Is hospital admission a sufficiently sensitive outcome measure for evaluating medication review services? A descriptive analysis of admissions within a randomised controlled trial

Abstract

Abstract Objective The aims of the study were: to describe and assess hospital admissions occurring during a randomised controlled trial (RCT) of a pharmacist-led medication review service; to describe the admissions in terms of emergency status and main cause; to estimate the potential contribution of pharmaceutical care issues (PCIs) to admission; and to assess the proportion of admissions that could be influenced by a pharmacist intervention. Setting Within the context of a RCT of pharmacists providing medication review for 332 elderly patients living at home, taking at least four repeat medicines, carried out in one region of Scotland. Method Hospital data were obtained for all admissions occurring during the 9-month period studied, summarised and evaluated by two independent medical reviewers for the contribution of PCIs to admission. Two pharmacists assessed the extent to which PCIs were preventable by pharmacist intervention. Key findings Approximately two-thirds of the 77 admissions were unplanned, and two-thirds were to medical wards. Only 17 (22%) of all admissions were considered to be related to PCIs and 10 (13%) possibly preventable by pharmacist intervention. Although the majority of surgical admissions were considered to be unrelated to PCIs (26/29), both unplanned and planned medical admissions were related to PCIs. One of these occurred as a direct result of the pharmacist's recommendation. Conclusion The overall numbers of hospital admissions, medical admissions and unplanned admissions may not be sufficiently sensitive outcome measures for evaluating the impact of pharmacist interventions. Consideration could be given to developing categories of admission that are related to medicines or are likely to be preventable as more relevant measures. Including more details of hospital admissions in future studies may be useful.