Background: Immune checkpoint inhibitors (ICIs) are known to cause immune-related adverse events. Patients with autoimmune diseases (AID) were excluded from most ICI clinical trials due to the potentially high risk of adverse effects. Data on the safety of ICIs in patients with a diagnosis of AID is therefore limited. Methods: A retrospective cohort study was conducted using a de-identified large oncology health care and pharmacy claims database with data from March 2010 until April 2017. Patients analyzed had a diagnosis of either melanoma or lung cancer and were treated with either of the anti-PD-1 inhibitors nivolumab or pembrolizumab. We assessed whether patients with AID compared with no AID were more likely to require medical interventions within 180 days of ICI therapy. We determined the percentage of patients receiving oral prednisone, IV methylprednisolone, or were hospitalized, which may represent responses to ICI toxicity. Results: 16.7% (16/96) of patients with either melanoma or lung cancer and AID received oral prednisone treatment within 180 days of ICI treatment, while 8.3% (131/1573) of patients without AID received oral prednisone during the same period. 8.4% (16/190) of patients with AID received IV methylprednisolone compared to 3.6% (79/2190) of patients without AID. Among melanoma patients, 24.1% (13/54) of patients with AID were hospitalized following ICI treatment, compared to 5.8% (28/480) of patients without ICI. Among lung cancer patients, 38.2% (52/136) of patients with AID were hospitalized compared to 31.6% (541/1711) of patients without AID. All comparisons are significant at p < 0.05 except hospitalizations in lung cancer patients. Conclusions: Patients with AID were more likely to receive interventions after ICI treatment that may represent responses to immune-related adverse events, suggesting that patients with AID are at increased risk for toxicity when being treated with ICIs.