Evaluation of features to support safety and quality in general practice clinical software

BMC Medical Informatics and Decision Making

Research Article

Peter Schattner¹  Jennifer A O'Neill²  Margaret Williamson³  Michelle Sweidan³  James F Reeve³  Teri Snowdon⁴  Ken Harvey⁵ 

[1] Department of General Practice, School of Primary Health Care, Monash University, 3168, Notting Hill, VIC, Australia;Medical Software Industry Association Inc, PO Box 1293, 2076, Wahroonga, NSW, Australia;NPS: Better choices, Better health, Level 7, 418A Elizabeth St, 2010, Surry Hills, NSW, Australia;Royal Australian College of General Practitioners, 1 Palmerston Cres, 3205, South Melbourne, VIC, Australia;School of Public Health, La Trobe University, 3086, Bundoora, VIC, Australia;

关键词: Electronic Health Record;    Clinical Decision Support;    Software Feature;    Software Vendor;    Test Script;   

DOI  :  10.1186/1472-6947-11-27

 received in 2011-02-20, accepted in 2011-05-03,  发布年份 2011

来源: Springer

【 摘 要 】

BackgroundElectronic prescribing is now the norm in many countries. We wished to find out if clinical software systems used by general practitioners in Australia include features (functional capabilities and other characteristics) that facilitate improved patient safety and care, with a focus on quality use of medicines.MethodsSeven clinical software systems used in general practice were evaluated. Fifty software features that were previously rated as likely to have a high impact on safety and/or quality of care in general practice were tested and are reported here.ResultsThe range of results for the implementation of 50 features across the 7 clinical software systems was as follows: 17-31 features (34-62%) were fully implemented, 9-13 (18-26%) partially implemented, and 9-20 (18-40%) not implemented. Key findings included: Access to evidence based drug and therapeutic information was limited. Decision support for prescribing was available but varied markedly between systems. During prescribing there was potential for medicine mis-selection in some systems, and linking a medicine with its indication was optional. The definition of 'current medicines' versus 'past medicines' was not always clear. There were limited resources for patients, and some medicines lists for patients were suboptimal. Results were provided to the software vendors, who were keen to improve their systems.ConclusionsThe clinical systems tested lack some of the features expected to support patient safety and quality of care. Standards and certification for clinical software would ensure that safety features are present and that there is a minimum level of clinical functionality that clinicians could expect to find in any system.

【 授权许可】

CC BY   
© Sweidan et al; licensee BioMed Central Ltd. 2011