Prospective cohort study of adverse events monitored by hospital pharmacists
1Angela Emerson BPharm, MSc, MRPharmS, 2Richard M. Martin BM BS, MSc, MRCGP, 1Mark Tomlin BPharm, MSc, MRPharmS, and 1,3Ronald D. Mann MD, FRCP, FRCGP, FRCP(Glas), FFPM, FCP On behalf of the Hospital Adverse Event Monitoring Study (HAEMS) group
1Pharmacy Department, Southampton University Hospitals NHS Trust, Southampton General Hospital, Tremona Road. Southampton S016 6YD, UK
2Drug Safety Research Unit, Bursledon Hall, Blundell Lane Southampton, SO31 1AA, UK
3Faculty of Medicine, Health and Biological Sciences, University of Southampton, Biomedical Sciences Building, Bassett Crescent East, Southampton, Hampshire, 5016 7PX, UK
Abstract
Purpose To examine the feasibility of pharmacist-led intensive hospital monitoring of adverse events (AEs) associated with newly marketed drugs.
Subjects/setting 303 patients admitted to Southampton Hospitals who were prescribed selected newly marketed drugs during their inpatient stay in 1998.
Methods Prospective observational study. Patients were identified from computerized pharmacy records, clinical pharma-cist ward rounds, dispensary records or via nursing staff. The pharmacist reviewed medical notes and recorded AEs, sus-pected adverse drug reactions (ADRs) and reasons for stopping drugs.
Outcomes Incidence of AEs, ADRs; proportionate agreement between the physician's and pharmacist's event recording.
Results 303 patients were monitored. Of the patients taking newly marketed drugs 9296 were identifiable using pharmacy computer systems and pharmacist ward visits. There were 21 (796) suspected ADRs detected during this pilot study. The types of adverse events detected were broadly similar to those identified by general practice-based prescription-event monitoring. However, biochemical changes featured more frequently than in general practice. Differences between adverse events recorded by pharmacist and physician were systematic and attributed to differences in event coding.
Conclusion Pharmacist-led monitoring in a typical NHS hospital setting was effective at detecting ADRs in newly mar-keted drugs. However, this effort might have been substantially less time-consuming and more effective were electronic patient records (EPRs) available. Pharmacy computer systems are not designed to be patient focused and are therefore unable to identify patients taking newly marketed drugs. It is argued that future EPR and computerised patient-specific prescribing systems should be designed to capture this data in the same way as some US systems are currently able to do.
KEY WORDS - cohort studies, adverse reaction, adverse drug reaction,adverse reaction monitoring, prescription-event monitoring
