Analyses Statistical analyses were conducted using SPSS 14 for Windows

All participants provided written informed consent. Analyses Statistical analyses were conducted using Idelalisib SPSS 14 for Windows. The distribution of data was assessed regarding uni Fingolimod variate and multivariate normality and described and analyzed accord ingly. Group comparisons of rating scale response category interpretations Nonparametric multivariate analysis of variance was used to compare Idelalisib VAS values from PD and control respondents. If no significant differences between groups were identified, the pooled data was used to explore whether VAS values differed according to educational level, age, physical func tioning and mental health. For the PD group, differences in VAS values according to PD duration and whether patients experienced dyskinesias or not were also explored. To determine the best response options for the three types of ratings, the mean, standard deviations and 95% confidence intervals of the VAS values were examined. The criterion was that mean VAS values should be distributed equally across the 0 100 mm continuum, assuming the values of 0 and 100 for the predefined extreme anchor categories. This was done for three, four, five and six category response scales. For example, for a five category response scale, the three response categories with mean VAS values closest to 25, 50 and 75 mm were identified and each 95% CI was examined to determine if it covered the criterion value. For three, four and six category response scales the corresponding reference locations were 50 mm, 33 and 67 mm and 20, 40, 60 and 80 mm. In addi tion to roughly equal distances between mean locations, the 95% CIs for the VAS values of the selected response categories should not overlap. If two or more response categories met these criteria, the one with the smallest dispersion was selected. Finally, participants com ments were also taken into account when determining response category suitability. Sample characteristics are reported in Table 2.

There were no differences between people with PD and controls regarding age, educational levels or mental health scores, but there were more men in the PD sample, and controls had better physical functioning scores than people with PD. Data collection also took significantly lon ger for people with PD than for controls to com plete. In the PD group, 94% were treated with levodopa, 82% were on dopamine ago nists, COMT and MAO inhibitors were used among 43% each, and 25% had undergone a neurosurgical inter vention for their PD. The proportion of categories whose 95% CI covered the crite rion VAS values was highest for the six category agree ment scale, followed by the five category scales for all three types of ratings. Discussion This appears to be the first controlled study on the inter pretation of patient reported rating scale response cate gories in the clinical neurosciences. As such, it provides a first evidence base and initial guidance for selection of rating scale response categories when developing new or modifying available patient reported rating scales for PD. This is highly relevant as clarity, distinctiveness and equality of response category intervals represent funda mental assumptions underpinning traditional rating scale construction that are recognized by, e.