Living with chronic illness scale in Parkinson's disease: Longitudinal metric properties and meaningful change

Abstract

AimTo analyze the responsiveness and interpretability of the Living with Chronic Illness Scale in patients with Parkinson's disease (LW-CI-PD). MethodsLongitudinal, international study, with a convenience sample of 153 PD Spanish and Latin-American patients assessed at baseline and one year later. The LW-CI-PD and other clinical measures were applied. For responsiveness, Wilcoxon-Mann-Whitney test of differences, correlation of change between rating scales, standard error of difference, relative change, Cohen's effect size and standardized response mean of LW-CI-PD were computed. The minimally clinical important difference was calculated using anchor- (applying the Patient Global Impression of Severity) and distribution-based methods. A triangulation of interpretability indexes was performed to determine the range of the minimally clinical important difference values. ResultsThe LW-CI-PD scored 65.7 (11.7, range: 33–101) at baseline, and 68.6 (10.3, range: 33–102) one year later (p < 0.001). Change in LW-CI-PD correlated −0.26 with change in psychosocial status, 0.18 with change in motor function and −0.15 with change in social support. Responsiveness statistics were: relative change = 4.5%; effect size = 0.25; standardized response mean = 0.46. Using PGI-S as anchor, 29 patients worsened, and the value of minimally clinical important difference for worsening in LW-CI-PD total score was 4.7. Minimally clinical important difference values using distribution-based methods were between 4.5 (1 standard error of measurement) and 10.4 (10% of total score), with a mean of 6.9. ConclusionsOur study suggest the LW-CI-PD is responsive to changes over time. The use of different methods for calculating the minimally clinical important difference allows to determine a range of the real change for the LW-CI-PD.