ABSTRACT: Importance:While observational studies show that physical inactivity is associated with worse outcomes in chronic obstructive pulmonary disease (COPD), there are no population-based trials to date testing the effectiveness of physical activity (PA) interventions to reduce acute care use or improve survival. Objective:To evaluate the long-term effectiveness of a community-based PA coaching intervention in patients with COPD. Design, Setting, and Participants:Pragmatic randomized clinical trial with preconsent randomization to the 12-month Walk On! (WO) intervention or standard care (SC). Enrollment occurred from July 1, 2015, to July 31, 2017; follow-up ended in July 2018. The setting was Kaiser Permanente Southern California sites. Participants were patients 40 years or older who had any COPD-related acute care use in the previous 12 months; only patients assigned to WO were approached for consent to participate in intervention activities. Interventions:The WO intervention included collaborative monitoring of PA step counts, semiautomated step goal recommendations, individualized reinforcement, and peer/family support. Standard COPD care could include referrals to pulmonary rehabilitation. Main Outcomes and Measures:The primary outcome was a composite binary measure of all-cause hospitalizations, observation stays, emergency department visits, and death using adjusted logistic regression in the 12 months after randomization. Secondary outcomes included self-reported PA, COPD-related acute care use, symptoms, quality of life, and cardiometabolic markers. Results:All 2707 eligible patients (baseline mean [SD] age, 72 [10] years; 53.7% female; 74.3% of white race/ethnicity; and baseline mean [SD] percent forced expiratory volume in the first second of expiration predicted, 61.0 [22.5]) were randomly assigned to WO (n?=?1358) or SC (n?=?1349). The intent-to-treat analysis showed no differences between WO and SC on the primary all-cause composite outcome (odds ratio [OR], 1.09; 95% CI, 0.92-1.28; P?=?.33) or in the individual outcomes. Prespecified, as-treated analyses compared outcomes between all SC and 321 WO patients who participated in any intervention activities (23.6% [321 of 1358] uptake). The as-treated, propensity score-weighted model showed nonsignificant positive estimates in favor of WO participants compared with SC on all-cause hospitalizations (OR, 0.84; 95% CI, 0.65-1.10; P?=?.21) and death (OR, 0.62; 95% CI, 0.35-1.11; P?=?.11). More WO participants reported engaging in PA compared with SC (47.4% [152 of 321] vs 30.7% [414 of 1349]; P?