A crossover design study was conducted to determine if reducing pain, by using alternative sites off the finger tip, would increase testing frequency and improve clinical outcome as measured by glycosylated hemoglobin. Subjects with type I and type 2 diabetes tested with the FreeStyle alternative-site meter (group I) or tested with their original meter (group 2). After 3 months the subjects used the alternate meter. Testing frequency and blood glucose concentrations were recorded for the month before the study began and monthly thereafter. Glycosylated hemoglobin was tested initially, at the crossover point, and at study conclusion. Insulin users increased testing frequency from 2.4 to 3.0 tests per day. Testing frequency for non-insulin users remained the same at 1.5 tests per day. Testing frequency was essentially the same with the FreeStyle and the original meters. The average hemoglobin A1c was 7.4% (standard deviation 1.5%) initially, 7.3% (standard deviation 1.5%) at the crossover point, and 6.9% (standard deviation 1.1%) after 6 months. There was no significant difference in hemoglobin A1c measurements between meter types after 6 months. Thirteen months later a final hemoglobin A1c, testing frequency, and a questionnaire regarding meter preference and pain rating were obtained. Seventy-four percent of participants preferred the alternative-site meter, which was rated as significantly (p < .05) less painful. Testing frequency significantly improved (p = .001) while free strips were being provided. Testing frequency 13 months later was not significantly different from the baseline (p = .101). Hemoglobin A 1 c was significantly lower 6 months after the study began (p = .000) and 13 months later (p = .008) at baseline.