We investigated the incidence of type II diabetes mellitus (T2DM) among people with COPD and whether exposure to inhaled corticosteroid (ICS) and exacerbation status was associated with T2DM. This descriptive cohort study used primary care data from the Clinical Practice Research Datalink (CPRD). The patient cohort included people with a diagnosis of COPD and previous smoking history registered at a CPRD practice between January 2010 and December 2016. We determined incidence rates by age, gender and deprivation. Using a nested case–control design—where cases and controls are drawn from the cohort population—we matched 1:5 with patients by age, gender and GP practice and estimated odds of T2DM using logistic regression (adjusting for smoking status, deprivation, BMI, hypertension, coronary heart disease and heart failure). We identified 220,971 COPD patients; mean age at COPD diagnosis was 66 years (SD 12) and 54% were male. The incidence rate of T2DM in COPD patients was 1.26 per 100 patient years (95% CI: 1.24–1.28) and was higher among men (1.32 vs 1.18 among women). The adjusted odds ratio for T2DM was 1.47 (95% CI: 1.36–1.60) among frequent exacerbators (≥2 treated exacerbations per year) compared to infrequent exacerbators and the odds ratio for patients receiving high-dose ICS (>800 mcg budesonide equivalent dose) was 1.73 (95% CI 1.65–1.82) compared to patients receiving no ICS therapy. Incidence of T2DM among COPD patients is high and exposure to ICS and frequent exacerbations are associated with a higher risk of T2DM among patients with COPD.