CONTEXT: Mortality data in patients with adrenal insufficiency are inconsistent, possibly due to temporal and geographical differences between patients and their reference populations. OBJECTIVE: To compare mortality risk and causes of death in adrenal insufficiency with an individually-matched reference population. DESIGN: Retrospective cohort study. SETTING: UK general practitioner database (CPRD). PARTICIPANTS: 6821 patients with adrenal insufficiency (primary, 2052; secondary, 3948) and 67564 individually-matched controls (primary, 20366; secondary, 39134). MAIN OUTCOME MEASURES: All-cause and cause-specific mortality; hospital admission from adrenal crisis. RESULTS: With follow-up of 40799 and 406899 person-years for patients and controls respectively, the hazard ratio (HR; [95%CI]) for all-cause mortality was 1.68 [1.58 - 1.77]. HRs were greater in primary (1.83 [1.66 - 2.02]) than in secondary (1.52 [1.40 - 1.64]) disease; (HR; primary versus secondary disease, 1.16 [1.03 - 1.30]). The leading cause of death was cardiovascular disease (HR 1.54 [1.32-1.80]), along with malignant neoplasms and respiratory disease. Deaths from infection were also relatively high (HR 4.00 [2.15 - 7.46]). Adrenal crisis contributed to 10% of all deaths. In the first two years following diagnosis, the patients' mortality rate and hospitalisation from adrenal crisis were higher than in later years. CONCLUSION: Mortality was increased in adrenal insufficiency, especially primary, even with individual matching and was observed early in the disease course. Cardiovascular disease was the major cause but mortality from infection was also high. Adrenal crisis was a common contributor. Early education for prompt treatment of infections and avoidance of adrenal crisis hold potential to reduce mortality.