To study the appropriateness of phlebotomy for digoxin therapeutic drug monitoring (TDM) in outpatients, we conducted a retrospective chart review, a computer search of all previous TDM testing, and a questionnaire of all outpatients (n = 86) who had serum digoxin determinations between April 10 and April 28, 1992 (585 tests). In patients who took digoxin at the same time daily (40 patients, 300 tests), 52% of tests were performed on inappropriate samples drawn within 6 h of the last dose. No patient who took digoxin after 1700 had inappropriate tests. Phlebotomy for serum digoxin determinations before distribution of digoxin is complete is a common problem in outpatients, leading to clinically uninterpretable test results. Postdistribution sampling can be assured by nighttime dosing, and this recommendation has been implemented at our hospital. INDEXING TERMS: blood sampling #{149} pharmacokinetics Digoxin is one of the most widely prescribed drugs for both inpatients and outpatients /1], and digoxin serum concentration is the most commonly ordered therapeutic drug monitoring (TDM) test in the US. Digoxin serum concentrations have been shown to be useful in corroborating the clinical diagnosis of digoxin toxicity as well as correlating with therapeutic effects of digoxin [2, 3]. Digoxin pharmacokinetics are complex hut well characterized (for a review see [4]). For oral dosing of digoxin, rapid absorption occurs in the intestine /4/; reaching distributional equilibrium after a digoxin dose requires at least 8 h, the time at which the semilog plots of plasma and tissue concentrations vs time reach terminal linear and parallel slopes [4, 5]. Samples obtained within 8 h after a digoxin dose can be very misleading as an index of digoxin response /4/.