The psychometric properties of Observer OPTION(5), an observer measure of shared decision making

Abstract

OBJECTIVES: Observer OPTION(5) was designed as a more efficient version of OPTION(12), the most commonly used measure of shared decision making (SDM). The current paper assesses the psychometric properties of OPTION(5). METHODS: Two raters used OPTION(5) to rate recordings of clinical encounters from two previous patient decision aid (PDA) trials (n=201; n=110). A subsample was re-rated two weeks later. We assessed discriminative validity, inter-rater reliability, intra-rater reliability, and concurrent validity. RESULTS: OPTION(5) demonstrated discriminative validity, with increases in SDM between usual care and PDA arms. OPTION(5) also demonstrated concurrent validity with OPTION(12), r=0.61 (95%CI 0.54, 0.68) and intra-rater reliability, r=0.93 (0.83, 0.97). The mean difference in rater score was 8.89 (95% Credibility Interval, 7.5, 10.3), with intraclass correlation (ICC) of 0.67 (95% Credibility Interval, 0.51, 0.91) for the accuracy of rater scores and 0.70 (95% Credibility Interval, 0.56, 0.94) for the consistency of rater scores across encounters, indicating good inter-rater reliability. Raters reported lower cognitive burden when using OPTION(5) compared to OPTION(12). CONCLUSIONS: OPTION(5) is a brief, theoretically grounded observer measure of SDM with promising psychometric properties in this sample and low burden on raters. PRACTICE IMPLICATIONS: OPTION(5) has potential to provide reliable, valid assessment of SDM in clinical encounters.