Medication Discrepancies Associated With a Medication Reconciliation Program and Clinical Outcomes After Hospital Discharge.

STUDY OBJECTIVE

To identify the frequency of unintended medication discrepancies 30 days postdischarge from medicine wards with interprofessional medication reconciliation processes and clinical import.

METHODS

Prospective cohort study of adults discharged between October 2013 and November 2014 from two teaching hospitals in Edmonton, Canada. The Best Possible Medication Discharge Plan (BPMDP) was prepared for all patients. Patients were called 30 days postdischarge to determine the medication discrepancy rate from the BPMDP and whether this was intentional or unintentional; three clinicians used standardized criteria to determine if the discrepancy was inconsequential. Electronic health records and patient contact were used to ascertain death, hospital readmissions, and emergency department (ED) visits at 90 days.

RESULTS

Of 433 patients (mean age 64 yrs, 52% female, median discharge prescriptions 6 [interquartile range 4-9]), 168 (38.8%) had at least one unintentional medication discrepancy at 30 days (325 total discrepancies; median one [interquartile range 1-2 discrepancies per patient]). Patients with unintentional medication discrepancies were older (65.9 vs 61.9 yrs, p=0.03) with more discharge medications (7 vs 6, p=0.03). Most unintentional discrepancies (91.1%) were judged inconsequential. The presence of an unintentional medication discrepancy was not associated with 90-day readmission or death (42/167 [25.1%] vs 64/263 [24.3%], adjusted odds ratio 0.96 [95% confidence interval 0.60-1.54]) or ED visits (69 [41.3%] vs 101 [38.4%], adjusted odds ratio 1.11 [95% confidence interval 0.74-1.67].

CONCLUSION

Despite the presence of an interprofessional medication reconciliation process, over one-third of patients had a medication discrepancy within 30 days of discharge, although most were inconsequential and there was no association between unintended medication discrepancies and risk of readmission, ED visit, or death 3 months after discharge.