Predictors of adherence to a step count intervention following total knee replacement: An exploratory cohort study

Objective: To explore the person-level predictors of adherence to a step count intervention following total knee replacement (TKR). • DESIGN: Prospective cohort study, nested within the PATHway trial.

Methods: Participants who had recently undergone TKR were recruited from 3 rehabilitation hospitals in Sydney, Australia, for the main trial. Only data from participants who were randomized to the TKR intervention group were analyzed. Participants in the intervention group (n = 51) received a wearable tracker to monitor the number of steps taken per day. Step count adherence was objectively measured at 3 months as the number of steps completed divided by the number prescribed and multiplied by 100 to express adherence as a percentage. Participants were classified into 4 groups: withdrawal, low adherence (0%-79%), adherent (80%-100%), and >100% adherent. Ordinal logistic regression was used to identify which factors predicted adherence to the prescribed step count.

Results: Of the 51 participants enrolled, nine (18% of 51) withdrew from the study before 3 months. Half of participants were classified as >100% adherent (n = 24%, 47%). Ten were classified as low adherence (20%), and 8 participants were classified as adherent (16%). In the univariable model, lower age (OR 0.90; 95% CI 0.83-0.97), higher patient activation (OR 1.03; 95% CI 1.00-1.06), and higher technology self-efficacy (OR 1.03; 95% CI 1.00-1.06) were associated with higher adherence. After adjusting for age in the multivariable model, patient activation and technology self-efficacy were not significant.

Conclusion: Younger age, higher patient activation, and higher technology self-efficacy were associated with higher adherence to a step count intervention following TKR in the univariable model. Patient activation and technology self-efficacy were not associated with higher adherence following adjustment for age.