Background: There has been a trend towards better management of glucose, lipids, BP, smoking, and other aspects of diabetes care in the last decade. The goals of this study are to assess changes in quality of diabetes care over a 14-month period of time in a large multi-specialty U.S. medical group, to quantify treatment costs associated with the improved care, and to estimate the impact of care improvement on long-term costs and quality adjusted life expectancy. Methods: : Study subjects included 7,054 persons with diabetes age 40-75 years who at baseline had one or more of: SBP > 140 mmHg, LDL-cholesterol > 129 mg/dl, or current smoking. We quantified their clinical status including A1c, BP, lipids, and smoking status both at baseline and after a median 14-month follow-up period. We similarly quantified their visit frequency and medication use and associated costs in the year prior and year following their baseline visit. We employed these clinical risk factors and observed costs as data inputs into a log-term simulation model of diabetes outcomes - the UKPDS Outcomes Model (Version 2) – in order to estimate changes in quality-adjusted life years (QALYs), and costs associated with changes in clinical care, projected over a 40-year time period. We applied costs of complications that were derived from a previous study within this health system. We then estimated the cost per QALY gain for these adult diabetes subjects who are experiencing better clinical care over time. Results: Observed improvements in clinical care significantly increased expected QALY from 10.83 to 11.06, for a gain in 0.22 QALY. Incremental costs associated with outpatient visits and intensification of pharmacotherapy were $167 per year and $2,323 over the study period. Total costs increased by $4,453. Cost per QALY was estimated to be $19,866. Sensitivity analysis indicated that estimates of cost per QALY were more favorable in simulations with longer follow up periods and in simulations that more narrowly targeted BP control among those with high BP at baseline. Conclusion: Observed improvements in diabetes care over a recent 14-month period of time are sufficient to significantly improve clinical and health outcomes. The cost- effectiveness of the slightly more intensive diabetes care provided appears to be satisfactory using standard thresholds for cost per QALY, both in the base case and across a range of sensitivity analysis scenarios. However, improvements in diabetes care are not cost saving from the point of view of the payer.