Patients eligible for total knee arthroplasty: Associations between psychological factors and patients’ decisions to participate in a randomized controlled trial that includes a non-surgical treatment group A comparative study

Background: We know little about psychological factors´ influencing patients´ decisions to participate in randomized controlled trials (RCTs) and the differences between those who participate and those who decline. Objective: To compare levels of pain, other symptoms, knee-related quality of life (QoL), knee awareness and psychological factors (anxiety, depression, pain-related catastrophizing and pain-related fear of movement) in two groups of patients with knee osteoarthritis deemed eligible for total knee arthroplasty – those who agreed to participate in a randomized controlled trial including education, exercise therapy and internet-delivered cognitive therapy, with a one-third chance of postponing the surgery by one year, and those who declined participation. Methods / Study design: In this comparative study, anxiety and depression (Hospital Anxiety and Depression Scale (HADS)), pain-related fear of movement (Fear-Avoidance Belief Questionnaire (FABQ)), pain-related catastrophizing (Pain Catastrophizing Scale (PCS)), knee pain and other symptoms (Knee Osteoarthritis and Outcome Scale (KOOS)), and knee awareness (Forgotten Joint Score (FJS)), were measured in patients included in the MultiKnee trial (N=280) and in patients who declined to participate in the trial but were willing to complete baseline questionnaires (N=373); reasons for declining were also recorded. Simple and multiple logistic regression models were used to explore the associations between psychological and clinical characteristics and the patients´ willingness to participate in the randomized controlled trial. Results: In adjusted analyses, each 1-point increase in HADS Anxiety score was associated with a 5.4% increase in the odds of participating in the RCT (OR=1.054, 95%CI 1.001, 1.110). Each 1-point increase in FABQ Physical activity score was associated with a 3.0% decrease in the odds of participating in the RCT (OR=0.970, 95%CI 0.944, 0.996). HADS Depression and PCS were not significantly associated with RCT participation. Compared to patients who participated in the RCT, those who declined had more fear-avoidance (FABQ 13.3 versus 11.7, p<0.001), more knee awareness (FJS 14.5 versus 18.5, p<0.001), and worse scores on all KOOS subscales: KOOS-Symptoms (47.3 vs 50.7, p=0.018), KOOS-Pain (42.8 vs 48.5, p<0.001), KOOS-ADL (53.6 vs 58.0, p=0.002), KOOS-Sport & Recreation (14.6 vs 18.8, p<0.001), and KOOS-QoL (26.1 vs 30.7, p<0.001). Conclusions: Patients with higher anxiety levels were more willing to participate in the RCT, while patients with pain related fear-avoidance of physical activity were less willing. These findings might provide a selection bias that weakens generalizability of RCT findings. Trial registration: The MultiKnee trial was approved (May ^9th 2017) by the Ethics Committee of South-Eastern Norway Regional Health Authority (2017/968) and registered in ClinicalTrials.gov (NCT03771430).