Treatment Fidelity in a Multi-Centre Study of Early Physical Rehabilitation in Mechanically Ventilated Patients: A Secondary Analysis of the CYCLE Pilot RCT

Purpose: To determine the relationship between patient characteristics and treatment fidelity in adult ICU patients enrolled in the CYCLE Pilot randomized controlled trial. Method: We conducted a secondary analysis of a pilot trial in which patients were randomly assigned to in-bed cycling plus usual physiotherapy (PT) or usual PT alone. We calculated protocol fidelity for each patient as the proportion of planned days when the assigned treatment was delivered or did not occur due to medical instability. Using multiple linear regression, we assessed the association of protocol fidelity with admission severity of illness, sex, assigned treatment, baseline comorbidities, and frailty. Results: Sixty-six patients were enrolled in this pilot trial (36 cycling, 30 usual PT) with a median (1st–3rd quartiles) protocol fidelity of 100% (83%–100%), 89% (75%–100%) in cycling and 100% (90%–100%) in usual PT. By regression analysis, assignment to in-bed cycling plus usual PT was associated with a 9.7% (95% CI: –17.8, –1.5) lower protocol fidelity compared to usual PT alone. Conclusions: We found those in the usual PT group had higher protocol fidelity; no other associations were found. This study offers an analytic approach for trialists to assess treatment fidelity for rehabilitation interventions in clinical trials.