Several fixed-dose combination inhaled corticosteroid (FDC ICS/LABA) inhalers are licensed for chronic obstructive pulmonary disease (COPD) in the UK. This study compares effectiveness of Fostair 100/6 (BDP/FOR) metered dose inhalers (MDI) against other licensed FDC ICS/LABAs, namely; Seretide Accuhaler 500 (FP/SAL) dry powder inhaler (DPI) and Symbicort Turbohaler 200/6 and 400/24 DPI (BUD/FOR) in a patient group with recent exacerbations and poor lung function. A historical cohort study using data extracted from the Optimum Patient Care Research Database. Patients with diagnostic read codes for COPD, smoking history, age >=35 years, postdose FEV1 percent predicted lt;55>=1 previous long acting bronchodilator ever, and >=1 previous exacerbation in the 18 months prior initiation of FDC ICS/LABA therapy. The observation period consisted of one year after the initiation of FDC ICS/LABA for follow up and one year prior for patient characterisation. Patients were excluded if they switched or halted ICS/LABA treatment during the 1 year follow up period. The primary outcome was the proportion of patients with >=1 exacerbation. The non-inferiority margin was defined as 20 Patients were directly matched 1:1 on categorised age, smoking status, FEV1 percent predicted and exacerbations. Mean total costs consisting of medication/resource costs were compared between treatment arms after adjusting for confounders. Out of the patients prescribed BDP/FOR (n=537) and 537 FP/SAL (n=537), the median age was 70 and 69 respectively and 41.7% were current smokers. In the BDP/FOR (n=540) and BUD/FOR (n=540) comparison, the median age was 70 and 69 respectively, and 42% were current smokers. The risk of >=1 exacerbation in BDP/FOR group was non-inferior during the year following the initiation of ICS/LABA compared to FP/SAL (OR 0.89, 95% CI: 0.67textendash1.19) or BUD/FOR (OR 0.79, 95% CI: 0.58textendash1.08). Cost was significantly lower for BDP/FOR versus FP/SAL (adjusted mean textsterling730.0 versus textsterling850.1 respectively, plt;0.001) and lower for BDP/FOR versus BUD/FOR (adjusted mean textsterling732.4 versus textsterling757.2 respectively, p=0.054). Treatment with BDP/FOR is non-inferior in terms of exacerbation risk and is additionally associated with a lower point estimate of exacerbation risk, a lower cost compared to FP/SAL and BUD/FOR, and a lower ICS dose compared to FP/SAL.