132 Chapter 6 Akaike Information Criterion, Bayesian Information Criterion, and entropy values of the different models were compared (39). When less than 5% of the sample was assigned to a class, the model with k-1 classes was chosen instead to maintain the clinical usefulness of the final model. To test for differences in baseline characteristics between participants based on class membership from the LCGA, chi-square tests were used for categorical variables. The Kruskal–Wallis test was used for continuous variables due to the nonnormal distribution of the data. Role of the Funding Source The funder played no role in the design, conduct, or reporting of this study. RESULTS Data on adherence and received treatment during the study were available for 191 of the 208 participants. The unavailability of adherence data for a patient was caused by physical therapists not using the case report form properly during treatment or not returning the case report form to the researchers after the treatment ended. Eighteen patients received fewer than 3 treatments, leaving data from 173 participants avail- able for analysis. Data on patient adherence were collected during 5.1 (SD 2.5) treatment sessions and total treatment duration lasted for 51 (SD 41.7) days. Baseline characteristics of included patients can be found in Table 1. Results from the LCGA are limited to 11 treatment sessions (10 timepoints), because only 1 patient received more than 11 treatment sessions. Table 1 Baseline demographic and clinical characteristics for all patientsa Patient characteristics n = 173 Sex (female), n (%) 88 (50.9) Age (y), median (IQR) 48 (24.3) Height (cm), median (IQR) 175 (12) Weight (kg), median (IQR) 80 (20) BMI (kg/m2), median (IQR) 25.5 (4.7) Educational level, n (%) Low 30 (17.3) Middle 60 (34.7) High 83 (48.0) Central sensitization (score = 0-100), median (IQR) 27 (18)
RkJQdWJsaXNoZXIy MTk4NDMw