Participation in regular physical activity after total knee or hip arthroplasty for osteoarthritis: prevalence, associated factors, and type
Arnaldo Libman
Original Article
Participation in regular physical activity after total knee or hip arthroplasty for osteoarthritis: prevalence, associated factors, and type
Abstract
Objective
This study aimed to describe the: rates of participation in regular physical activity (PA) pre‐ and 3‐years following knee or hip arthroplasty; factors associated with participation post‐surgery; and types of activity undertaken.
Methods
A previously acquired, multicentre, prospective cohort of knee or hip arthroplasty recipients was followed‐up 3‐years post‐surgery. Regular participation in PA was defined as participation in PA ≥1 times/week, excluding incidental activities. Participants were interviewed about current participation as well as participation in the year pre‐surgery. Joint‐specific and health‐related quality of life scores and experience of major complications were obtained. Co‐morbid information and body weight were updated. Factors associated with 3‐year PA participation were determined using multivariable logistic regression modelling.
Results
74% (1289/1773) of the eligible cohort were followed‐up (718 knee, 571 hip). Participation profiles were similar regardless of joint replaced. Participation in PA increased post‐surgery [combined cohort, 45.2 to 63.5%, P<0.001]. Participation at 3‐years was associated with participation pre‐surgery (P<0.0001), better 3‐year quality of life (P<0.001), younger age (P=0.002), better 3‐year joint scores (P= 0.01), > 1 lifetime arthroplasty (P=0.02), and higher education (P=0.04). Low‐impact and non‐ambulatory activities significantly increased post‐surgery with no change in high‐impact activities.
Conclusion
Participation rates increase post‐surgery when recovery is stable, but approximately one‐third of arthroplasty recipients do not engage in PA at least once per week. As participation is associated with habitual activity pre‐surgery, a potential role for behaviour change interventions is suggested. That non‐ambulatory activities increase indicates current devices measuring ambulatory activity alone are inadequate for capturing PA.