Clinical decision support systems: Some core concepts and questions


Year:

Session type:

Jeremy C Wyatt1
1Institute of Health Sciences, Leeds, UK

Abstract

Clinical decision support systems (CDSS) are "active knowledge systems using two or more items of patient data to produce encounter specific advice"1 and have existed since the 1950s. Designing a bias-free RCT of CDSS is challenging1,2 but over 160 RCTS have been carried out. In some settings they are effective3, even when their advice is printed on paper.4 Many different methods can be used to represent the knowledge in CDSS including rules, frames and causal probabilistic networks5, but clinicians are rightly sceptical about neural networks6 whose knowledge base cannot be inspected, as this opens them up to legal liability for damages.7 While CDSS are an attractive way to disseminate evidence based guideline recommendations or help implement tested clinical prediction rules, some key questions are:

1. Against which implementation barriers are CDSS an effective tool?8

2. How to build the CDSS knowledge base for easy maintenance?8

3. How to ensure that the CDSS can access complete patient data in suitably coded form?

4. How to design the advice to maximise its acceptability and impact9 while minimising the known potential for automation bias (professionals unthinkingly following advice even when it is incorrect).10

1. Wyatt J, Spiegelhalter D. Field trials of medical decision-aids: potential problems and solutions. In Clayton P (ed). Proc. 15th Symp. on Computer Applications in Medical Care, Washington 1991: 3-7

2. Randolph AD, Haynes RB, Wyatt JC, Cook DJ, Guyatt GH. Users' Guides to the Medical Literature: XVIII. How to Use an Article Evaluating the Clinical Impact of a Computer-Based Clinical Decision Support System. JAMA 1999; 282: 67-74

3. Roshanov PS, Fernandes N, Wilczynski JM, Hemens BJ, You JJ, Handler SM, Nieuwlaat R, Souza NM, Beyene J, Van Spall HG, Garg AX, Haynes RB. Features of effective computerised clinical decision support systems: meta-regression of 162 randomised trials. BMJ. 2013 Feb 14;346:f657. doi: 10.1136/bmj.f657.

4. Arditi C, Rège-Walther M, Wyatt JC, Durieux P, Burnand B. Computer-generated reminders delivered on paper to healthcare professionals; effects on professional practice and health care outcomes. Cochrane Database Syst Rev. 2012 Dec 12;12:CD001175. doi: 10.1002/14651858.CD001175.pub3.

5. Wyatt J. Computer-based knowledge systems. Lancet 1991; 338: 1431-1436

6. Wyatt J. Nervous about artificial neural networks ? Lancet 1995; 346: 1175-1177

7. Brahams D, Wyatt J. Decision-aids and the law. Lancet 1989; 2: 632-34

8. John Loder, Laura Bunt, Jeremy C Wyatt. Doctor Know: a Knowledge Commons in Health. London 2013: NESTA www.nesta.org.uk/home1/assets/features/doctor_know_a_knowledge_commons_in_health.

9. Scott GP, Shah P, Wyatt JC, Makubate B, Cross FW. Making electronic prescribing alerts more effective: scenario-based experimental study in junior doctors. JAMIA. 2011 Aug 11.

10. Kate Goddard, Abdul Roudsari and Jeremy C Wyatt. Automation bias: a systematic review of frequency, effect mediators, and mitigators. JAMIA June 16, 2011 doi: 10.1136/amiajnl-2011-000089