Evaluating the effectiveness of a pharmaceutical intervention at the hospital discharge interface: An evidence based approach

Enhancing the communication of drug-related information at patient discharge has been a long term aim of the Department of Health and, despite many initiatives to improve the transfer of information, continuity of patient care remains problematic. Developing and evaluating the effectiveness of interventions to improve service delivery, in areas like the discharge interface, is complex, and a different approach is needed to that traditionally used in the evaluation of health care treatments and outcomes. The alternative approach outlined in this document requires the application of evidence achieved by rigorous methodology, in combination with a more pragmatic approach, ensuring relevance to practice and a high level of practitioner involvement. The purpose of this study was to develop a pharmaceutical intervention across the health care interface and the challenges to using this type of approach, and the practitioners' attitude towards involvement in the research project, were also explored in this study. The intervention was developed by a three stage process. Firstly, current practice was described from a range of different perspectives. Then, a theoretical model of pharmaceutical care was developed using the best available research evidence. This model was then adapted to meet the needs of practitioners; integration into current practice was ensured through practitioner involvement in the study design and a feasibility study. The effectiveness of the final intervention was the explored through a randomised controlled trial. The principal outcome measure used to evaluate the effectiveness of the intervention was re-admission to hospital. The main results were as follows: Survival analysis, measuring time to re-admission to hospital, identified a complex post-discharge period, where the pharmaceutical intervention appeared to delay but not prevent re-admission to hospital within 60 days of discharge. A proportional hazards model identified the number of drugs prescribed at discharge and the number of previous admissions to hospital as variables having the most effect on re-admission to hospital. The effect of patient cohort on number and time to re-admission to hospital was found to be less important than other patient factors, such as the number of drugs prescribed at discharge and number of previous admissions to hospital. This study also identified many of the challenges to undertaking research in a practice setting and further supported the necessity to combine rigorous methodology with a pragmatic approach to ensure relevance to practice. It identified a continuing gap between research evidence, health care policy and practice, and questions the reality of evidence based practice within the current health care system.