To date, it remains unknown which patients report a clinically-relevant improvement in fatigue following pulmonary rehabilitation (PR). The purpose of this study was to identify and characterize these responders. Demographics, lung function, anxiety (anxiety subscale of the 90-item symptom checklist (SCL-90-A)), depression (Beck depression inventory for primary care (BDI-PC)), exercise tolerance (six-minute walking distance test (6MWD)), and health status (Nijmegen clinical screening instrument (NCSI)) were assessed before and after a 12-week PR programme. Fatigue was assessed using the checklist individual strength (CIS)-Fatigue. Patients with a decline ≥ 10 points (minimally clinically important difference, MCID) on the CIS-Fatigue were defined as responders. Chronic obstructive pulmonary disease (COPD) patients (n = 446, 61 ± 9 years, 53% male, forced expiratory volume in 1 s (FEV1) 43% ± 18% predicted, 75% severe fatigue) were included. Mean change in fatigue after PR was 10 ± 12 points (p < 0.01) and exceeded the MCID. In total, 56% were identified as fatigue responders. Baseline CIS-Fatigue (45 ± 7 vs. 38 ± 9 points, respectively, p < 0.001) and health-related quality-of-life (HRQoL; p < 0.001) were different between responders and non-responders. No differences were found in demographics, baseline anxiety, depression, lung function, 6MWD, and dyspnoea (p-values > 0.01). Responders on fatigue reported a greater improvement in anxiety, depression, 6MWD, dyspnoea (all p-values < 0.001), and most health status parameters. PR reduces fatigue in COPD. Responders on fatigue have worse fatigue and HRQoL scores at baseline, and are also likely to be responders on other outcomes of PR.