A Modified Version of the Function In Sitting Test (mFIST): Development and Initial Reliability of the mFIST in Outpatient Clinic in Veterans with Spinal Cord Injury

IntroductionTrunk control and seated balance are important for various activities of daily living in patients with spinal cord injuries (SCI). A clinically appropriate and reliable gold standard seated balance outcome measure for SCI is lacking. The Function in Sitting Test (FIST) has become popular in stroke and in-patient clinics for tracking seated balance. A modified version of the Function in Sitting Test for SCI patients may prove similarly useful for measuring and tracking seated balance in the clinic. ObjectiveThe primary objective of this study was to modify the FIST for the SCI population (mFIST) and assess initial reliability of the measure in a busy outpatient, real-world, clinical setting. The secondary objective was to determine trends in scoring differences if reliability was below excellent. DesignProspective reliability study. The mFIST was administered to Veterans with SCI twice within 2 weeks by the same evaluator (test re-test), while being video-recorded for subsequent scoring. SettingBusy Outpatient Clinic in Veteran Affairs Healthcare System. ParticipantsA total of 42 Veterans with SCI (mean age = 62, 91% male), between C4-L1 and with an American Spinal Injury Association Impairment Scale grade of A to D, participated in this study during routine outpatient clinic. Participants had to be primary wheelchair users and able to sit unsupported for at least 1 min. InterventionsNot Applicable. Main Outcome MeasuresTest re-test, intra-rater and inter-rater reliability using intra-class correlation coefficients were determined. Mean rating differences were calculated for individual mFIST items, total scores across all subjects, and when separated by SCI injury level grouping to investigate reliability trends. ResultsTest re-test and intra-rater reliability for the mFIST for the entire cohort were good with ICCs of 0.88 (CI 0.77 - 0.93) and 0.89 (CI 0.51-0.97), respectively. Inter-rater reliability was excellent with ICC = 0.92 (CI 0.85 - 0.96) across the entire cohort, however reliability dropped (0.78; p= 0.02) for certain SCI subgroups. The largest differences in scoring across reliability testing were seen during dynamic tasks in those with cervical injuries. ConclusionThe mFIST displays good to excellent reliability during routine use in a busy outpatient clinic. Inconsistencies in scoring of some of the dynamic items, however, indicate that the measure might need further wording and/or scoring refinements before being widely distributed. Determining the validity and sensitivity of the mFIST is also needed.