VP.13 Updates on the development of the spinal muscular atrophy - person-reported outcome (SMA-PRO): A caregiver and self-proxy performance measure for child

Abstract

There is a need for person-reported outcome measures for SMA. The PEDI-CAT is a caregiver-reported measure used to describe functional skill performance. Convergent validity of the PEDI-CAT with SMA phenotypes was established, and Rasch analysis of the full item bank revealed representation of abilities for SMA III and for daily activities for SMA II, but a floor effect for SMA I. Results led to the development of a new measure to include a larger item bank with caregiver and self-proxy options. Goals were 1) To describe the SMA PRO test development, including focus groups and cognitive-testing for item content 2) To describe the development of branching logic for data collection and 3) To report interim results on data collected with the new measure. Clinicians with expertise in SMA participated in 4 focus groups. Clinicians reviewed the results of the Rasch analysis and discussed content, scoring and clarity for the 76 daily activities and 105 mobility items. Thirty mobility and 69 daily activities items were added to address the floor effect and include items for infants and adults. A power wheelchair domain was created with 35 items and illustrations including "maintaining hand on wheelchair joystick while driving on uneven terrain." Illustrations were re-created to increase relevance for adults and children. The new items and illustrations were reviewed and several items were discarded due to perception of redundancy. Clinicians then conducted cognitive interviews with 15 study participants (7 caregivers and 8 adolescents and adults with SMA). A maximum of 30 items with accompanying illustrations were reviewed at each interview with probing questions including, "In your own words, what do you think this question is asking?" and "Does the illustration match the item wording?" Cognitive interview results were collated and items were further edited. The final items and illustrations were added to a database, and branching logic was established based on pre-determined "stopping rules" to minimize the total number of items required for each respondent. Item surveys were sent to participants and caregivers via a secure email, and data was collected prospectively. At the time of this interim study point, 58 surveys had been completed (16 "non-sitters," 38 "sitters" and 25 "walkers") across 4 functional domains. Data collection is underway, and a full analysis will be performed in order to validate the measure.