OBJECTIVE: To examine the impact of a nurse-coordinated intervention delivered to patients with non-insulin-dependent diabetes mellitus between office visits to primary care physicians. DESIGN: Randomized, controlled trial. SETTING: Veterans Affairs general medical clinic. PATIENTS: 275 veterans who had NIDDM and were receiving primary care from general internists. INTERVENTION: Nurse-initiated contacts were made by telephone at least monthly to provide patient education (with special emphasis on regimens and significant signs and symptoms of hyperglycemia and hypoglycemia), reinforce compliance with regimens, monitor patients' health status, facilitate resolution of identified problems, and facilitate access to primary care. MEASUREMENTS: Glycemic control was assessed using glycosylated hemoglobin (GHb) and fasting blood sugar (FBS) levels. Health-related quality of life (HRQOL) was measured with the Medical Outcomes Study SF-36, and diabetes-related symptoms were assessed using patients' self-reports of signs and symptoms of hyper- and hypoglycemia during the previous month. MAIN RESULTS: At one year, between-group differences favored intervention patients for FBS (174.1 mg/dL vs 193.1 mg/dL, p = 0.011) and GHb (10.5% vs 11.1%, p = 0.046). Statistically significant differences were not observed for either SF-36 scores (p = 0.66) or diabetes-related symptoms (p = 0.23). CONCLUSIONS: The intervention, designed to be a pragmatic, low-intensity adjunct to care delivered by physicians, modestly improved glycemic control but not HRQOL or diabetes-related symptoms.