• Te-Fen Lo and Chee-Ruey Hsieh.
• Global Health Research Center, Duke Kunshan University, Jiangsu Province, 215347, China, chee.ruey.hsieh@duke.edu.
• J Ment Health Policy Econ. 2014 Sep 1; 17 (3): 107-17.

BackgroundAs the population ages, the number of people living with Alzheimer's disease (AD) has been increasing over time. Between 1999 and 2006, four new AD drugs were approved for insurance coverage in Taiwan.Aims Of The StudyWe investigate the heterogeneous influences of adopting these new drugs on healthcare expenditures. We also evaluate whether the adoption of pharmaceutical innovation creates any "offsetting" effect in the sense that an increase in one component of the health care costs is offset by the decrease in the other components of health care costs.MethodsAD patients are defined in this study as those who are diagnosed with ICD-9-CM code 331.0. Based on longitudinal insurance claims data, we identify 1,088 AD patients whose first diagnosis occurred between 1997 and 2007. After excluding the period before each patient's AD diagnosis, 4,629 patient-year observations are found to constitute an unbalanced panel data set used in this study. We employ the correlated-random-effects quantile regression (CREQR) method to explicitly control for the unobserved heterogeneity and to consider the heterogeneous influences of adopting new drugs on different points of the conditional distribution function of health expenditures.ResultsOur empirical findings are consistent with previous evidence that the adoption of pharmaceutical innovation is costly. To be specific, the expenditure-increasing effect is mainly reflected by the increase in expenditure on drugs in the outpatient sector. In addition, we find evidence of a significant offsetting effect in the sense that new-drug users tend to make less use of inpatient services. As a result, the net effect of adopting pharmaceutical innovation is heterogeneous across AD patients: the use of new drugs is expenditure-increasing for patients whose health care costs are distributed below the 75th quantile; by contrast, the use of new drugs is expenditure-neutral above the 75th quantile, that is, the increase in the drug costs is almost completely offset by the decrease in the inpatient expenditure.DiscussionThe adoption of pharmaceutical innovation for treating AD is associated with a significant offsetting effect for higher cost patients. Our results also show that the CREQR method supplements the traditional ordinary least squares (OLS) method to provide interesting information beyond the conditional mean of the distribution. In our study, CREQR estimates suggest that the marginal impact of adopting pharmaceutical innovation on health care costs is heterogeneous across AD patients.Implications For Health PoliciesGiven that the impact of adopting new AD drugs on health care costs is not uniform among patients, current payment regulations that impose simple clinical criteria to decide the eligibility of using new drugs, that is, a policy that adopts one size to fit for all, may become an access barrier to realizing the potential benefits of pharmaceutical innovation.Implications For Future ResearchAnother potential source of the offsetting effect is that adopting new AD drugs may be beneficial to the reduction in the cost of long-term care. This is an important avenue for future research.

24 keywords...

hide…