Assessment of patients’ expectations: development and validation of the Expectation for Treatment Scale (ETS)

Abstract

ObjectiveTo develop a short self-report instrument for the assessment of expectations (Expectation for Treatment Scale(ETS)) using acupuncture as a case example.DesignA cross-sectional assessment with retest after 1 week.SettingA web-based survey with patients suffering from pain.MethodsIn a three-step approach, we reduced the initially collected number of items from 17 to 9 and to 5, including expectations about coping ability, vitality, physical health and reduction of patient complaints. Items were selected according to internal consistency (Cronbach’s alpha); convergent and divergent validities with related constructs (optimism, pessimism, resilience, perceived sensitivity to medicines, depression and others); 1-week retest reliability (intraclass correlation coefficient (ICC)); and exploratory and confirmatory factor analysis (CFA).ResultsA total of 102 patients suffering from pain were included, and 54 of these patients completed the retest assessment. The final version of the ETS consisted of five items and had an excellent Cronbach’s alpha (0.90), with 72.33% variance on one single factor. Depression, pessimism and perceived sensitivity to medicines showed positive correlations with our expectation measure (r=0.23, r=0.20 and r=0.34, respectively); the correlation between the ETS and optimism was low (r=−0.07) and no correlation between the ETS and resilience was found (r=−0.07). Convergent validity was confirmed with a high correlation (r>0.90) between ETS and a treatment-specific measure of expectations. The retest ICC was 0.86, which showed high stability over 1 week. A CFA (n=439) with data from patients with low back pain confirmed the single-factor structure of the instrument.ConclusionThe ETS showed strong psychometric properties and covered a distinct construct. As the next step, the ETS might be implemented in different clinical conditions and settings to investigate psychometrics and its predictive power for treatment outcomes.