The current study was performed to document the cost savings and need for a therapeutic drug monitoring program for phenytoin. The methodology employed a prospective, randomized, crossover design that utilized two control and two therapeutic drug monitoring phases. The therapeutic drug monitoring program significantly decreased the average number of assays performed per patient from 2.14 to 0.61. Withdrawal of the program resulted in a significant increase in the average number of assays performed per patient (from 0.61 to 2.41), the average number of assays drawn incorrectly (from 0.39 to 1.89), and in the average number of assays used inappropriately (from 0.50 to 2.07). Withdrawal of the program was also associated with a significant increase in the average number of readmissions (from 0 to 0.19) within 3 months of discharge. Reinstitution of the program was associated with a significant decrease in the average number of readmissions (from 0.19 to 0.03) within 3 months of discharge. The cost savings from decreasing the number of assays performed was estimated to be 100.00 for the first year.