Abstract
Introduction
Patient satisfaction becomes an important aspect in clinical practice causing a shift from clinician-administered scales (CAS) towards patient-administered measurement outcomes (PROMs). Besides, clinical outcome can objectively be evaluated using inertia-based motion analysis (IMA). This study evaluates different outcome measures by investigating the 1) effect of replacing CAS by PROMS on outcome assessment, 2) redundancy between scales, 3) additional value of IMA in outcome scoring.
Methods
This cross-sectional study included 27 primary unilateral total knee arthroplasty patients (m/f=12/19; age=66.2 yrs), 6 weeks (n=12) and 6 months (n=15) postoperative, who covered a wide range of the scores. One CAS (Knee Society Score (KSS; knee and function subscore), two PROMs (Knee Injury and Osteoarthritis Outcome Score Physical Shortform (KOOS-PS), Visual Analogue Scale satisfaction (VAS)) and a functional test (IMA block step test) were completed. For IMA, patients stepped up and down a 20cm block starting with the affected and followed by the non-affected leg, while wearing an inertia-sensor (3D accelero- and gyrometer) at the lower back (fig. 1). IMA-parameters like performance time (s), bending angle (°), pelvic-obliquity angle (°), were calculated using self-designed algorithms. Differences between legs were determined by ratios (affected/non-affected leg). Pearson's correlations were done, considering r<0.4 poor, 0.4<r<0.7 moderate, r>0.7 strong.
Results
KSS-subscores and KOOS-PS encountered a broad range of the total scale (e.g. KSS-function [40-100]), while VAS-satisfaction range was limited [0-3]. Most questionnaires were moderately intercorrelated (r-range 0.4–0.6). Correlations were lacking between VAS-satisfaction and KSS-subscores. The KSS-function correlated most with IMA-parameters (r-range 0.4–0.5). VAS-satisfaction and KOOS-PS correlated only with one IMA parameters (resp. pelvic-obliquity ratio, time-to-perform). Correlations were lacking between KSS-knee and IMA (table 1). For all correlations applies that a better outcome in one score was associated with a better outcome in the other score.
Discussion
The correlations between KOOS-PS and KSS-subscores indicate that they capture similar aspects of function, showing redundancy. VAS-satisfaction correlated with KOOS-PS but not with KSS-subscores, showing that KOOS-PS captures some satisfactory dimensions, which are lacking with KSS. The strongest correlation with VAS-satisfaction was found with IMA pelvic-obliquity ratio, a measure showing asymmetry in unilateral pathologies, indicating that satisfaction is best captured by IMA. Most correlations with IMA-parameters were found for KSS-function showing that KSS-function, which is the PROM like part of the KSS, is the most objective questionnaire-based measure. The KSS-knee lacked any correlation with IMA showing that clinician-based measurements are not so relevant to patients and not related to objective measures either. Also the KOOS-PS lacks objective aspects of function as was shown by the limited amount of correlations with IMA. This may be due to the lack of stair climbing assessment in the KOOS-PS in contrary to KSS-function.
Conclusions
The shift from CAS to PROMs may result in a loss of (objective) information, but will add a satisfaction aspect. Improvements in PROMS (e.g. add moderate-demanding activities like stair climbing) are therefore recommended. The use of IMA may be an alternative as it provides an objective assessment capturing satisfaction and PROMs-like (KSS-function) aspects. Thereby, CAS will be improved (e.g. new KSS) which may be promising as well.
