The management of Type 2 diabetes mellitus with currently available oral agents may be complicated in the elderly by an increased frequency of side-effects. The effects of troglitazone, an insulin action enhancer, were studied in elderly patients with Type 2 diabetes in a double-blind, parallel-group, placebo-controlled trial. A total of 229 patients (41% male), mean age 75 (range 69-85) years, with two fasting capillary blood glucose values > or =7 and < or =15 mmol l(-1) (and within 4.0 mmol l(-1) of each other) and previously treated with either diet alone (30%) or oral hypoglycaemic agents, were randomized to placebo or troglitazone 400 mg once daily or 200 mg twice daily, or 800 mg once daily or 400 mg twice daily, for 12 weeks. After 12 weeks' treatment, fasting serum glucose was significantly lower in troglitazone-treated patients (troglitazone, adjusted geometric mean 9.4-10.4 mmol l(-1) vs placebo 12.7 mmol l(-1), p < 0.001). Adjusted geometric mean fructosamine was also lower in troglitazone-treated patients by 5 to 15% compared to placebo (P < 0.05 at all doses except 400 mg od). There was no significant difference between troglitazone doses for improvement in glycaemic control. Troglitazone lowered adjusted geometric mean fasting plasma insulin by 27-34% compared to placebo (P < 0.001) and insulin sensitivity (HOMA-S) improved by 9-15% in all troglitazone dose groups (p < 0.001). Troglitazone also lowered serum non-esterified fatty acids and triglyceride. Adverse event incidence in troglitazone-treated patients was similar to that in patients treated with placebo. No weight gain or symptomatic hypoglycaemia was recorded at any of the doses studied. Troglitazone is effective and well tolerated in elderly patients with Type 2 diabetes mellitus, providing improved glycaemic control in the absence of weight gain.