Objective: Dosing and therapeutic range of digoxin has recently changed based on results from clinical trials. We examined the epidemiology, mortality and length of stay (LOS) of patients with serum digoxin level results during hospitalization. Methods: We retrospectively analyzed 4765 cases with serum digoxin levels from 2 institutions that electronically provided laboratory data from 2004-2006. Cases were stratified into the groups based on maximum serum digoxin level; <1.0, 1.0-2.0, 2.1-2.4, and >2.4. The actual to predicted hospital mortality and length of stay was compared across each strata with predicted mortality and LOS determined by previously described admission based clinical models. Results: Approximately 3 in 5 cases (57.8%) had a serum digoxin level higher than the recommended range of <1.0. The crude mortality for cases with digoxin levels <1.0 was 4.1% and 9.1% for those with digoxin levels = 1.0. After adjustment for severity of illness on admission cases with a digoxin level = 1.0 had a significantly higher actual to predicted mortality ratio (1.3 [CI = 1.1-1.4]) than cases <1.0 (0.8 [0.7-1.0]). While crude LOS was higher for serum digoxin levels = 1.0 (7.2 ± 7.7 days vs. 5.5 ± 5.6 days) after risk adjustment the actual to predicted LOS ratio was not significantly different from 1. Conclusion: The majority of cases have serum digoxin levels that are above the currently recommended range and cases with serum digoxin levels = 1.0 have a higher mortality. Cases with serum digoxin levels = 1.0 did not have higher risk-adjusted LOS. Digoxin dosing during hospitalization requires careful monitoring and has the potential of improving safety and related outcomes.