The Effect of Taiwan's National Health Insurance on Mortality of the Elderly: Revisited

Abstract

A recent paper estimates the effects of Taiwan's National Health Insurance (NHI) on the elderly and concludes that NHI greatly increased the medical care utilization of the elderly but did not reduce their mortality. Using more recent and more accurate mortality data of the same group of elderly, this note re-estimates the NHI effect on mortality and finds that the mortality hazard of the previously uninsured elderly in the post-NHI period was on average 24% lower than it would have been in the absence of NHI. However, the NHI effect on the mortality hazard is only evident in the first 6 years following the enactment of NHI, suggesting that it may be difficult to undo the damage caused by the lack of insurance in early life.

The Impacts of China's Urban Employee Basic Medical Insurance on Healthcare Expenditures and Health Outcomes

“…Following the literature (Chen et al, 2007;Ding and Zhu, 2007;Chang, 2012;Liu and Zhao, 2014), we also control for a set of demographic, socioeconomic, and health characteristics in the model. The demographic characteristics that we consider are age, gender, marital status, and residential area; the socioeconomic characteristics that we control for are education and household income level.…”

Universal Coverage on a Budget: Impacts on Health Care Utilization and Out-of-Pocket Expenditures in Thailand

“…It is more relevant than evidence on the effects of health insurance obtained from marginal changes in coverage, even when this is obtained experimentally (Newhouse, 1993;Finkelstein et al, 2012), because supply-side responses can make the impact of a large scale expansion of coverage disproportionate to that of a small scale increase (Finkelstein, 2007;Kondo, Shigeoka, 2013). Evidence from a middle-income country that implemented universal coverage a decade ago is more relevant to emerging economies contemplating this step than is evidence from the experience of the US and Japan fifty years ago (Finkelstein, 2007;Finkelstein, McKnight, 2008;Kondo, Shigeoka, 2013), and even from Taiwan (Chen et al, 2007;Chou et al, 2011;Chang, 2012;Keng, Sheu, 2013) that implemented universal coverage in 1995 at a much higher level of income than that of Thailand in 2002. Our evidence is distinguished from that obtained from major expansions of health insurance in the middle income settings of Colombia and Mexico not only by geography and the lower income of Thailand, but also by the nature of the reform in Colombia, which offered a means-tested subsidy for the purchase of private insurance covering care at a restricted network of providers (Miller et al, 2012), and the nature of the evidence in Mexico, which comes from an experiment that offers only a ten-month window within which to identify the effect on treated individuals offered the opportunity to voluntarily enroll relative to controls (King et al, 2009;Barofsky, 2011).…”

Recent Trends in Taiwanese Medical Spending

“…Average end-of-life hospital spending is more than 30 times average hospital spending by the total population, and it accounts for 15.9 per cent of total hospital care expenditure (or 5.0 per cent of overall medical spending), even though the relative population near death accounts for 6 See the literature review in De Nardi et al (this issue). For the Taiwanese context, see Chang (2012) and Chang, He and Hsieh (2014). 7 For example, Spillman and Lubitz (2000).…”