Validity and reliability of the Turkish version of the Spinal Cord Independence Measure (SCIM III) self-report.

Study design: A cross-sectional validation study.

Objectives: The aim of the study was to create a Turkish version of the Spinal Cord Independence Measure Self-Report (SCIM-SR) scale and verify its validity and reliability.

Setting: Rehabilitation Department at Istanbul Physical Medicine and Rehabilitation and Training Research Hospital.

Methods: The Turkish version of SCIM-SR was translated from English using forward and backward translation in accordance with standard guidelines and 90 participants completed the SCIM-SR Turkish form. At the same time, a Spinal Cord Independence Measure III (SCIM III) was administered to the participants by a health professional through observation. Cronbach's alpha, Pearson and Spearman correlation coefficients were used to evaluate the internal consistency and concurrent validity.

Results: The Cronbach's α coefficients for the total scale was 0.883. For the subscales, it was 0.929 for self-care, 0.571 for respiration and sphincter management and 0.885 for mobility. Test-retest reliability after two weeks yielded Spearman coefficients above 0.92 for the total scale and all subscales. When analyzing the correlation between SCIM-SR and SCIM III, the intraclass correlation average and its lower and upper bound values were as follows: 0.984 (0.976-0.990) for self-care; 0.970 (0.955-0.980) for respiration and sphincter management; 0.968 (0.951-0.979) for mobility and 0.980 (0.969-0.987) for the total. There was good agreement between SCIM-SR and SCIM III when evaluating the total score in Bland-Altman graphs. The Bland-Altman analysis revealed no significant proportional bias (0.76; 95% CI [-21.34, 22.87]), but a significant fixed bias was observed.

Conclusions: The Turkish version of SCIM-SR demonstrated excellent internal consistency and good validity in Turkish individuals with spinal cord injury.