SPECIAL CASE OF SCIENTIFIC IC REASONING? THEORY AND PRACT CTIC - PowerPoint PPT Presentation

CLIN INICAL DECISION MAKING AS A SPECIAL CASE OF SCIENTIFIC IC REASONING? THEORY AND PRACT CTIC ICAL IM IMPLIC ICATIONS FOR MEDIC ICAL EDUCATION

Disclosure sclosure Note te I serve as Board Director of Instruct AG (CASUS) in Munich, Germany) and hold shares of this company. Instruct AG is a LMU spin-off for-profit company.

Outline Ou line  Clinical reasoning and diagnostic competence: definitions and context  System 1 and 2 and beyond  Error reduction a key goal of diagnostic competence – some study results  Clinical decision making as a special case of scientific reasoning? Proposal of a model  Summary

Clinical Reasoning Clinical Cognition Diagnostic Reasoning Clinical Problem Solving Medical Problem Solving Clinical Decision Making … The thinking and/or decision-making processes that are used in clinical practice (Higgs and Jones 2000, Edwards et al 2004)

Defi finition Dia iagnosing and Dia iagnostic competence Diagnosing is the purposeful collection and integration of information [for the purpose of uncertainty reduction] to make medical decisions. Diagnostic competence is demonstrated on the basis of diagnostically relevant knowledge and diagnostic skills by • the accuracy of the diagnosis • the critical consideration of effectiveness (expected value in terms of uncertainty reduction) and • the efficiency (the necessary expenditure for the uncertainty reduction). Helmke 2010; Zentralantrag DFG-Forschergruppe FOR 2385, 2016

Clinical nical Re Reasoning oning – Cont ntext ext Dependence? endence? Kiesewetter & Fischer 2014

Mental Processes: Permanent “dual processing” of analytic and non- analytic thinking (“systems 1 and 2“) Eva K Med Edu 2005, Norman G Med Edu 2007, Norman G Med Edu 2010

Ment ntal al Proce cess sses es in Cl Clini nical cal Reasoning: soning: Permanent “dual processing” of analytical and non -an analy alyti tica cal th thinking nking (Sy Syst stems ems 1 a 1 and d 2) 2) Syste stem 2: anal alytica ytical Syste stem 1: non on-analyt analytical ical fast, impulsive, unconscious match to prior slow, logical, conceptual, amounting to the examples stored in memory (pattern logical application of “rules” (hypothetical - recognition) deductive)  Influenced by the representativeness of  Heavy load on working memory, which has the new problem and the availability of real limitations in speed and size prior similar cases Eva K Med Edu 2005, Norman G Med Edu 2007, Norman G Med Edu 2010

Estimate of Diagnotic errors for Adult Outpatients in the US: About 5% or 12 Millons Adults per Year! Singh H, et al. BMJ Qual Saf 2014

Diag agnos nostic tic Errors rs in Inte ternal rnal Medici dicine: ne: 228 syste tem m rela lated ed factors ors technical failure, equipment problems, teamwork, supervision, management, coordination of care, expertise unavailable, policy/procedures, … 320 cogniti itive ve factor tors 11 faulty knowledge 45 faulty data gathering 159 faulty information processing 106 faulty verification 39 premature closure (most frequent cognitive error) Graber ML Arch Int Med 2005

How could we better understand and operationalize diagnostic competence to design educational interventions and ultimately reduce errors?

Op Operat erationaliz ionalizin ing g Di Diagn gnostic ostic Co Comp mpetenc etence Conceptual Procedural knowledge Knowledge Strategical Conditional • Basic concepts • Problem-solving • Rationale for a strategies procedure Scientific knowledge Individual experience (Stark, Kopp, & Fischer M., 2011) (Paris, Lipson, & Wixson, 1983; van Gog, Paas, & van Merrienboer, 2004)

Schmidmaier et al. BMC Med Edu 2013

Schmidmaier et al. BMC Med Edu 2013

Onli line ne Learning rning Envir viron onme ment: nt: Case se Exampl mple e

Exam ampl ple e with thou out t error or and d elabo borate rate feedb dback ck from om expe pert rt

Medical Education 2008

Kiesewetter et al. PLOSone 2013

Mod odel l für com ompl plex x pr prob oblem em-solving olving in Medicine icine (modifi fied d from Schoenfe enfeld ld) Operationalised definition Cognitive Action Retrieve information; read Denomination Analyse information; generate Analysis differential diagnostic ideas Associate, compare, vaguely Exploration propose strategies how to understand the problem Generate plans, weigh up these Plan plans against each other, decide on a plan State and justify one definite plan; Implementation request certain additional information and/or examinations Verify or dismiss hypotheses with Evaluation regard to new information or examination results; evaluative thinking Inner representation of the case; Representation statement of the situation as far as it is summarized in the mind of the student Decision for one working diagnosis, Integration differential diagnoses and/or therapy Kiesewetter et al. PLOSone 2013

Complete Incomplete Kiesewetter et al. PLOSone 2013

Incorrect Correct solution solution Incomplete 29 cases; 26/29; 90% 3/29; 10% model 44% Complete 37 cases; 13/37; 35% 24/37; 65% model 56% 39/66; 59% 27/66; 41% Kiesewetter et al. PLOSone 2013

Cl Clinica nical Reas asoning oning – a Sp Speci cial al Ca Case se of S f Sci cien entifi tific c Reasoning soning and d Ar Argumenta mentati tion? on?

“ Scienfic reasoning (…) includes the thinking and reasoning skills involved in inquiry, experimentation, evidence evaluation, inference, and argumentation that supports the formation and modification of concepts and theories about the natural and social world.” Bao et al., 2009, Science

On One-dime dimensional sional Mo Mode del Use Underst erstanding anding

Two-dimens dimension ional al Mo Mode del Underst erstanding anding Use Donald E. Stokes, Pasteur's Quadrant – Basic Science and Technological Innovation, Brookings Institution Press, 1997

Two-dimens dimension ional al Mo Mode del: l: Fou our r Quadrants drants ?! ?!? Donald E. Stokes, Pasteur's Quadrant – Basic Science and Technological Innovation, Brookings Institution Press, 1997

Problem 1. Problem Identification identification Communica- 2. Questioning tion/ Questioning Scrutinizing 3. Hypothesis generation soning for reasoning 4. Construction of artefacts Drawing Hypothesis conclusions generation 5. Evidence generation ities fo ctivities 6. Evidence evaluation ic act 7. Drawing conlcusions stemic Evidence Construction of pistem evaluation artefacts 8. Communcation/Scrutinizing Epi Evidence generation

Advancing theory- Artefact‐centered Science-based building about natural scientific reasoning in and social phenomena Reasoning practice ( Bohr’s quadrant ( Pasteur’s quadrant ( Edison’s quadrant of basic research) of use-inspired of applied research) basic research) Problem Problem Problem identificatio identificatio identificatio n n n Communica- Communica- Communica- tion/ tion/ tion/ Questioning Questioning Questioning Scrutinizing Scrutinizing Scrutinizing Drawing Hypothesis Drawing Hypothesis Drawing Hypothesis conclusions generation conclusions generation conclusions generation Constructio Constructio Constructio Evidence Evidence Evidence n of n of n of evaluation evaluation evaluation artefacts artefacts artefacts Evidence Evidence Evidence generation generation generation

Clinical Reasoning Modelle Mot Model of Clinical Model of Clinical Model of Scientific Reasoning Reasoning Reasoning (Charlin et al., 2012) (Bowen, 2006) (Fischer et al., 2014) Inital representation Problem Identification Patient’s history of the problem Formulating Questions Data acquisition Generating Hypothesis Dynamic Constructing Artifacts Accurate problem representation representation Generating Evidence of the problem Generation of hypothesis Evidence Evalutation Conclusions Search for and selection of illness script Final Communcation Diagnosis representation of the problem Meta -cognition Knowledge Context Experience PhD-Thesis Ch. Strobel 2016

Summ mmary ary (1 (1) • Clinical reasoning is multi-contextual and multifacetted • Diagnostic competence is a relevant part and measurable contruct of clinical reasoning • Erroneous case examples and elaborate feedback have potential to improve diagnostic competence

Summ mmary ary (2 (2) • The cognitive part of clinical reasoning is a form of problem-solving • The completeness of a problem-solving process predicts diagnostic accuracy • Clinical reasoning may ideally be an example of science- based reasoning in practice (Edison’s quadrant of applied research)

Many thanks for your attention! martin.fischer@med.uni-muenchen.de

Those who can’t do teach. Those who can’t teach teach how to teach. Those who can’t teach how to teach do research on teaching. adapted from Bernhard Shaw The doctor's Dilemma 1906