Background: Following National Institute for Clinical Excellence approval of inhaled insulin Exubera (R) (Pfizer, New York, NY) in 2006, we established a dedicated clinic in January 2007 to monitor the efficacy and safety of Exubera. Between January and October 2007, eight patients started Exubera: six because of needle phobia (DSM-IV criteria) and two with injection site problems. Methods: Data were collected at the clinic over a 12-month period from February 2007 at 3-, 6-, 9-, and 12-month intervals. The clinic is jointly led by a consultant diabetologist and a diabetes specialist nurse within the secondary care setting. Results: Inhaled insulin was well tolerated in all eight patients who had previously experienced significant problems with initiation or intensification of subcutaneous insulin injections. Mean hemoglobin A1c was 10.7% (range, 8.1-14.2%) at initaition, 8.3% (7.2-9.4%) at 3 months, 7.7% (6.9-9.0%) at 6 months, 7.4% (6.7-8.4%) at 9 months, and 7.5% (6.5-8.7%) at 12 months. At 6 months, six patients had a reduction in forced expiratory volume in the first second (FEV1) by 4-12%, whereas five patients had a reduction of 2-12% at 12 months. One developed dyspnea, with a 29% fall in FEV1, which was transient and secondary to an upper respiratory tract infection. Two patients with the highest starting and most improved hemoglobin A1c developed significant retinopathy. Conclusions: Our 12-month audit data demonstrate that the initiation of inhaled insulin in this difficult-to-treat group of patients resulted in a significant improvement in glycemic control. The subsequent withdrawal of an alternative and acceptable form of insulin treatment now presents a renewed challenge for our patients and healthcare professionals.