The Tilburg frailty indicator: a psychometric systematic review

Abstract

The Tilburg Frailty Indicator (TFI) is one of the most prominent multidimensional frailty assessment instruments. This review aimed to critically appraise and summarise its measurement properties. Methods Reports were eligible if they included results of studies aimed at developing the TFI or evaluating its measurement properties. We performed a literature search in MEDLINE, CINAHL, and PsycINFO databases from their inception until September 17, 2020. We also searched grey literature databases. We assessed the methodological quality of the included studies using the 'COSMIN Risk of Bias'. The measurement properties were evaluated using specific criteria. We graded the quality of the evidence using a GRADE approach. Results Fifty-five studies were included. We found moderate sufficient evidence for TFI content validity, although it is still insufficient for the comprehensiveness of its items. TFI construct validity was based on sufficient evidence from a single study of its structural validity as well as multiple hypothesis-testing for construct validity studies with inconsistent results. We did not find any studies that assessed cross-cultural validity. The TFI showed high sufficient concurrent validity with the comprehensive geriatric assessment. We identified several studies assessing its predictive validity for adverse frailty-related outcomes, although most of the evidence from these studies was insufficient. We did not find any studies that assessed the responsiveness of TFI scores. Conclusions The TFI is a simple measurement instrument that may be helpful in the assessment of frailty. However, more studies are needed to strengthen its usefulness as a clinical decision-making tool.