A lack of knowledge of patients’ medication use at the time of prescribing combined with polypharmacy is an important cause of medication-prescribing errors. Inconsistent medication lists, involvement of several healthcare professionals and frequent changes in prescribing when the patient is hospitalised can lead to erroneous discharge letters; this lack of updated and uniform medication lists poses a problem for the continuity in patient care with the risk of less effective treatment and adverse drug reactions.
The study was carried out at a university college hospital in Denmark. Patients who were being discharged and consented to participation in the study were enrolled consecutively until 200 home visits had been carried out (83 surgical patients and 117 medical patients). They were visited in their homes one week after discharge and information on their current medication use was collected via interview, and observation of packages and containers. This information was compared with the written registrations in 1) the discharge letter and 2) the full hospital file (including drug list registered upon admission and that prescribed during admission). Discrepancies were noted and the possible consequences evaluated.
The median age of the participants was 75 years and the majority (70%) were women. Six patients had no prescription-only medicines (POM). The main results were: 194 patients (97%) stored 1189 POM in their homes (median 5 POM per patient, range 0–19) of which 955 were currently used daily or on demand (median 4 POM per patient, range 0–19). Among the 955 currently-used POM, 444 (46%) were registered in discharge letters and another 305 (31%) were registered elsewhere in hospital files giving an overall registration of 78%. The overall registration (discharge letter or hospital file) and the registration in the discharge letter was significantly higher among medical compared with surgical patients (p<0.001). Sixty-six POM users had no medication list in their discharge letter. Local treatments (eg, topical creams, eye preparations, inhalers) were registered less frequently than drugs administered orally. In total, 179 of the currently-used POM (19%) were not mentioned anywhere in hospital files, probably because of insufficient medication lists made at admission, and the prescribed regimen was unclear. At least 63 POM (7% of currently-used POM) were used in disagreement with the prescribed regimen.
These data show that the hospital has no knowledge of one-fifth of used POM and reports only half of used drugs in the discharge letter. Erroneous medication lists are most likely introduced because of insufficient registrations at admission and omissions at discharge. It is important to address these issues further and to improve the communication between primary and secondary care in order to prevent inappropriate medication use and adverse drug effects. Systematic follow-up after discharge focusing on an updated medication list might be effective in reducing medication errors.