Comparing Two Frailty Tools for the Prediction of Mortality in Older People Admitted to Hospital in Tanzania: A Prospective Cohort Study.

BACKGROUND: Healthcare systems in low- and middle-income countries are under pressure to adapt services for ageing populations, and frailty screening has gained interest as a means of identifying older adults at risk of adverse outcomes. AIMS: The aim of this study was to examine two frailty instruments for the prediction of mortality in older adults admitted to hospital in Tanzania. METHODS: This study followed 308 people, aged ≥ 60 years, admitted to four hospitals in the Kilimanjaro Region between March and August 2022. Baseline frailty was assessed using the Clinical Frailty Scale (CFS) and the Brief Frailty Instrument for Tanzania (B-FIT). The primary outcome was all-cause mortality, with secondary outcomes of length of stay and change in functional status. Cox regression estimated hazard ratios (HRs) for mortality, with Harrell's C statistic and receiver operator characteristics used to compare instrument performance. RESULTS: Follow-up data were available for 194 participants (63%). After a mean of 10.8 months of follow-up, adjusted models demonstrated frailty was associated with higher mortality (CFS HR 2.26, 95% CI 1.34-3.83; B-FIT HR 2.63, 95% CI 1.45-4.77, p = 0.001). The two instruments showed similar discriminatory power. Neither was associated with length of stay. Among survivors, frailty was associated with decline in physical function, but not social function. CONCLUSIONS: Frailty screening, using either the CFS or B-FIT, offered a feasible method for identifying older adults at the greatest risk of adverse outcomes. Such tools should be investigated as a means of targeting limited resources towards the most vulnerable older adults in an African setting.