Potential drug-drug interactions in the medication of medical patients at hospital discharge

Eur J Clin Pharmacol. 2003 Mar;58(11):773-8. doi: 10.1007/s00228-002-0557-z. Epub 2003 Feb 21.


Objective: Information on the frequency of drug combinations with the potential to induce dangerous drug-drug interactions (DDIs) in patients discharged from the hospital is scarce. With the present study, we assessed the frequency and potential clinical significance of DDIs in the prescriptions of discharged medical patients.

Methods: We retrospectively screened the medication for potential DDIs of 500 patients consecutively discharged with at least two prescriptions using a computerised drug-interaction program.

Results: The 500 patients (56.6% male, mean age 67.0+/-15.9 years, median length of stay 13 days) were prescribed a median of six drugs (range 2-18) at discharge. Three hundred patients (60.0%; 95% confidence interval 55.7-64.3%) had at least one potentially interacting drug combination. Of 747 potential DDIs at discharge overall, 402 (53.8%) were new at the time of discharge due to a change of the medication during the hospital stay. Of these, 72 (17.9%) were of potentially minor, 281 (69.9%) of moderate and 49 (12.2%) of major severity. Of 44 patients with a potential DDI with major severity, 1 patient was re-hospitalised within 2 months after discharge due to a probable drug-related problem associated with the potential DDI.

Conclusions: Using a computerised drug-interaction program, a high proportion of patients was detected with at least one potential DDI in the medication prescribed at discharge. However, the proportion of DDIs associated with potentially relevant clinical consequences appeared to be relatively low.

MeSH terms

  • Aged
  • Drug Interactions*
  • Drug Prescriptions / statistics & numerical data*
  • Drug Therapy, Combination
  • Drug Utilization
  • Female
  • Hospitals, University / statistics & numerical data
  • Humans
  • Length of Stay
  • Male
  • Patient Discharge / statistics & numerical data*
  • Risk Factors