Comparative cost-benefit analysis of tele-homecare for community-dwelling elderly in Japan: Non-Government versus Government Supported Funding Models
Miki Akiyama,Chon Abraham
DOI: https://doi.org/10.1016/j.ijmedinf.2017.04.017
Abstract:Objective: Tele-homecare is gaining prominence as a viable care alternative, as evidenced by the increase in financial support from international governments to fund initiatives in their respective countries. The primary reason for the funding is to support efforts to reduce lags and increase capacity in access to care as well as to promote preventive measures that can avert costly emergent issues from arising. These efforts are especially important to super-aged and aging societies such as in Japan, many European countries, and the United States (US). However, to date and to our knowledge, a direct comparison of non-government vs. government-supported funding models for tele-homecare is particularly lacking in Japan. The aim of this study is to compare these operational models (i.e., non-government vs. government-supported funding) from a cost-benefit perspective. This simulation study applies to a Japanese hypothetical cohort with implications for other super-aged and aging societies abroad. Methods: We performed a cost-benefit analysis (CBA) on two operational models for enabling tele-homecare for elderly community-dwelling cohorts based on a decision tree model, which we created with parameters from published literature. The two models examined are (a) Model 1-non-government-supported funding that includes monthly fixed charges paid by users for a portion of the operating costs, and (b) Model 2-government-supported funding that includes startup and installation costs only (i.e., no operating costs) and no monthly user charges. We performed base case cost-benefit analysis and probabilistic cost-benefit analysis with a Monte Carlo simulation. We calculated net benefit and benefit-to-cost ratios (BCRs) from the societal perspective with a five-year time horizon applying a 3% discount rate for both cost and benefit values. The cost of tele-homecare included (a) the startup system expense, averaged over a five-year depreciation period, and (b) operation expenses (i.e., labor and non-labor) per user per year. The benefit of tele-homecare was measured by annual willingness to pay (WTP) for tele-homecare by a user and medical expenditures avoided. Both costs and benefits were inflated using the relevant Japanese consumer price index (CPI) and converted into 2015 US dollars with purchasing power parity (PPP) adjusted. Results: Base case net benefits of Model 1 and Model 2 were $417.00 and $97.30, respectively. Base case BCR of Model 1 tele-homecare was 1.63, while Model 2 was 1.03. The probabilistic analysis estimated mean (95%CI) for BCRs of Model 1 and Model 2 was 1.84 (1.89, 1.88) and 1.46 (1.43, 1.49), respectively. Sensitivity analysis showed robustness of Model 1 in 7 parameters but Model 2 was sensitive in all key parameters such as initial system cost, device cost, number of users, and medical expenditure saved. Break-even analysis showed that the system cost of Model 2 had to be under $187,500. Conclusions: Our results for each model collectively showed that tele-homecare in Japan is cost-saving to some extent. However, the government-funded model (i.e., Model 2), which typically requires use of all startup funding to be spent within the first year on system costs, was inferior to the monthly fee model (i.e., Model 1) that did not use the government funding for installation or continued operations, but rather incorporated a monthly fee from users to support the receipt of services via tele-homecare. While the benefits of Model 1 outweighed the benefits of Model 2, the government-subsidized method employed in Model 2 could be more beneficial in general if some explicit prequalifying estimated metrics are instituted prior to funding. Thus, governments need to require applicants requesting funding to note, at a minimum, (a) estimated costs, (b) the expected number of tele-homecare users, and expected benefits such as (c) WTP by the user, or (d) medical expenditure saved by tele-homecare as a means of financing some of the operational costs.