Citation Context

...rofessor, Department of Agricultural and Resource Economics, North Carolina State University, 3332 Nelson Hall, Raleigh, NC 27695-8109, USA e-mail: tom_vukina@ncsu.edu Using the age-based insurance eligibility criterion to estimate moral hazard in medical care consumption YAN ZHENG, PhD* TOMISLAV VUKINA, PhD* Article** JEL: C14, D82, I11 doi: 10.3326/fintp.39.3.3 ya n zh en g a n d to m islav v u k in a: u sin g th e a g e-b a sed in su r a n c e elig ib ility c r iter io n to estim ate m o r a l h a za r d in m ed ic a l c a r e c o n su m ptio n fin a n c ia l th eo ry a n d pr a c tic e 40 (3) 337-356 (2016) 338 Abstract This paper uses fuzzy regression discontinuity design to estimate the moral hazard effect in health care consumption in the population of young adults. We use invoice data for outpatient hospital services from a regional hospital in Croatia. The estimation is complicated by the fact that the data set consists only of users of medical services, which would tend to underestimate the moral hazard effect. To address this issue we use a modified version of the instrumental variables approach. We find a 92% reduction in the number of hospital visits for individuals who l...

Citation Context

...nkelstein, 2011; Chetty and Finkelstein, 2012). In this paper, we disentangle price effects from selection effects (those going into retirement might be less healthy), by exploiting a jump in the probability of retirement that occurs at the legal retirement age (65 years). This allows us to apply a regression discontinuity (RD) design, which is highly regarded for its internal validity (Lee and Lemieux, 2010). Indeed, discontinuities in insurance coverage at specific ages have yielded robust estimates of the effect of health insurance on health care use and health outcomes (Card et al,. 2008; Card et al., 2009; Shigeoka, 2011 and Anderson et al., 2012). Closer to our main object of interest, Chandra et al. (2010) examined policy changes for Californian civil servants under Medicare program that increased the level of co-payment both for physician visits and prescription drugs. They provide one of the first robust estimations of price elasticity of pharmaceuticals for the elderly, as this group was excluded from the RAND Health Insurance Experiment. 4 In this paper, we combine two different data sources to obtain the price-elasticity estimates. Administrative data on individuals aged 63-67 covered b...

Citation Context

...ts are not reported), SES hardly explains all the correlation between subjective bias and insurance participation. It is possible that people who are insured or go to the doctor frequently are more likely to know their health status by taking more examinations and getting alerts from their doctors. Thus the associations in first two columns may be driven by reverse causality. It is difficult to rule out this possibility. But the robustness checks by restricting the sample to those who are never notified by the doctors about osteoporosis yield very consistent results. Following the strategy in Card et al. (2009), we also did another set of regressions to use the large jump at 65 in the fraction who have Medicare as their primary insurer as an exogenous shock in health insurance coverage to see whether insurance helps people to realize their shrinkage. The results, not provided here, provide no evidence for this hypothesis. 3.4 Insufficient Insurance due to Subjective bias Based on the estimates in Table 2, we are able to calculate how much distortion there is due to subjective bias and the distribution of such distortion over health status. First, we use the 14 estimates in column 1 directly to predi...

Citation Context

...ysiotherapy was removed from the benefits package (Mladovsky et al., 2012) – it remains to be established whether the cuts have had a relevant impact on long-term care needs. Again, there is an identification problem which makes it difficult to draw clear inference and there is no general agreement as to whether increases or reductions in coverage lead to discernible changes in population health (Moreno-Serra and Smith, 2012). Some American studies use the eligibility for Medicare, which leads to a sharp change in health care coverage at age 65, to estimate the effect in the older population. Card et al. (2007) reports that Medicare eligibility is associated with a significant reduction in patient mortality. The Oregon experiment, which randomly assigned health insurance coverage, has also been evaluated in various studies (Allen, 2015). Finkelstein et al. (2011) observe a positive effect on utilization and on self-reported health. The evidence for objective health measures is much more mixed: no effect was observed for detection of hypertension or indicators such as blood pressure and cholesterol, but there is some evidence that diabetes was better diagnosed and treated, and that mental health impr...

Citation Context

... by the demand for health care (δM/δφ) and its effectiveness (δH/δM). Both are areas of major preoccupation within the field of health economics (e.g., Manning et al., 1987; Finkelstein et al., 2011; =-=Card et al., 2009-=-; Cutler, 2007; Lichtenberg, 2007). When combined they reveal the transient and perhaps residual, untreatable health effects due to pollution (as implied by δH/δφ). To be concrete, imagine a patient t...