Poor compliance with prescribed medications limits the effectiveness of many pharmacologic therapies and enhances their potential toxicities. Traditional methods of measuring drug‐taking behavior, including direct observation, patient self‐report, pill counts, and therapeutic drug level monitoring, all have well‐described limitations in validity and interpretability. Electronic medication event monitoring has been used to assess compliance with therapies for hypertension, glaucoma, anemia, and epilepsy, overcoming many problems of traditional approaches. However, no published reports describe the use of electronic monitoring with immunosuppressive agents, despite their increasing use for non‐life‐threatening conditions and their many dose‐dependent toxicities. Transplant recipients are thought to be at particular risk from noncompliance. Therefore, we undertook this study to assess the feasibility of electronically monitoring compliance with immunosuppressive drugs among renal allograft recipients. Twenty‐five kidney transplant patients receiving immunosuppressive medications from a single pharmacy were enrolled. Each subject received electronic monitors with their immunosuppressive serum drug refills for cyclosporine and azathioprine. Each subject returned their monitors after the first month of this 2‐month study for downloading data. The frequency distribution of interdose intervals were described. Two measures of average non‐compliance were calculated for both drugs: the proportion of monitored days that had missed doses, and the proportion of missed doses. Once daily and twice daily regimens of cyclosporine were compared. Concordance in drug compliance between the two drugs was calculated for each subject and averaged over the study population. Twenty‐two of 25 subjects missed one or more doses of cyclosporine or azathioprine. Seventeen (68%) subjects never missed four or more consecutive doses. Subjects were non‐compliant with cyclosporine on 8.7% of monitored days, and non‐compliance with azathioprine on 9.8% of monitored days. Subjects were non‐compliant with 6.8% of their cyclosporine doses and 9.8% of their azathioprine doses. Patients were compliant with both drugs on 86.6% of days and were non‐compliant with both drugs on 5.1% of days. Subjects were non‐compliant with cyclosporine during 5% and 13.2% of monitored days for once and twice daily dosing regimens, respectively. Concordance analysis demonstrated that for 91.7% of days of monitoring, compliance information was identical for both drugs. This study demonstrated the feasibility of electronic medication event monitoring among kidney transplant patients. This methodology represents an important tool for monitoring compliance of immunosuppressive agents essential to their safe and effective use, and should be considered for use in future studies of these drugs and others with substantial dose‐dependent toxicity. Copyright © 1999 John Wiley & Sons, Ltd.