A Hybrid Effectiveness/Implementation Clinical Trial of Adherence to Long-Term Oxygen Therapy for Chronic Obstructive Pulmonary Disease

Interventions to promote adherence to long-term oxygen therapy (LTOT) in chronic obstructive pulmonary disease (COPD) are needed. To examine the real-world effectiveness of phone-based peer coaching on LTOT adherence and other outcomes in a pragmatic trial of patients with COPD. In a hybrid effectiveness/implementation pragmatic trial, patients were randomized to receive phone-based proactive coaching (educational materials, five phone-based peer coaching sessions over 60 d), reactive coaching (educational materials, peer coaching when requested), or usual care. Study staff members collected baseline and outcome data via phone at 30, 60, and 90 days after randomization. Adherence to LTOT over 60 days, the primary effectiveness outcome, was defined as mean LTOT use ⩾17.7 h/d. LTOT use was calculated using information about home oxygen equipment use in worksheets completed by study participants. Comparisons of adherence to LTOT between each coaching group and the usual care group using multivariable logistic regression models were prespecified as the primary analyses. Secondary effectiveness outcomes included Patient Reported Outcome Management Information System measures for physical, emotional, and social health. We assessed early implementation domains in the reach, adoption, and implementation framework. In 444 participants, the proportions who were adherent to LTOT at 60 days were 74% in usual care, 84% in reactive coaching, and 70% in proactive coaching groups. Although reach, adoption by stakeholder partners, and intervention fidelity were acceptable, complete LTOT adherence data were available in only 73% of participants. Reactive coaching (adjusted odds ratio, 1.77; 97.5% confidence interval, 0.80-3.90) and proactive coaching (adjusted odds ratio, 0.70; 97.5% confidence interval, 0.34-1.46) did not improve adherence to LTOT compared with usual care. However, proactive coaching significantly reduced depressive symptoms and sleep disturbance compared with usual care and reduced depressive symptoms compared with reactive coaching. Unexpectedly, LTOT adherence was significantly lower in the proactive compared with the reactive coaching group. The results were inconclusive about whether a phone-based peer coaching strategy changed LTOT adherence compared with usual care. Further studies are needed to confirm the potential benefits of proactive peer coaching on secondary effectiveness outcomes and differences in LTOT adherence between proactive and reacti