Search

Abstract Background While there are a few studies on measurement properties of PROMIS short forms for pain and function in patients with knee osteoarthritis, nothing is known about the measurement properties in patients with knee arthroplasty. Therefore, this study examined the measurement properties of the German Patient-Reported Outcomes Measurement Information System (PROMIS) short forms for pain intensity (PAIN), pain interference (PI) and physical function (PF) in knee arthroplasty patients. Methods Short forms were collected from consecutive patients of our clinic's knee arthroplasty registry before and 12 months post-surgery. Oxford Knee Score (OKS) was the reference measure. A subsample completed the short forms twice to test reliability. Construct validity and responsiveness were assessed using scale-specific hypothesis testing. For reliability, Cronbach’s alpha, intraclass correlation coefficients, and agreement using standard error of measurement (SEMagr) were used. Agreement was used to determine standardised effect sizes and smallest detectable changes (SDC90). Individual-level minimal important change (MIC) was calculated using a method of adjusted prediction. Results Of 213 eligible patients, 155 received questionnaires, 143 returned baseline questionnaires and 119, 12-month questionnaires. Correlations of short forms with OKS were large (│r│ ≥ 0.7) with slightly lower values for PAIN, and specifically for men. Cronbach’s alpha values were ≥ 0.84 and intraclass correlation coefficients ≥ 0.90. SEMagr were around 3.5 for PAIN and PI and 1.7 for PF. SDC90 were around 8 for PAIN and PI and 4 for PF. Follow-up showed a relevant ceiling effect for PF. Correlations with OKS change scores of around 0.5 to 0.6 were moderate. Adjusted MICs were 7.2 for PAIN, 3.5 for PI and 5.7 for PF. Conclusion Our results partly support the use of the investigated short forms for knee arthroplasty patients. The ability of PF to differentiate between patients with high perceived recovery is limited. Therefore, the advantages and disadvantages should be strongly considered within the context of the intended use.

Abstract Purpose To examine the measurement properties of the German International Knee Documentation Committee Subjective Knee Form (IKDC-SKF) in knee disorder patients. Methods Three hundred twelve consecutive patients undergoing surgery for anterior cruciate ligament, meniscus and/or cartilage injuries completed the IKDC-SKF, Lysholm Score, Tegner Activity Scale, and Short Form-12 Health Survey before and 6 months post-surgery. IKDC-SKF measurement properties were calculated and patients were also asked to rate the relevance/comprehensibility of the questionnaire items. Results Reliability was good with high Cronbach’s alpha and intraclass correlation coefficients, and standard error of measurement values of 4.4 to 6.0. The smallest detectable change (SDC) ranged from 12.3 to 16.7 points. Validity was good with 90% of all hypotheses confirmed. Confirmatory factor analysis did not show adequate fitting indices within the model. Over half of the items were rated as essential, and all were well comprehended. The majority of hypotheses for responsiveness were confirmed. No floor and ceiling effects were observed. The area under the curve ranged from 0.82 to 0.89 and the minimal important difference was smaller than the SDC. Conclusions The German IKDC-SKF is a reliable outcome measure with good hypotheses testing and responsiveness, but its MIC and structural/content validity need further analysis.