Accessing patient's veins for drug level sampling is not always feasible. The use of capillary sampling techniques is often utilized when venous access is hampered. In the therapeutic monitoring of patients, unexpected drug level results often occur that can be caused by a number of different factors. The possibility that differences in assay results might occur if samples were collected by capillary stick vs. venous phlebotomy was examined by simultaneous sampling in 18 patients. Although correlation was very high (0.999) and percentage differences fairly low (range of 0 to 15.4%), a statistical difference was noted in the sampling methods. The precision was 6.5 +/- 6.58% and there was a slight negative bias (-3.76%), with capillary samples less than venous samples. Although there were statistical differences for the drugs studied in the concentration ranges evaluated, capillary samples should provide fairly small errors when compared to venous samples.