Psychometric properties of the German Self-efficacy Expectation Scale (SES-G) in patients undergoing primary total hip arthroplasty: A validation study.

Objectives

To translate and culturally adapt the Self-efficacy Expectation Scale (SES) into German and examine its psychometric properties for total hip arthroplasty (THA) patients.

Methods

The English-language SES was translated into German (SES-G). 116 patients completed the SES-G together with the Self-Efficacy for Rehabilitation Outcome Scale-German version (SER-G), Hip Disability and Osteoarthritis Outcome Scale (HOOS), and Hospital for Special Surgery Hip Expectation Survey (HSS-HRES). Structural validity was assessed using an exploratory factor analysis (EFA). Construct validity was examined by testing predefined hypotheses on relationships between the SES-G and total- and subscores of the SER-G, HOOS, HSS-HRES. The SES-G was readministered after two weeks to assess test-retest reliability. To evaluate test-retest reliability 2-way random-effects absolute agreement intraclass correlation coefficients and Bland-Altman analysis were calculated. MacDonald's Omega was used to evaluate internal consistency. Standard error of measurement (SEM) and minimal detectable change (MDC) were explored.

Results

EFA revealed three components explaining 75% of the total variance. For construct validity, 80% of the hypotheses were confirmed. Internal consistency was good (MacDonald's Omega = 0.81), as was test-retest reliability (ICC 0.71). Mean difference between the first and second administrations was -1.6 ± 4.2. The corresponding 95% CI did not contain zero [-2.4, -0.8], indicating systematic bias. Limits of agreement were 6.7 and 9.8. Proportional bias can be ruled out (B = 0.187, p = 0.05). SEM was 3.08, MDCind and MDCgroup were 5.95 and 0.55, respectively.

Conclusion

The psychometric analysis provides preliminary evidence that the SES-G is a valid and reliable instrument to measure self-efficacy in THA patients.