Minimal clinically important difference of the functional independence measure in older adults with hip fracture

Figures & data

Table 1. Hypotheses formulated a priori to assess the responsiveness of the FIM scale in older adults with hip fracture.

Hypothesis	Met/unmet hypothesis
1. The ES of rehabilitation program for older adults with hip fracture measured with the FIM scale is moderate to high and comparable to the values reported for heterogeneous inpatients [29].	+
2. The correlation between the FIM change total score and the BI is strong (≥0.70), as reported in previous studies [16,30].	−
3. The correlation between the FIM change total score and the FAC score is at least moderate (≥0.40).	+
4. The correlation between the FIM change total score and the checklist of transfer independence is at least moderate (≥0.40).	+
5. At discharge, individuals classified as “improved” with the BI (MCID of 7.1) display a larger change in the FIM total score than those “not-improved”.	+
6. At discharge, individuals classified as “functional ambulators” with the FAC score (≥3) display a larger change in the FIM total score than those “non-functional ambulators” (<3).	+
7. At discharge, individuals classified as “independent” at the checklist of transfers display larger change in the FIM total score than those “not-independent”.	+
8. The AUC for the FIM, using the BI as the anchor, is above 0.70. In order to calculate the AUC, individuals were divided into “improved” and “not-improved” according to the MCID of BI specific for people with hip fracture during rehabilitation.	+
9. The AUC for the FIM, using the FAC score as the anchor, is above 0.70. In order to calculate the AUC, individuals were divided into “functional ambulators” and “non-functional ambulators” according to their FAC score (<3 or ≥3, respectively).	+
10. The AUC for the FIM, using the checklist of transfers as the anchor, is above 0.70. Individuals were divided into “independent” and “not-independent” to calculate the AUC.	+
Total hypotheses met	9/10 (90.0%)

+: met hypothesis; −: unmet hypothesis.

Table 2. Clinical characteristics of older adults with hip fracture included in the study (n = 701).

Characteristics	All participants (n = 701)
Age, mean (SD), y	83.3 (7.2)
Sex, female, n (%)	543 (77.5)
BMI, mean (SD), kg/m^2	24.1 (4.2)
Type of fracture, n (%)
Intracapsular hip fracture	380 (54.2)
Extracapsular hip fracture	321 (45.8)
Time between surgery and admission, mean (SD), days	8.8 (3.9)
Length of stay, mean (SD), days	30.3 (8.4)
Discharge destination, n (%)
Home	655 (93.4)
Nursing home	46 (6.7)
FIM at admission, mean (SD), score
Motor	31.7 (9.5)
Cognitive	28.7 (6.2)
Total	60.4 (14.0)
FIM at discharge, mean (SD), score
Motor	55.0 (15.9)
Cognitive	29.7 (5.7)
Total	84.8 (20.3)
Barthel Index, mean (SD), score
At admission	39.0 (12.6)
At discharge	62.4 (21.0)
FAC, n (%)
Non-functional ambulators (FAC < 3)	272 (38.9)
Functional ambulators (FAC ≥ 3)	429 (61.1)
Checklist question about transfers, n (%)
Not-independent	306 (43.6)
Independent	395 (56.4)

Figure 1. ROC curves of the FIM total (A) and motor (B) score considering the MCID of the BI, the FAC score, and the checklist question as anchors. ROC shows the overall accuracy of the FIM in identifying an improvement in the level of independence in individuals with hip fracture according to the three anchors. (A) For the total score, the AUCs (95% C.I.) were 0.82 (0.78 − 0.86) when the MCID of the BI was used as an anchor, 0.81 (0.78 − 0.85) when cut-off of 3 of the FAC score was the anchor, 0.85 (0.81 − 0.87) for the checklist on independence in performing transfers. (B) For the motor score, the AUCs (95% C.I.) were 0.83 (0.79 − 0.87) when the MCID of the BI was used as an anchor, 0.82 (0.79 − 0.85) when cut-off of 3 of the FAC score was used, 0.86 (0.83 − 0.89) for the checklist on independence in performing transfers.

Two receiver operating characteristics (ROC) graphs of the Functional Independent Measure total score and of the motor subscore. The areas under the ROC curves plotted with the three anchors are similar.

Figure 2. MCID for the FIM total and motor score using both anchors- and distribution-based methods. The dark grey circles represent the MCID of the FIM total score, while the light grey square the MCID of the FIM motor score. The dark and light grey dashed lines represent the minimum detectable change (MDC) of the total (10.3 points) and the motor (7.7 points) FIM score, respectively.

Graph of estimated Minimally Clinical Important Difference (MCID) of both the Functional Independent Measure total score and the motor subscore where the values calculated with the distribution-based methods are below the Minimal Detectable Change. Values of each estimated MCID are reported in Supplementary Material.

Darzins SW, Imms C, Shields N, et al. Responsiveness, construct and criterion validity of the personal care-participation assessment and resource tool (PC-PART). Health Qual Life Outcomes. 2015;13:125.

 PubMed Web of Science ®Google Scholar

Hsueh IP, Lin JH, Jeng JS, et al. Comparison of the psychometric characteristics of the functional independence measure, 5 item barthel index, and 10 item barthel index in patients with stroke. J Neurol Neurosurg Psychiatry. 2002;73(2):188–190.

 PubMed Web of Science ®Google Scholar

Houlden H, Edwards M, McNeil J, et al. Use of the barthel index and the functional independence measure during early inpatient rehabilitation after single incident brain injury. Clin Rehabil. 2006;20(2):153–159.

 PubMed Web of Science ®Google Scholar

Data availability statement

The data that support the findings of this study are available from the corresponding author, MG, upon reasonable request.