health insurance equation is obtained by applying Joe90 copula regression spline model. The rug-plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are all statistically significant at less than one percent (<0.000) level of significance. Source: Graph by Authors.

Source: Authors’ own calculations. “py < 0.001; “p< 0.01; “p < 0.05; ‘p< 0.10; (0): Hospitalized Individuals; (1): Non- Hospitalized Individuals. Descriptive statistics and two-sample mean-test of different variables for hospitalized and non-hospitalized individuals. Table A3 Table A2

Source: Authors’ own calculations. “py < 0.001; “p< 0.01; “p < 0.05; ‘p< 0.10; (0): Hospitalized Individuals; (1): Non- Hospitalized Individuals. Descriptive statistics and two-sample mean-test of different variables for hospitalized and non-hospitalized individuals. Table A3 Table A2

Source: Authors’ own calculations. p< 0.001; “"p<0.01; “p<0.05; “p< 0.10; (0): uninsured; (1): Insured. Table A1 (continued)

Source: Authors’ own calculations. p< 0.001; “"p<0.01; “p<0.05; “p< 0.10; (0): uninsured; (1): Insured. Table A1 (continued)

Source: Authors’ own calculations. “py < 0.001; “p< 0.01; “p< 0.05; ‘p<0.10. Semi-parametric estimation results from both treatment and outcome equations with exclusion restriction for the aggregate model.

Source: Authors’ own calculations. “py < 0.001; “p< 0.01; “p< 0.05; ‘p<0.10. Semi-parametric estimation results from both treatment and outcome equations with exclusion restriction for the aggregate model.

Loading...

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

References (75)

  1. Ahlin, T., Nichter, M., & Pillai, G. (2016). Health insurance in India: What do we know and why is ethnographic research needed. Anthropology & Medicine, 23(1), 102-124. https://doi.org/10.1080/13648470.2015.1135787.
  2. Akerlof, G. (1970). The market for ''Lemons": Quality uncertainty and the market mechanism. The Quarterly Journal of Economics, 84(3), 488-500.
  3. Arrow, K. (1963). Uncertainty and the welfare economics of medical care Retrieved from. The American Economic Review, 53(5), 941-973 http://www.jstor.org/ stable/i304624.
  4. Atella, V., Holly, A., & Mistretta, A. (2016). Disentangling adverse selection, moral hazard and supply-induced demand: An empirical analysis of the demand for health care services. CEIS Research Paper 389, Tor Vergata University, CEIS, 14(10). https://doi.org/10.2139/ssrn.2801679.
  5. Azam, M. (2018). Does social health insurance reduce financial burden? Panel Data Evidence from India. World Development, 102, 1-17. https://doi.org/10.1016/ j.worlddev.2017.09.007.
  6. Bajari, P., Hong, H., & Khwaja, A. (2011). A semi-parametric analysis of adverse selection and moral hazard in health insurance contracts. Working Paper. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.576.7750&rep= rep1&type=pdf Re
  7. Bajari, P., Dalton, C., Hong, H., & Khwaja, A. (2014). Moral hazard, adverse selection, and health expenditures: A semiparametric analysis. The RAND Journal of Economics, 45(4), 747-763. https://doi.org/10.1111/1756-2171.12069.
  8. Balooni, K., Gangopadhyay, K., Turakhia, S., & Karthik, R. G. (2012). Challenges in the sustainability of a targeted health care initiative in India. IIM Kozhikode Society & Management Review, 1(1), 21-32. https://doi.org/10.1177/ 227797521200100104.
  9. Barik, D., & Thorat, A. (2015). Issues of unequal access to public health in India. Frontiers in Public Health., 3, 245. https://doi.org/10.3389/fpubh.2015.00245.
  10. Barros, P. P., Machado, M. P., & Sanz-de-Galdeano, A. (2008). Moral hazard and the demand for health services: A matching estimator approach. Journal of Health Economics, 27(4), 1006-1025. https://doi.org/10.1016/j.jhealeco.2008.02.007.
  11. Bennett, S., Ozawa, S., & Rao, K. (2010). Which path to universal health coverage? Perspectives on the World Health Report 2010. PLoS Medicine, 7(11). https://doi. org/10.1371/journal.pmed.1001001 e1001001.
  12. Bhatia, J. C., & Cleland, J. (2001). Health-care seeking and expenditure by young Indian mothers in the public and private sectors. Health Policy and Planning., 16 (1), 55-61. https://doi.org/10.1093/heapol/16.1.55.
  13. Buchmueller, T. C., Grumbach, K., Kronick, R., & Kahn, J. G. (2005). Book review: The effect of health insurance on medical care utilization and implications for insurance expansion: A review of the literature. Medical Care Research and Review, 62(1). 3-3.
  14. Cameron, A. C., Trivedi, P. K., Milne, F., & Piggott, J. (1988). A microeconometric model of the demand for health care and health insurance in Australia. The Review of Economic Studies, 55(1), 85-106. https://doi.org/10.2307/2297531.
  15. Cardon, J., & Hendel, I. (2001). Asymmetric information in health insurance: Evidence from the National Medical Expenditure Survey. RAND Journal of Economics, 32(3), 408-427. https://doi.org/10.2307/2696362.
  16. Chomi, E. N., Mujinja, P. G., Enemark, U., Hansen, K., & Kiwara, A. D. (2014). Health care seeking behaviour and utilisation in a multiple health insurance system: Does insurance affiliation matter? International Journal for Equity in Health, 13 (1), 25. https://doi.org/10.1186/1475-9276-13-25.
  17. Coulson, N. E., Terza, J. V., Neslusan, C. A., & Stuart, B. C. (1995). Estimating the moral-hazard effect of supplemental medical insurance in the demand for prescription drugs by the elderly Retrieved from. The American Economic Review, 85(2), 122-126 https://www.jstor.org/stable/2117904.
  18. Cutler, D. M., & Zeckhauser, R. J. (2000). The anatomy of health insurance. Handbook of health economics. Amsterdam: Elsevier.
  19. Devadasan, N., Ghosh, S., Nandraj, S., & Sundararaman, T. (2014). Monitoring and evaluating progress towards universal health coverage in India. PLoS MEDICINE, 11(9). https://doi.org/10.1371/journal.pmed.1001697 e1001697.
  20. Dong, Y. (2013). How health insurance affects health care demand-A structural analysis of behavioral moral hazard and adverse selection. Economic Inquiry, 51 (2), 1324-1344. https://doi.org/10.1111/j.1465-7295.2012.00457.x.
  21. Dutta, M., & Husain, Z. (2013). Does health insurance ensure equitable health outcomes? An analysis of hospital services usage in urban India. World Health & Population, 14(4), 38-49. https://doi.org/10.12927/whp.2013.23581.
  22. Economou, A., Nikolaou, A., & Theodossiou, I. (2008). Socioeconomic status and health-care utilization: A study of the effects of low income, unemployment and hours of work on the demand for health care in the European Union. Health Services Management, 21(1), 40-59. https://doi.org/10.1258/hsmr.2007.007013.
  23. Einav, L., Finkelstein, A., Ryan, S. P., Schrimpf, P., & Cullen, M. R. (2013). Selection on moral hazard in health insurance. American Economic Review, 178-219. https:// doi.org/10.1257/aer.103.1.178.
  24. Fang, H., Keane, M. P., & Silverman, D. (2008). Sources of advantageous selection: Evidence from the Medigap insurance market. Journal of Political Economy, 116 (2), 303-350. https://doi.org/10.1086/587623.
  25. Ghosh, S. (2014). Towards equitable coverage and more inclusive social protection in health: Health Inc-financing health care for inclusion. Studies in Health Services Organisation and Policy (SHSOP) (pp. 56-73). Antwerp, Belgium: ITG Press.
  26. Ghosh, S., & Gupta, N. D. (2017). Targeting and effects of Rashtriya Swasthya Bima Yojana on access to care and financial protection. Economic and Political Weekly, 52(4), 61-70. https://www.researchgate.net/profile/Soumitra\_Ghosh6/publicat.
  27. Giedion, U., Andrés Alfonso, E., & Díaz, Y. (2013). The impact of universal coverage schemes in the developing world: A review of the existing evidence No. 25. UNICO Studies Series. Washington DC: World Bank.
  28. Grossman, M. (1972). On the concept of health capital and the demand for health. Journal of Political Economy, 80(2), 223-255.
  29. Grossman, M. (1975). The correlation between health and schooling. In N. E. Tertecky (Ed.), Household Production and Consumption (pp. 147-211). New York: Columbia University Press.
  30. Guo, Z. Y. (2017). Health insurance and the demand for medical care: A case study from China. Asian Journal of Economics and Empirical Research, 4(1), 8-13. https://doi.org/10.20448/journal.501.2017.41.8.13.
  31. Haddad, G. K., & Anbaji, M. Z. (2010). Analysis of adverse selection and moral hazard in health insurance market of Iran. The Geneva Papers on Risk and Insurance - Issues and Practice, 35(4), 581-599. https://doi.org/10.1057/gpp.2010.20.
  32. Han, S., & Vytlacil, E. J. (2013). Identification in a generalization of bivariate probit models with dummy endogenous regressors. Journal of Econometrics., 199(1), 63-73. https://doi.org/10.1016/j.jeconom.2017.04.001.
  33. Harmon, C., & Nolan, B. (2001). Health insurance and health services utilization in Ireland. Health Economics, 10(2), 135-145. https://doi.org/10.1002/hec.565.
  34. Hopkins, S., & Kidd, M. (1996). The determinants of the demand for private health insurance under Medicare. Applied Economics, 28(12), 1623-1632. https://doi. org/10.1080/000368496327598.
  35. Hopman, P., Heins, M. J., Korevaar, J. C., Rijken, M., & Schellevis, F. G. (2016). Health care utilization of patients with multiple chronic diseases in the Netherlands: Differences and underlying factors. European Journal of Internal Medicine, 35, 44-50. https://doi.org/10.1016/j.ejim.2016.08.025.
  36. Hsiao, W. C. (1995). Abnormal economics in the health sector. Health Policy, 32(1- 3), 125-139. https://doi.org/10.1016/0168-8510(95)00731-7.
  37. Keane, M., & Stavrunova, O. (2016). Adverse selection, moral hazard and the demand for Medigap insurance. Journal of Econometrics, 190(1), 62-78. https:// doi.org/10.1016/j.jeconom.2015.08.002.
  38. Madhukumar, S., Sudeepa, D., & Gaikwad, V. (2012). Awareness and perception regarding health insurance in Bangalore rural population. International Journal of Medicine and Public Health, 2(2). https://doi.org/10.5530/ijmedph.2.2.5.
  39. Mahapatro, S. R., Singh, P., & Singh, Y. (2018). How effective health insurance schemes are in tackling economic burden of healthcare in India. Clinical Epidemiology and Global Health, 6(2), 75-82.
  40. Manning, W. G., Newhouse, J. P., Duan, N., Keeler, E. B., & Leibowitz, A. (1987). Health insurance and the demand for medical care: Evidence from a randomized experiment Retrieved from. The American Economic Review, 77(3), 251-277 https://www.jstor.org/stable/1804094.
  41. Marquis, M. S., & Phelps, C. E. (1987). Price elasticity and adverse selection in the demand for supplementary health insurance. Economic Inquiry, 25(2), 299-313.
  42. Marra, G., & Radice, R. (2011). Estimation of a semi-parametric recursive bivariate probit model in the presence of endogeneity. Canadian Journal of Statistics., 39 (2), 259-279. https://doi.org/10.1002/cjs.10100.
  43. Marra, G., & Radice, R. (2013). Estimation of a regression spline sample selection model. Computational Statistics & Data Analysis, 61, 158-173. https://doi.org/ 10.1016/j.csda.2012.12.010.
  44. Marra, G., & Radice, R. (2017). A joint regression modeling framework for analyzing bivariate binary data in R. Dependence Modeling, 5(1), 268-294. https://doi.org/ 10.1515/demo-2017-0016.
  45. Mavalankar, D. V., & Bhat, R. (2010). Health insurance in India: Opportunities, challenges and concerns. In D. C. Srivastava & S. Srivastava (Eds.), Indian insurance industry: Transition and prospects. New Delhi: New Century Publications.
  46. Morgan, M. (1980). Marital status, health, illness and service use. Social Science & Medicine. Part A: Medical Psychology & Medical Sociology, 14(6), 633-643. https:// doi.org/10.1016/S0271-7123(80)80073-3.
  47. Nandi, A., Ashok, A., & Laxminarayan, R. (2013). The socioeconomic and institutional determinants of participation in India's health insurance scheme for the poor. PloS One, 8(6) e66296.
  48. Nandi, S., Schneider, H., & Dixit, P. (2017). Hospital utilization and out of pocket expenditure in public and private sectors under the universal government health insurance scheme in Chhattisgarh State, India: Lessons for universal health coverage. PloS One, 12(11) e0187904.
  49. National Sample Survey (2015). Key indicators of social consumption in India health, NSS 71st Round, January-June 2014. Ministry of Statistics and Programme Implementation, National Sample Survey Office.
  50. Nelsen, R. B. (2006). An introduction to copulas (2nd ed.). New York: Springer Science Business Media.
  51. Nyman, J. (1999). The value of health insurance -The access motive. Journal of Health Economics, 18(2), 141-152. https://doi.org/10.1016/S0167-6296(98) 00049-6.
  52. Nyman, J. A. (2004). Is 'moral hazard' inefficient? The policy implications of a new theory. Health Affairs, 23(5), 194-199. https://doi.org/10.1377/hlthaff.23.5.194.
  53. Pauly, M. V. (1968). The economics of moral hazard: Comment. The American Economic Review, 58(3), 531-537.
  54. Powell, D., & Goldman, D. (2016). Disentangling moral hazard and adverse selection in private health insurance. NBER Working Paper No. 21858. doi: 10.3386/ w21858.
  55. Prinja, S., Kaur, M., & Kumar, R. (2012). Universal health insurance in India: Ensuring equity, efficiency, and quality. Indian Journal of Community Medicine, 37(3), 142-149. https://doi.org/10.4103/0970-0218.99907.
  56. Radice, R., Marra, G., & Wojtys ´, M. (2016). Copula regression spline models for binary outcomes. Statistics and Computing, 26(5), 981-995. https://doi.org/ 10.1007%2Fs11222-015-9581-6.
  57. Rajasekhar, D., Berg, E., Ghatak, M., Manjula, R., & Roy, S. (2011). Implementing health insurance: The rollout of Rashtriya Swasthya Bima Yojana in Karnataka. Economic and Political Weekly, 46(20), 56-63.
  58. Reddy, K. S., Patel, V., Jha, P., Paul, V. K., Kumar, A. S., & Dandona, L. (2011). Towards achievement of universal health care in India by 2020: A call to action. The Lancet, 377(9767), 760-768. https://doi.org/10.1016/S0140-6736(10)61960-5.
  59. Rothschild, M., & Stiglitz, J. (1976). Equilibrium in competitive insurance markets: An essay on the economics of imperfect information. The Quarterly Journal of Economics., 90(4), 629-649. https://doi.org/10.1186/s12889-015-1603-y.
  60. Saeed, B. I., Xicang, Z., Yawson, A. E., Nguah, S. B., & Nsowah-Nuamah, N. N. (2015). Impact of socioeconomic status and medical conditions on health and health care utilization among aging Ghanaians. BMC Public Health, 15(1), 276. https:// doi.org/10.1186/s12889-015-1603-y.
  61. Selvaraj, S., & Karan, A. K. (2012). Why publicly-financed health insurance schemes are ineffective in providing financial risk protection. Economic and Political Weekly, 47(11), 60-68.
  62. Selvaraj, S., Farooqui, H., & Karan, A. (2018). Quantifying the financial burden of households' out-of-pocket payments on medicines in India: A repeated cross- sectional analysis of National Sample Survey data, 1994-2014 Retrieved from. BMJ open, 8(5) https://bmjopen.bmj.com/content/8/5/e018020.
  63. Shamika, R., & Bergkvist, S. (2015). Are Publicly Financed Insurance Schemes Working in India? In A. P. Shekhar Shah, (Ed.), India Policy Forum 2014-15.
  64. NCAER.
  65. Shane, D., & Trivedi, P. (2012). What drives differences in health care demand? The role of health insurance and selection bias. Health, Econometrics and Data Group (HEDG) Working Papers. Department of Economics, University of York, 12(9). Retrieved from https://www.york.ac.uk/media/economics/documents/ herc/wp/12_09.pdf.
  66. Sinha, R. K. (2013). A critical assessment of Indian national health insurance scheme-Rashtriya Swasthya Bima Yojna (RSBY). European Academic Research, 1, 2299-2325.
  67. Spence, M., & Zeckhauser, R. (1971). Insurance, information, and individual action Retrieved from. American Economic Review, 61(2), 380-387 https://econpapers. repec.org/article/aeaaecrev/v_3a61_3ay_3a1971_3ai_3a2_3ap_3a380-87.htm.
  68. Srivastava, P., & Zhao, X. (2008). Impact of private health insurance on the choice of public versus private hospital services. Health, Econometrics and Data Group (HEDG) working papers. Retrieved from https://www.york.ac.uk/media/ economics/documents/herc/wp/08_17.pdf.
  69. Thomas, T. K. (2016). Role of Health insurance in enabling universal health coverage in India: A critical review. Health Services Management Research, 29(4), 99-106.
  70. Vera-Hernández, Á. M. (1999). Duplicate coverage and demand for health care. The case of Catalonia. Health Economics, 8(7), 579-598. https://doi.org/10.1002/ (SICI)1099-1050(199911)8:7<579::AID-HEC478>3.0.CO;2-P.
  71. Vuong, Q. H. (1989). Likelihood ratio tests for model selection and non-nested hypotheses. Econometrica, 307-333.
  72. Wagstaff, A. (1986). The demand for health: Some new empirical evidence. Journal of Health Economics, 5(3), 195-233. https://doi.org/10.1016/0167-6296(86) 90015-9.
  73. Wolfe, J. R., & Goddeeris, J. H. (1991). Adverse selection, moral hazard, and wealth effects in the Medigap insurance market. Journal of Health Economics, 10(4), 433-459. https://doi.org/10.1016/0167-6296(91)90024-H.
  74. Zweifel, P., & Breuer, M. (2006). The case for risk-based premiums in public health insurance. Health Economics, Policy and Law, 1(02), 171-188. https://doi.org/ 10.1017/S1744133105001064.
  75. Zweifel, P., Breyer, F., & Kifmann, M. (2009). Health economics. In Optimal health insurance contracts, pp. 293-310). New York: Springer [chapter 6].