In this study the investigators evaluate the reliability, validity and responsiveness of the SPEQ and providing an estimate for Minimal Important Difference MID. They will use a longitudinal multicenter design with an adequate sample size of n=250 for validity and n=80 for reliability evaluation. Private practices and the Department of Physiotherapy of the Inselspital, Bern University Hospital, will participate in the study.
Reliability will be evaluated in a test-retest setting and with ordinal alpha. The SPEQ will be compared to the Euroqual-5-Dimensions-5-Levels (EQ-5D-5L), Short Form 12 (SF-12) and Patient Satisfaction Questionnaire (PSQ) to evaluate the validity. Responsiveness will be evaluated using a criterion validity approach and hypothesis testing of the correlation between SPEQ and the other questionnaires. In addition, the investigators will calculate ES, Standardized Response Mean (SRM), Modified SRM (MSRM) and relative efficiency between SPEQ and the criterions. For MID estimates, they use a combined anchor and distribution based approach. The anchors for the items are provided by the T0- and T1-questionnaire.
Patients will be recruited by the treating physiotherapists and provide written informed consent. At baseline, they will complete the T0-questionnaire, SF-12 and EQ-5D-5L. At the end of treatment or after 6 month of treatment for persons in long term treatment, the patients will fill out SPEQ, the T1-questionnaire, SF-12, EQ-5D-5L, and the PSQ. Two to seven days after the second test series, a subset of patients(n=80) will be asked to answer the SPEQ questions again by telephone.
Descriptive statistics will describe missing data, distribution of data over scale levels, frequency, floor and ceiling effects. Scaling will be tested by comparing pearson and spearman correlations and effect size (ES) for ordinal data and for metric data. Independence will be evaluated with the Mann-Whitney U test. The assumptions for building a sum of score will be evaluated comparing and assessing the spearman correlations and a categorical principle component analysis of the matrix of the 4 items of the SPEQ.
The reliability will be evaluated. Correlations between the SPEQ and the criterions EQ-5D-5L, SF-12 and PSQ will demonstrate validity.
Responsiveness will be evaluated analyzing the correlations between SPEQ and criterion and hypothesis testing of change of scores (Cosmin). Furthermore we will calculate the several traditional statistics; Effect Size, Standardized Response Mean, Modified Standardized Response Mean and relative efficiency between SPEQ and the criterion; distribution based Minimal Important Change estimates will be calculated with 0.5 Standard Deviation, Standard Error Measurement and Minimal Detectable Change at the 95% confidence interval; the mean between T0 and T1 values are anchor-based MID estimates. The range of anchor and distribution based MIC estimates will be analysed by Receiver Operating Characteristics curves. The result of the analysis provides an estimate that will be compared to the MID estimate of the criterion validity approach. Finally the threshold for clinical relevant change for the items and the sum of scores of the SPEQ will be determined.
The investigators expect the SPEQ to be a reliable, valid and responsive patient rated outcome assessment that can be used in the evaluation of the results of physiotherapy treatment.
