Outcomes of a financial incentive scheme for dementia care by dementia specialist teams in acute-care hospitals: A difference-in-differences analysis of a nationwide retrospective cohort study in Japan.

OBJECTIVES: The quality of care for dementia in acute-care settings has been criticised. In 2016, the Japanese universal health insurance system introduced a financial incentive scheme for dementia care by dementia specialist teams in acute-care hospitals. This study aimed to investigate the effectiveness of this financial incentive scheme on short-term outcomes (in-hospital mortality and 30-day readmission). DESIGN AND METHODS: Using a Japanese nationwide inpatient database, we identified older adult patients with moderate-to-severe dementia admitted for pneumonia, heart failure, cerebral infarction, urinary tract infection, intracranial injury or hip fracture from April 2014 to March 2018. We selected 180 propensity score-matched pairs of hospitals that adopted (n = 180 of 185) and that did not adopt (n = 180 of 744) the financial incentive scheme. We then conducted a patient-level difference-in-differences analysis. In a sensitivity analysis, we restricted the postintervention group to patients who actually received dementia care. RESULTS: There was no association between a hospital's adoption of the incentive scheme and in-hospital mortality (adjusted odds ratio [aOR]: 0.97; 95% confidence interval [CI]: 0.88-1.06; p = 0.48) or 30-day readmission (aOR: 1.04; 95% CI: 0.95-1.14; p = 0.37). Only 29% of patients in hospitals adopting the scheme actually received dementia care. The sensitivity analysis showed that receiving dementia care was associated with decreased in-hospital mortality. CONCLUSIONS: The financial incentive scheme to enhance dementia care by dementia specialist teams in Japan may not be working effectively, but the results do suggest that individual dementia care was associated with decreased in-hospital mortality.

Comparison of progression of care-need levels among long-term care recipients with and without advanced care management in a rural municipality of Japan: A population-based observational study.

BACKGROUND: Little is known about the association between advanced care management and patient outcomes in home settings. In 2009, the Japanese government introduced a financial incentive scheme for advanced care management by long-term care agencies with at least one advanced care manager. However, it remains unclear whether advanced care management in rural areas is associated with improved outcomes for recipients. OBJECTIVE: To compare the progression of care-need levels among long-term care recipients in home settings with and without advanced care management. DESIGN: A population-based observational study. SETTING: A rural municipality in Japan. PARTICIPANTS: Recipients of long-term care (n = 2005). METHODS: We used individual-level secondary data provided from a rural municipal government that was collected as part of the Survey of Long-Term Care Benefit Expenditures and medical care claim records between April 2012 and March 2017. We linked these two databases using unique identifiers. The inclusion criteria for study subjects were that they: (1) were aged ≥65 years; (2) were newly certified as care-need level 1, 2, or 3; and (3) used long-term care insurance services in home settings from April 2012 through March 2017. We excluded individuals using long-term care insurance services for less than six months. Ultimately, we selected 1722 propensity-matched recipients with and without advanced care management, and then conducted Kaplan-Meier survival analyses and a log-rank test. The outcome measure was progression of care-need levels. RESULTS: The proportions of five-year cumulative progression-free survival in the groups with and without advanced care management were 50.3% and 42.2% for recipients of care-need level 1 (p < .01), 34.3% and 32.3% for recipients of care-need level 2 (p < .01), and 22.3% and 24.5% for recipients of care-need level 3 (p > .05), respectively. The progression-free period lasted a median of 12 (interquartile range, 10-24) months for recipients of care-need level 1, 14 (interquartile range, 11-28) months for recipients of care-need level 2, and 12 (interquartile range, 10-24) months for recipients of care-need level 3. CONCLUSIONS: Long-term care recipients with advanced care management in home settings had a higher probability of progression of care-need levels in a rural municipality of Japan. This finding suggests that the governmental policy of providing financial incentive for advanced care management may not be effective in improving the outcome of long-term care recipients in a rural municipality of Japan.