BACKGROUND: In the concordance model, physician and patient discuss treatment options, explore the impact of treatment decisions from the patient's perspective, and make treatment choices together. We tested, in a concordance setting, whether the availability of AIR inhaled insulin (developed by Alkermes, Inc. [Cambridge, MA] and Eli Lilly and Company [Indianapolis, IN]; AIR is a registered trademark of Alkermes, Inc.), as compared with existing treatment options alone, leads to greater initiation and maintenance of insulin therapy and improves glycemic control in patients with type 2 diabetes. METHODS: This was a 9-month, multicenter, parallel, open-label study in adult, nonsmoking patients with diabetes not optimally controlled by two or more oral antihyperglycemic medications. Patients were randomized to the Standard Options group (n = 516), in which patients chose a regimen from drugs in each major treatment class excluding inhaled insulin, or the Standard Options + AIR insulin group (n = 505), in which patients had the same choices plus AIR insulin. The primary end points were the proportion of patients in each group using insulin at end point and change in hemoglobin A1C (A1C) from baseline to end point. RESULTS: At end point, 53% of patients in the Standard Options group and 59% in the Standard Options + AIR insulin group were using insulin (P = 0.07). Both groups reduced A1C by about 1.2% and reported increased well-being and treatment satisfaction. The most common adverse event with AIR insulin was transient cough. CONCLUSIONS: The opportunity to choose AIR insulin did not affect overall use of insulin at end point or A1C outcomes. Regardless of group assignment, utilizing a shared decision-making approach to treatment choices (concordance model), resulted in improved treatment satisfaction and A1C values at end point. Therefore, increasing patient involvement in treatment decisions may improve outcomes.