Objectives: To describe hospital stays and the characteristics of hospitalized patients with diabetes, and to identify factors associated with these hospitalizations. Methods: In 2007, 8,926 adults with diabetes who had received reimbursement for oral antidiabetic medications and/or insulin at least three times over the past 12 months were randomly selected from health insurance data. Medical records were available for all participants; a patient questionnaire was available for 48% (N=4,277) of the patients, and a physician questionnaire for 28%. Hospital stays for individuals who did not refuse to participate in the study (N=7,534, 84%) were extracted from the PMSI between August 2006 and July 2009. Determinants of inpatient admissions (24-hour stays) were analyzed for hospital stays recorded between August 2008 and July 2009. Results: Nearly one-third (31%) of people with diabetes (type 1: 45%; type 2: 31%, p<0.0001) had at least one hospital stay during the year: 13% for stays of less than 24 hours (type 1: 23%; type 2: 13%, p<0.0001), 24% for inpatient stays (type 1: 31%; type 2: 24%, p<0.0001). Patients admitted for inpatient care were older than other patients (median age, 69 years vs. 65 years), were more often covered 100% for a chronic condition (91%), reported financial difficulties more often (59%), had long-standing diabetes (10+ years, 54%), and experienced complications. They sought medical care and received insulin treatment (29%) more frequently. They had an average of 1.6 hospital stays, amounting to a cumulative 11 days of hospitalization per person. In multivariate analysis, advanced age, financial difficulties, a history of microvascular or coronary complications, inadequate glycemic control, and treatment with insulin alone were independently associated with the need for inpatient hospitalization among people with type 2 diabetes. Conclusion: Hospitalizations remain common among people with diabetes, particularly among the elderly, frail, and disadvantaged. It is therefore essential to strengthen secondary prevention measures for these individuals. (R.A.)