FRAILTY, SARCOPENIA AND LONG TERM CARE UTILIZATION IN OLDER POPULATIONS: A SYSTEMATIC REVIEW.

This systematic literature review documents the link between frailty or sarcopenia, conceptualized as dimensions of physical health, and the use of long-term care services by older individuals. Long-term care services include formal and informal care provided at home as well as in institutions. A systematic review was performed according to PRISMA requirements using the following databases: PubMed-Medline, Embase, CINAHL, Web of Science, and Academic Search Premier. We included all quantitative studies published in English between January 2000 and December 2018 focusing on individuals aged 50 or more, using a relevant measurement of sarcopenia or physical frailty and a long-term care related outcome. A quality assessment was carried out using the questionnaire established by the Good Practice Task Force Report of the International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Five subsets of long-term care outcome were considered: 1/ nursing home placement (NHP), 2/ nursing home short stay (NHSS), 3/ formal personal care (FPC), 4/ formal home help (FHH), 5/ informal care (IC). Out of 1943 studies, 17 were finally included in the review. With some studies covering several LTC outcomes, frailty and / or sarcopenia were associated with increased LTC use in 17 out of 26 cases (NHP: 5/6, NHSS: 3/4, FPC: 5/7, FHH: 1/4, IC: 3/5) The association was not consistent in 5 cases (NHP: 1/6, NHSS: 1/4, FPC: 2/7, FHH: 0/4, IC: 1/5) and the association was either not significant or the results inconclusive in the remaining 9 cases. Overall, while results on sarcopenia are scarce, evidence support a positive association between frailty and LTC use. The evidence is stronger for the association of physical frailty with nursing home placement / short stay as well as on FPC. There is less (more heterogeneous) evidence regarding the correlation between physical frailty and FHH or IC use. Results need to be confirmed by more advanced statistical methods or design based on longitudinal data.

Déclaration de conflit d'intérêts

None of the authors report any conflict of interest whatsoever.