Improving the quality of prescribing in elderly hospital inpatients: evaluating the effect of a clinical pharmacist on ward rounds

Abstract

Introduction: Elderly patients are at increased risk of adverse medication events (ADEs) andpotentially inappropriate medicines (PIMs). The STOPP START tool has been validated to assessmedications of elderly patients for potentially inappropriate prescribing. There is little data onprescribing patterns for elderly Australian hospital inpatients. This Thesis aims to describe changesin prescribing as measured by the STOPP START tool in elderly patients throughout a hospitaladmission including admission and discharge from a specialised geriatric evaluation management(GEM) unit. This Thesis also assessed the effect of inclusion of a pharmacist on a physician-led wardround on the quality of prescribing in elderly hospitalised patients. The number of medicationsprescribed for each patient on admission to hospital, on transfer to the GEM unit and at dischargefrom the GEM unit for both the pre-intervention and post-intervention group were also compared.Method: This Thesis used an observational retrospective design to study the quality of prescribing intwo groups of patients, pre- and post-intervention, using the STOPP START tool at three pointsduring hospital stay; admission to hospital, transfer to a specialised geriatric unit and discharge fromhospital. Data was collected over 4 months pre- and post-introduction of a pharmacist to a physicianledward round. Demographic and clinical data, including total number of medications and STOPPSTART criteria met, were collected. The number of STOPP START criteria at the different timepoints, was compared between the pre- and post-intervention groups to determine whether there wasa change in potentially inappropriate prescribing after the intervention of a pharmacist participatingon the ward round. The mean number of STOPP START criteria (the total number of criteria metdivided by number of patients) at each time point were compared pre- and post-introduction of apharmacist using a Mann-Whitney U test. The mean number of criteria for each time point withinboth the pre- and post-intervention groups were compared using a paired Wilcoxon test.Results: The demographics of the participants in the pre- and post-intervention groups were similar.In the ninety-six pre-intervention group patients, 58 (60.4%) were female, the median age was 83[IQR 76-87] years and the mean number of co-morbidities was 5.10. The post-intervention group hadone hundred patients, 55 (55%) were female and the median age was 84 [IQR 78-89]. The postinterventiongroup had 21% less STOPP START criteria at discharge, mean 1.18 (SD 1.37) comparedto the pre-intervention group 1.50 (SD 1.41), p=0.07. The pre-intervention group had no significantchange in the criteria from admission 1.78 (SD 1.57) to geriatric unit transfer 1.72 (SD 1.54) (p=0.37) however there was a significant 13% decrease from geriatric unit transfer 1.72 (SD 1.54) to discharge1.50 (SD 1.41) (p=0.02). The post-intervention group had a 26% decrease in criteria from hospitaladmission 2.30 (SD 1.91) to geriatric unit transfer 1.59 (SD 1.60) (p<0.01) and again to discharge1.18 (SD 1.37) (p<0.01).The total number of medications prescribed per patient in the pre-and post-intervention groups werenot significantly different between the groups at admission (median 7 [IQR 5-10] vs 8 [IQR 5-10],p=0.4) or discharge (median 8 [IQR 6-11] vs 9 [IQR7-11], p=0.5). However, in the individual groups,the number of medications increased from admission to discharge (pre-intervention group 7 [IQR 5-10] to 8 [IQR 6-11], p<0.01), (post-intervention group 8 [IQR 5-10] to 9 [IQR 7-11], p<0.01).Conclusion: This Thesis found that prescribing quality changed for elderly patients during admissionto an Australian hospital. During the acute hospital stay potentially inappropriate prescribing did notchange significantly for the pre-intervention group however a decrease in STOPP START criteriawas seen in the post-intervention group. This Thesis found that after admission to the GEM unitprescribing quality improved, as measured by less STOPP START criteria on discharge in both thepre- and post-intervention groups. We observed that whilst the number of prescribed medicines perpatient increased, potentially inappropriate prescribing was reduced during the hospital stay. Thelesser amount of STOPP START criteria for the post-intervention group on discharge compared tothe pre-intervention group is evidence that pharmacist participation on physician ward rounds canimprove prescribing quality and supports pharmacist participation on ward rounds as a valuableaddition to the pharmaceutical care provided by clinical pharmacists.