Establishing anchor-based minimal important change values for the functional independence measure in older patients with vertebral compression fractures in Japan: A retrospective observational study

In rehabilitation medicine, the functional independence measure (FIM) is widely used to assess activities of daily living, but its minimal important change (MIC) in patients with vertebral compression fractures (VCFs) remains unclear. The MIC is defined as the smallest within-person change perceived by patients or clinicians to represent an important improvement. The MIC is crucial in evaluating the effectiveness of interventions across various studies and in setting of treatment goals. Accordingly, this study aimed to estimate the MIC of the FIM in patients with VCFs. This retrospective observational study estimated the MIC (the receiver operating characteristic method [minimal important change estimated using the receiver operating characteristic method [MICroc]], the predictive modeling method [minimal important change estimated using the predictive modeling method [MICpred]], the adjusted predictive modeling method [adjusted minimal important change [MICadj]]) of the FIM in 290 older patients with VCFs admitted at a convalescent rehabilitation ward in Japan between January 2020 and December 2022, using discharge destination (home vs elsewhere) as the anchor. The sample size was determined in accordance with the recommended criterion for MIC estimation (≥100 participants), and all eligible participants available during the study period were included. The 95% confidence intervals for the MIC estimated using each method (MICroc, MICpred, and MICadj) were calculated using the bootstrap method (n = 2000). The mean age of the participants was 82.61 ± 7.08 years, and 216 (74.48%) were female. While 254 participants (87.59%) were discharged home, 36 (12.41%) were discharged to other facilities, such as nursing homes. The MICroc, MICpred, and MICadj, respectively, were 31.5, 26.39, and 22.15 for functional independence measure motor items, while these were 0.5, 0.6, and -1.09 for functional independence measure cognitive items. For functional independence measure total score, these were 40.5, 27.01, and 21.53, respectively. This study is the first to estimate the MIC of the FIM using multiple methods in older patients with VCFs admitted at a convalescent rehabilitation ward, providing new insights into the definition of clinically meaningful changes in rehabilitation assessment. Future studies should incorporate multidimensional anchors, including patient-reported outcomes, and employ more rigorous prospective study designs and multicenter collaborations.