Defining substantial clinical benefit for patient-rated outcome tools for shoulder impingement syndrome.

OBJECTIVE: To define for 2 shoulder outcomes scales the substantial clinical benefit (SCB)-a metric that defines the change amount associated with patient perception of a large meaningful improvement and that can be used to interpret change over time in the outcome score.
DESIGN: Cohort.
SETTING: Clinic. PARTICIPANTS: Patients (N=74) with shoulder impingement syndrome.
INTERVENTIONS: Standardized exercise and manual therapy for 6 weeks, and outcome measures completed at initial evaluation, discharge, and 6 to 8 weeks postdischarge. MAIN OUTCOME MEASURES: Disabilities of the Arm, Shoulder and Hand (DASH), Pennsylvania Shoulder Score (Penn), and a 13-point Global Rating of Change (GROC). Patients were classified as "substantially improved" when they reported "quite a bit better" (11) or greater on the GROC at discharge and again 6 to 8 weeks after discharge. Patients with GROC <11 at discharge or follow-up were classified as "nonsubstantially improved." The percentage and raw points change in the Penn and DASH that corresponded with patient-rated substantial improvement was determined with receiver operator characteristic (ROC) analyses.
RESULTS: ROC analyses revealed the SCB for the DASH was 40% (area under the curve [AUC]=.79; confidence interval [CI], .69-.89) and 11 points (AUC=.63; CI, .50-.76); and for the Penn, 20% (AUC=.76; CI, .65-.87) and 21 points (AUC=.80; CI, .69-.90).
CONCLUSIONS: The SCB of 40% for the DASH, and 20% and 21 points for the Penn represents substantial improvement over 6 weeks of care, which was sustained at 12 weeks. The SCB of 11 points for the DASH is not recommended for use because of poor discrimination. The SCB can be used to enable clinical decision-making and in future clinical trials. Alternative approaches such as the within- and between-group change values can produce different SCB values.