OBJECTIVE: To describe a visually-friendly approach (Drug-O-Gram) for summarizing patterns of drug use over time using medications to treat two common disease conditions – arthritis and hypertension. METHODS: Administrative claims data (Medstat MarketScan) was used to identify hypertensive patients newly treated with diuretics, angiotensin-receptor blockers [ARB], or calcium-channel blockers [CCB]; and arthritis patients newly treated with cyclooxygenase-2 [COX-2] inhibitors or traditional non-steroidal anti-inflammatory drugs [NSAID]). Patients were then classified according to medication class use during each subsequent day for one year. The Drug-O-Gram plots the distribution of drug use over the follow-up period. Statistical smoothing was applied to the plots to help visualize trends over time. Superimposed on the plots are plots of outpatient medical costs impacted by medication use. RESULTS: Among patients with hypertension, the percent of days on medication was higher for ARBs (65%) compared to CCBs (59%) and diuretics (48%). Additionally, outpatient medical payments for ARB patients were 69% of diuretic and 56% of CCB outpatient payments. Among patients with arthritis, discontinuation was faster and persistency worse for traditional NSAIDs than for the COX-2 inhibitors. CONCLUSIONS: The Drug-O-Gram graphically summarizes drug use over time by fitting graphical smooths to daily averages. It then stacks these graphs to visualize changes between medications. These plots convey more detail of drug use patterns than commonly used discontinuation and switching rates. Areas under the plots can be used to quantify and analyze the amounts and changes of drug use over time. By superimposing daily plots of other outcomes, one can assess their relationship to drug use.