A systematic review was conducted to evaluate the effectiveness of initiating insulin treatment with insulin glargine or insulin detemir for type 2 diabetes in a routine clinical practice setting. Medline, EMBASE and Cochrane literature databases were searched to identify published "real world" reports. Studies were included in the review if they reported the following; insulin dose, change in HbA1c, change in body weight and hypoglycemic events. In routine practice, decreases inHbA1c associated with insulin glargine and insulin detemir were variable and ranged between approximately -0.3% to -1.5% depending on prior treatment, but switching to insulin analogs was associated with less weight gain than previously reported. Compared with data reported in published trials, hypoglycemic event rates associated with basal analog insulin use in clinical practice were lower in patients initiating insulin and comparable in patients using basal-bolus regimens. Most patients were treated with low doses of insulin analog administered as once daily injections. In routine clinical practice most patients stopped concomitant use of sulfonylureas when initiating insulin analogs and it is likely that this together with the lower dose of insulin influenced outcomes. Nevertheless, despite initiating insulin therapy, it was also apparent that for many patients in routine clinical practice attainment of glycemic goals remains elusive.