Prevalence of diabetes in the elderly increases, and half of the French diabetics are over the age of 75 years. The GERODIAB study is the first French multicentre, prospective, observational study designed to analyse over 5 years the influence of glycaemic control on morbidity-mortality in type 2 diabetics patients 70 years old and over. This study analysed the diabetic and geriatric factors associated with the treatment modalities, particularly insulin, at inclusion in the cohort. The cohort of 987 type 2 diabetics was divided into three groups according to the method of treatment. Slightly fewer than one-third of these patients (26.4%) were treated with insulin alone, 31% received insulin and oral antidiabetic drugs, and 42.7% oral antidiabetic drugs alone. The patients that received insulin alone were significantly older, had poorer glycaemic control (HbA1c = 7.9 ± 1.4, 7.8 ± 1.0 and 7.1 ± 1.2%, respectively; P < 0.001) and had greater alterations of glomerular filtration rate (GFR). HbA1c was below 6.5% in 15% of patients and 37.3% of patients had a GFR below 60 mL/min. The patients treated with insulin alone had significantly more hypoglycaemic episodes (respectively 53.3, 36.3 and 19.5%, P < 0.001), retinopathy, cardiovascular involvement and more specific geriatric complications, such as cognitive disorders (respectively 34.1, 31.4 and 23.6%, P = 0.006). In this specific population of elderly type 2 diabetic patients, diabetic and geriatric conditions significantly differed between the types of drug treatments. Considering low values of HbA1c and GFR, some patients seemed overtreated and other patients received inappropriate drugs.