Potentially avoidable hospitalizations, non-potentially avoidable hospitalizations and in-hospital deaths among residents of long-term care facilities

Abstract

Aim: The present study aimed to examine the percentage of and risk factors for potentially avoidable hospitalizations (PAH), non-PAH and in-hospital deaths among residents of special nursing homes for the elderly (SNH) and geriatric health service facilities (GHSF).

Methods: Long-term care and national health insurance claims data (April 2012 to September 2013) were obtained from a suburban city in Chiba prefecture, Japan. Study participants were aged ≥75 years and resided in either SNH (n = 1138) or GHSF (n = 885). The PAH were defined using 17 medical condition groups, and the percentage of PAH, non-PAH and in-hospital deaths was identified, and associated factors were compared using multilevel logistic regression models for SNH and GHSF, respectively.

Results: A total of 34.5% SNH residents experienced any hospitalization, and this was composed of PAH (16.3%), non-PAH (12.2%) or in-hospital deaths (6.1%). Of the GHSF residents, 23.8% experienced any hospitalization, and this was comprised of PAH (9.5%), non-PAH (10.6%) and in-hospital death (3.7%). More than 70% of the PAH were related to respiratory infections, urinary tract infections or congestive heart failure. In both SNH and GHSF, artificial nutrition was positively associated with PAH and non-PAHs, and male sex was positively associated with non-PAHs and in-hospital deaths. However, there were also discrepancies between SNH and GHSF in terms of risk factors for PAH.

Conclusions: The percentage of PAH was higher in SNH than in GHSF, which might be related to their different personnel and managerial regulations. The linkage of health and long-term care claims data might facilitate data-based evidence on policy-making. Geriatr Gerontol Int 2018; 18: 1272-1279.

Keywords: health services research; hospitalization; long-term care; nursing homes; quality of care.