Background: SLS COPD was the first open-label randomised controlled trial demonstrating a reduction in moderate/severe COPD exacerbations with once-daily inhaled fluticasone furoate/vilanterol (FF/VI) in everyday clinical practice. Here we report FF/VI effectiveness and safety in predefined patient subgroups. Methods: Patients with COPD, exacerbation history, and receiving maintenance inhaler therapy, were randomised to initiate FF/VI 100/25 μg or continue usual care (UC) with 12 months’ follow-up. Annual rates of moderate/severe exacerbations (primary outcome), selected secondary outcomes, and incidence of pneumonia serious adverse events of special interest (SAESI) were compared between randomisation groups across various patient subgroups/baseline treatment strata. SAESI rates by actual treatment were also assessed. Results: Lower exacerbation rates were observed for FF/VI versus UC across all subgroups/strata, including ICS + LABA therapy subset (8.0% [0.1, 15.4]), except in patients without baseline airflow limitation (−0.5% [–29.8, 22.1]). Larger reductions compared to the overall analysis were observed for patients on ICS-containing regimens (excluding LAMA) before the study (15.6% [3.4, 26.3]), and with baseline CAT score <10 (25.3% [–0.4, 44.4]). Pneumonia SAESI rates were similar for FF/VI versus UC across all subgroups/strata, except the LABA, LAMA or LABA + LAMA stratum (incidence ratio 2.8 [0.9, 8.5]). SAESI rates were not increased for FF/VI versus other ICS + LABA. Conclusions: Initiating FF/VI versus continuing UC reduced exacerbation rates without increased pneumonia SAESI risk compared to other ICS-containing regimens and in various patient subgroups, consistent with primary study findings. FF/VI may be a therapeutic option for a broad population of COPD patients, including those with more severe disease.