Cut points and sensitivity to change of the Enfranchisement scale of the Community Participation Indicators in adults with traumatic brain injury

Background: Community participation is an important outcome of rehabilitation following traumatic brain injury. Yet, few measures assess inclusion and belonging (enfranchisement) as a dimension of community participation. The Enfranchisement scale of the Community Participation Indicators addresses this need. However, research on its psychometric properties is lacking. Objective: To examine cut points and sensitivity to change of the Enfranchisement scale of the Community Participation Indicators in adults with traumatic brain injury. Design: This was a repeated measures study with assessments administered twice (3 months apart). Setting: Assessments were administered either over the phone, virtually (Zoom), or in person at the participant's home. Participants: A total of 44 participants from community settings who had either experienced a traumatic brain injury within the previous year or were receiving rehabilitation interventions were recruited. Main Outcome Measure: The Enfranchisement scale has two subscales: the Control subscale (range: 13-65) and the Importance subscale (range: 14-70). On both subscales, lower scores indicate better enfranchisement. Methods: The software SAS PROC Logistic and the macro %ROCPlot were used to examine cut points at varying levels of sensitivity and specificity. The area under the receiver operating characteristics curve was calculated to determine overall classification accuracy. Minimum detectable change and minimal clinically important difference were also calculated. Results: For the Control subscale, a cut point of 44 (area under the curve =.75), a minimum detectable change of 8, and a minimal clinically important difference of 5 were found. For the Importance subscale, a cut point of 39 (area under the curve =.81), a minimum detectable change of 8, and a minimal clinically important difference of 5 were found. Conclusions: The cut points resulted in good classification accuracy, providing support for their reliability. The results provided evidence that both subscales are sensitive to change in adults with brain injury.