The safety and effectiveness of hospital e-prescribing systems: a controlled time series study
Prescribing errors represent a significant public health issue internationally. In the US over 770,000 people are harmed or die each year in hospitals as a result of adverse drug events. In Australia 2% of hospital patients experience harm or death due to medication errors. The risk of these errors is risingeach year with the increased intensity of medical care, use of complex & potent drug regimens, and the increasing age and severity of hospitalised patients.
Health informatics interventions are becoming an increasingly important component of public health strategies aimed at reducing errors and improving the safety and quality of health care delivery. These interventions are usually high cost and have limited scientific evidence of their effectiveness. Developing approaches to evaluate these technologies and providing evidence of their effectiveness and impact upon the way care is provided is an urgent challenge. Electronic prescribing systems are strongly recommended as an effective intervention to significantly reduce the number of prescribing errors which cause harm, and those that have the potential to cause harm, to patients. No studies however have demonstrated that these systems result in significantly reducing prescribing errors that cause harm to patients. Alarmingly, anecdotal evidence from overseas suggests that use of electronic prescribing systems introduces new types of errors. This is consistent with a growing body of scientific research which suggests new decision-making biases may be introduced when individuals use decision support systems.
The aim of this project is to undertake a controlled time series study to assess the safety & effectiveness of two electronic prescribing systems (e-PSs) to reduce prescribing errors in two Australian hospitals. The study will use comprehensive, prospective error surveillance methods and compare error rates before & after the introduction of the systems. Using innovative data analysis techniques the study will investigate the safety of these systems in terms of the extent to which new types of errors result from system use. This will be the first empirical study of new error types and the role of automation bias in e-PS error generation.
The study will provide critical information about the effectiveness and safety of these e-PSs which are starting to be introduced into Australian hospitals. Unlike other medical interventions, which require stringent safety testing before use with patients, there are no minimum safety requirements for electronic prescribing systems. The data from this study is urgent and of international significance in developing recommendations to ensure the safe design and use of electronic prescribing systems. The research team brings together national and international experts from multiple disciplines required to achieve both the pragmatic and scientific goals of this research and to ensure that the research findings influence the future safe design and use of e-PSs.
Lead Researcher: Professor Johanna Westbrook -
Research Funding: NHMRC Project Grant 2006-2008
Collaborators:
- Concord Repatriation General Hospital, Sydney
- Centre for Health Informatics University of NSW
- St. Vincent’s Hospital, Sydney
- National Prescribing Service