Effects of restricting levothyroxine dosage strength availability.

We conducted a prospective, randomized, controlled trial to assess whether hospital formulary restrictions involving limiting dosage strengths of levothyroxine affect physicians' ability to manage patients effectively and provide pharmacy cost savings in a tertiary care federal government research hospital. Thirty-three endocrinologists were randomly assigned to prescribe levothyroxine from a restrictive (dosage strengths of 25, 50, 100, 125, and 150 micrograms) or a nonrestrictive (dosage strengths of 25, 50, 75, 100, 112, 125, 150, 175, 200, and 300 micrograms) formulary through a central computer system. Their 241 respective outpatients' laboratory results and drug compliance were outcome measures. Achievement of treatment objectives was measured by thyroid function tests (free and total thyroxine, total triiodothyronine, thyrotropin), number of clinic visits, and compliance (survey method). Additional measures were drug distribution patterns, drug costs, and pharmacy inventory costs. Restriction of levothyroxine's dosage strength did not significantly alter therapeutic outcomes. However, the restricted formulary was associated with more complex dosing regimens, and resulted in no significant cost savings. It is not known whether such restriction would adversely affect the care of patients of nonspecialists. Prospective studies are required to verify presumed cost-containment measures before such measures are adopted for widespread application.

Automated therapeutic drug monitoring in an ambulatory care endocrine clinic.

OBJECTIVE: To develop and implement an automated therapeutic drug monitoring system for accessing data from endocrine clinic patients who had been prescribed insulin, oral hypoglycemic agents (OHA), or levothyroxine. DATA SOURCES: We designed a computer system to retrieve clinical data from the Medical Information System (MIS), a centralized hospital computer system, and import this information directly into a Macintosh personal computer. Physician entry of prescriptions for insulin, OHA, or levothyroxine into MIS formed the basis for a computer program to retrieve daily diagnostic and prescription information, demographics, and laboratory analyses, including blood glucose and glycosylated hemoglobin for insulin and OHA orders and free and total thyroxine, total triiodothyronine, and thyroid stimulating hormone for levothyroxine orders. The information was imported into a database program (4th Dimension). RESULTS: The system identifies laboratory values outside of predetermined therapeutic ranges, maintains an up-to-date patient profile, and edits and generates reports. Preliminary experience suggests that automation eliminates 75-90 percent of the time required to manually collect the same information, and improves the accuracy, comprehensiveness, and utility of reports. CONCLUSIONS: Automated therapeutic drug monitoring minimizes the time required to collect clinical data, alerts clinicians to potential problems, and provides a means to assess overall therapeutic management. Our methodology can be used to evaluate other medications in a variety of general or specialty clinics.