Medication Reconciliation in Continuum of Care Transitions: A Moving Target

Objective: To study medication discrepancies in clinical transitions across a large health care system. Design: Randomized chart review of electronic medical records and paper chart medication reconciliation lists across 3 transitions of care. Settings and participants: Subacute patient medication records were reviewed through 3 transition care points at a large health care system, including hospital admission to discharge (time I), hospital discharge to skilled nursing facility (SNF; time II) and SNF admission to discharge home or long term care (LTC; time III). Measurements: Medication discrepancies were identified and categorized by the principal investigator and a pharmacist. Discrepancies were defined as any unexplained documented change in the patients' medication lists between sites and unintentional discrepancies were defined as any omission, duplication, or failure to change back to original regimen when indicated. Results: We reviewed 1696 medications in the 132 transition records of 44 patients, identifying 1002 discrepancies. Average age was 71.4 years and 68% were female. Median hospital stay was 5.5 days and 14.5 SNF days. Total medications at hospital admission, hospital discharge, SNF admission, and SNF discharge were 284, 472, 555, and 392, respectively. Total medication discrepancies were 357 (time I), 315 (time II), and 330 (time III). All patients experienced discrepancies and 86% had at least 1 unintentional discrepancy. The average number of medications per patient increased at time I from 6.5 to 10.7 (P < .001), increased at time II from 10.7 to 12.6 (P < .0174), and decreased at time III from 12.6 to 8.9 (P < .001). Patients, on average, had 8.1, 7.2, and 7.6 medication discrepancies at times I, II, and III, respectively. Surgical patients had more discrepancies than medical at times I and III (8.94 vs 5.3, P < .019; 8.0 vs 5.8, P < .028). In the unintentional group, cardiovascular drugs represented the highest number of discrepancies (26%). Conclusion: This study is the first to follow medication changes throughout 3 transition care points in a large health care system and to demonstrate the widespread prevalence of medication discrepancies at all points. Our findings are consistent with previously published results, which all focused on single site transitions. Outcomes of the current reconciliation process need to be revisited to insure safe delivery of care to the complex geriatric patient as they transition through health care systems. Copyright (C) 2013 - American Medical Directors Association, Inc.