This study examined whether treatment of Crohn's disease (CD) and ulcerative colitis (UC) with immunosuppressant medications was associated with an increased risk of death in the era prior to antitumor necrosis factor (TNF) therapies. This retrospective cohort study used data from the General Practice Research Database from 1987 to 1997. CD and UC patients were matched to controls on age, sex, and primary care practice. CD and UC patients were stratified according to whether they used immunosuppressant medications during follow-up. Cox proportional hazards models adjusted for comorbidities were used to define the relative hazard of death. Additional models examined the relative hazard of death with current use of corticosteroids or thiopurines. The cohort included 5,539 patients with CD, 8,910 patients with UC, and 41,624 controls. Patients with CD had an increased mortality (not immunosuppressant-treated CD hazard ratio [HR] 1.27, 95% confidence interval [CI] 1.07-1.51; immunosuppressant-treated CD HR 2.44, 95% CI 1.84-3.25). Increased mortality was only observed among UC patients treated with immunosuppressant medications (HR 1.67, 95% CI 1.34-2.09). In both CD and UC, current corticosteroid therapy was associated with increased mortality (CD HR 2.48, 95% CI 1.85-3.31; UC HR 2.81, 95% CI 2.26-3.50). Current use of thiopurines was not associated with increased mortality (CD HR 0.83, 95% CI 0.37-1.86; UC HR 0.70, 95% CI 0.29-1.70). Patients treated with corticosteroids, but not thiopurines, are at increased risk of death, although this study could not clarify whether this was as a result of the medication or the underlying disease severity.