case-based clinical reasoning in feline medicine 3: …...418 jfms clinical practice special article...

418 JFMS CLINICAL PRACTICE

S P E C I A L art i c l e

case-based clinical reasoning

in feline medicine

3: Use of heuristics and illness scripts

Martin L Whitehead, Paul J Canfield, Robert Johnson, Carolyn R O’Brien and Richard Malik

Martin L WhiteheadBSc PhD BVSc CertSAM

MRCVSChipping Norton VeterinaryHospital, Banbury Road,Chipping Norton, Oxon,

OX7 5SY, UK

Paul J CanfieldBVSc PhD GradCertEdStudDVSc FANZCVS (Veterinary

Clinical Pathology)FRCPath MRCVSFaculty of Veterinary

Science, B14, University of Sydney, NSW 2006, Australia

Robert JohnsonBVSc MANZCVS (FelineMedicine) CertZooMed

BA CMAVASouth Penrith VeterinaryClinic, 126 Stafford Street,Penrith, NSW 2750, Australia

Carolyn R O’BrienBVSc (Hons) MVetClinStudFANZCVS (Feline Medicine)

Faculty of VeterinaryScience, The University ofMelbourne, Parkville, VIC 3152, Australia

Richard MalikBVSc DipVetAn

MVetClinStud PhD DVScFANZCVS (Feline Medicine)

FASM MASIDCentre for VeterinaryEducation, B22,

University of Sydney, NSW 2006, Australia

Corresponding author: [email protected]

Aim: This is Article 3 of a three-part series on clinical reasoning that encourages practitioners to explore and understand how they think and make case-based decisions. It is hoped that, in the process, they will learn to trust their intuition but, at the same time, put in place safeguards to diminish the impact of bias and misguided logic on their diagnostic decision-making.Series outline: Article 1, published in the January 2016 issue of JFMS, discussed the relative merits and shortcomings of System 1 thinking (immediate and unconscious) and System 2 thinking (effortful and analytical). In Article 2, published in the March 2016 issue, ways of managing cognitive error,particularly the negative impact of bias, in making a diagnosis were examined. This final article explores the use of heuristics (mental short cuts) and illness scripts in diagnostic reasoning.

lead under certain circumstances.So, if there is some negativity

about the use of heuristics, whyis it that they survive in decision-making? The answer may wellstem from the fact that they arepart of an ‘ancient’ form of thinking and are necessary for

survival in an adaptive world.2

That is possibly still true today for many of us.What is irrefutable, however, is that they savetime, and the brain has a natural tendency,especially when stressed or tired, to incorpo-rate them into System 1 thinking in order torapidly reach an approximate or ‘best guess’answer. Moreover, they can be powerful whenused sensibly and knowingly as general-purpose strategies in trained System 2 think-ing, with the proviso that any negative biasinvolved is acknowledged.

The medical profession has recognised, withvarying views onacceptance, theuse of heuris-tics alongwith illnessscripts inclinical rea-soning.3,4

What are heuristics and how do they playa role in diagnosticreasoning?

You may well be thinking, ‘Whydo I need to know anythingabout heuristics?’ Well, take itfrom us, you already use heuris-tics! Heuristics are used by everyone, every day,sometimes knowingly but often unknowingly.We often use heuristics unconsciously in

circumstances where a decision needs to bemade relatively quickly, whether it be ‘how doI know if this new car (or associate) is right forthe practice?’ or ‘how do I manage this frac-tious cat?’. Sometimes heuristics are moredeliberate and consciously used when wehave plenty of time to make a clinical deci-sion. Heuristics are ‘rules of thumb’ that proveuseful for diagnosis more times than they leadto error. Sometimes they are referred to as‘fast and frugal’ mental short cuts.1Since heuristics appear to be more common-

ly used in System 1 thinking and less commonly in trained System 2 thinking, theyhave been maligned as being cognitively undemanding. Moreover, some heuristics arecognitive biases and have the potential to mis-

‘Why can’t I alwaysunderstand how Ireached the correctdiagnosis – am I agenius, or was I justlucky?’

VIDEOAuthors Paul Canfield and Richard Malikdiscuss many of the concepts coveredin this article series on clinical reasoning

in a video available as Supplementary material at: jfms.comDOI: 10.1177/1098612X16643251

doi: 10.1177/1098612X16643251

© The Author(s) 2016

Journal of Feline Medicine and Surgery (2016) 18, 418–426

JFMS CLINICAL PRACTICE 419

SPEC IAL ar t icle / Use of heuristics and illness scripts in clinical reasoning

Definition Example

CATEGORY 1 Heuristics that are cognitive biases and which act as inherent parts of clinical reasoning, primarily in System 1thinking. They particularly come into play when choices have to be made between several diseases that haveoverlapping features. They can give rise both to correct and incorrect diagnostic decisions, but the former shouldoutweigh the latter

Availability heuristic Estimating the frequencies of events (ie, the probability of occurrence of an illness) on the basis of how easy it is to recall them from one’s memory(past experiences)– Care has to be taken to avoidoverestimation of the commonness of a disease just because it is easy to recall or treat due to some other factor(s)

< Good use: Progressive muscle weakness resulting in ascending paralysis in cats seen in your practice on the east coast of Australia is most commonlydue to Ixodes holocyclus (Australian paralysis tick); especially when the catlooks anxious, vomits or has difficulty breathing and/or a change in its meow.Because you see this so commonly, this is the first thing that springs to mind when the next case of ascending paralysis in a cat is presented. But confirm by searching for an engorged tick or tick bite ‘crater’< Bad use (misleading): There has been much recent publicity about leptospirosisas a zoonotic disease and the risk companion animals may pose to their owners.The next time you see an icteric cat, you immediately think of leptospirosis as acause of liver disease, even though domestic cats domiciled in the city (oranywhere for that matter) are most unlikely to have symptomatic leptospirosis

Representativenessheuristic

This is based on the acceptance of a high probability of commonality betweenobjects of similar appearance – The assumption is that the individualfeline patient will always present with (at least some) characteristic clinicalsigns and other features derived fromhistory, signalment and physical findingsreported for the stereotype of the disease(ie, representative of the disease)

< Good use: A young adult cat presents with signs of cystitis, but with severalpossible causes. Idiopathic cystitis is considered on the basis of a combinationof case features including that the cat is an overweight indoor cat in a multi-cathousehold, is very anxious, and hides under the bed whenever visitors come. We would possibly not have thought of idiopathic cystitis if the cat had beenslim, confident, with free access to outdoors and was the only cat in thehousehold. This is because the former cat is very representative of the ‘class’ ofidiopathic cystitis cats. The latter cat might well have idiopathic cystitis (indeed,idiopathic cystitis is probably the most likely cause of the cystitis signs), but isnot representative – in a vet’s mind – of the class of idiopathic cystitis cats< Bad use (misleading): A 3-year-old cat presents in late spring withappendicular weakness. The weakness seems to improve with rest. Because it is tick season and the clinical presentation is fairly typical, you think tickparalysis is the most likely diagnostic possibility (although you cannot find anengorged I holocyclus tick, or even a tick bite crater). You administer tickantiserum and monitor the cat. Your colleague takes over the case the next day,considers myasthenia gravis as a possibility because a tick was not found (andthey checked themselves) and administers edrophonium. The cat shows apositive response (improved muscle strength after intravenous injection) and thediagnosis of myasthenia is confirmed by demonstrating antibodies against theacetylcholine receptor using a specialist neuromuscular laboratory overseas

Familiarity/recognitionheuristic

Using this heuristic, greater emphasis isplaced on features or evidence of diseasewith which you are more familiar (ie, moreeasily recognised)– Potentially certain clinical signs orresults of tests have a greater impact onyour thinking, which could be a good or abad thing depending on what you ignore– While the familiarity and recognitionheuristics are closely related, the lattercan be regarded as a more ‘extreme’form, and is more prone to error

< Good use: You are presented with a young cat that is displaying a cough,fever, loss of appetite, weight loss, reduced breath sounds on auscultation and a bluish tinge to the gums. The fever in combination with respiratory signs(as a pattern or even generic illness script) immediately makes you think of thepossibility of pyothorax< Bad use (misleading): This cat also has heart sounds that are displacedcaudally, and a single large superficial (right) cervical lymph node. But thesefeatures for some reason are given less emphasis, and initially ignored. Furtherinvestigations demonstrate that the cat has mediastinal lymphoma and a largelymphocyte-rich pleural effusion. With the benefit of hindsight the significanceof cardiac displacement is now obvious

Table 1 Common heuristics that may be used in clinical decision making (continued on page 420)

There are many types of heuristics, and theyfall into two distinct categories (Table 1). Theavailability, representativeness and familiari-ty/recognition heuristics are cognitive biasesthat are an inherent part of our reasoning

processes (especially System 1 thinking);depending on circumstances, they result incorrect decisions more often than incorrectdecisions. Veterinarians are likely to use thiscategory of heuristics without recognisingthem when engaging in System 1 thinking.These heuristics will draw on long-term mem-ories concerning patterns of features or evencomplete generic illness scripts to help decideon a diagnosis. They come into play when sim-ple pattern recognition lets you down becausethe case data derived from history, signalment,clinical signs and physical findings couldmatch any of several possible diseases. inother words, heuristics in System 1 thinkinghelp when having to make a choice among diseases that have overlapping features.

Heuristics in the critical care settingApart from their general use in clinical diagnosis, heuristics can be especiallyimportant in a critical care setting (ie, in dealing with emergencies) – whereinterventions are stressful to the staff and patient, and where assessing theresponse to therapy may be more informative (and perhaps safer) thaninvestigations. The heuristic may be as simple as a strategy to keep theanimal alive while you decide on further investigations to find the cause; forexample, providing an oxygen-enriched environment for a cat with dyspnoea,while preparing to do other investigations such as thoracic imaging.


SPEC IAL ar t icl e / Use of heuristics and illness scripts in clinical reasoning

Definition Example

CATEGORY 2 These heuristics are not cognitive biases as they act as general-purpose strategies for approaching complex or difficult diagnoses. They tend to operate in System 2 thinking, except for the anchoring heuristic, which alsooften acts in System 1 settings

Anchoring andadjustment heuristic

The clinician starts withan implicitly suggestedreference point (the‘anchor’) and makesadjustments to it to reach a decisionabout diagnosis – The anchor maymislead if it is affectedby bias or if a piece ofinformation (a premise)was plainly wrong fromthe start (‘urban myth’)– This heuristic is usedin both System 1 andSystem 2 thinking

< Good use: A cat is presented with dyspnoea. You begin your investigation by employing theanchor that dyspnoea in cats is most commonly due to cardiorespiratory disease, and often iscaused by a pleural effusion. So diagnostic testing will likely involve ultrasound of the chest,echocardiography and thoracic radiography (plus thoracocentesis if fluid is present). In thisinstance, ultrasonography reveals a cranial mediastinal mass and a pleural effusion. With thefinding of the mass you ‘adjust’ from your anchor and assume that this mass lesion is moreimportant diagnostically. A chest tap produces red-tinged fluid, which cytologically shows auniform population of large lymphoblasts, consistent with mediastinal lymphoma< Bad use (misleading): You are presented with a young cat with dyspnoea and subtle stertor.You begin your investigation by employing the anchor that overt dyspnoea in cats is mostcommonly due to cardiorespiratory disease, and often is caused by a pleural effusion. Sodiagnostic testing will likely involve ultrasound of the chest, echocardiography and thoracicradiography. All these investigations are unremarkable. You then read that stertor is often a signof nasopharyngeal disease. Posterior rhinoscopy demonstrates a mass, which can be dislodgedby vigorous antegrade nasal flushing. Histology of the tissue specimen demonstrates large celllymphoma. In this instance, ‘dyspnoea is commonly due to disease of the chest’ proved to be a false anchor. Dyspnoea plus stertor is suggestive of nasopharyngeal disease and this wouldhave been a better anchor, and would not have misled the investigation (although no harm wasdone, some of the client’s money might have been better spent)

Means-end and hill-climbingheuristics

In the means-endsheuristic a largereasoning problem isdivided into smaller ‘subproblems’, in theknowledge (or hope!)that solving all of thesmaller problems willresult in solving thelarger problem. The hill-climbing heuristic differsin that there is greateruncertainty aboutwhether the choice of a smaller subproblemmight start you on yourway to solving the largerproblem– These heuristics arecommonly used inSystem 2 thinking

< Good use (means-end heuristic): A cat is presented with a complex, likely multiorgan,disease just before you are about to finish after a long day at work. Neither the history nor thephysical examination provide any useful clues as to the cause of the cat’s illness. You assess that part of the problem is severe dehydration with electrolyte and acid/base derangements. In-house blood tests demonstrate metabolic alkalosis and hypokalaemia. You decide initially to improve hydration and correct these secondary electrolyte changes. The combination ofmetabolic alkalosis and hypokalaemia is a ‘pattern’ that is usually associated with loss of gastricacid and potassium through vomiting (often with a fixed obstruction at the level of the pylorus or duodenum), with further loss of potassium into the urine. You plan on performing abdominalultrasound and contrast radiology. However, the investigation can wait until tomorrow, when the cat’s status is improved sufficiently to cope with imaging studies< Bad use (misleading) (hill-climbing heuristic): A cat is presented with a complex, likelymultiorgan, disease just before you are about to finish after a long day at work. You haven’t a clue what is going on, but notice that its fur is matted under the chin and you decide to clip it.You find what looks like a healing cat-fight abscess which has burst and is draining. You collectblood for FIV and FeLV testing, give the cat an injection of amoxicillin clavulanate, offer it foodand water, and head home. Although you have found two problems, and treated one of themeffectively, it is not likely to be the cause of the cat’s overall status (unless there is anotherabscess you have missed). You have climbed two small hills: the cat might have been in fightsbefore, may have become infected with the FIV virus, and this may have something to do withthe current problem, but you have a long way to go. Hill climbing usually pays off if you doenough tests and link the answers correctly – but be prepared, sometimes it doesn’t!

Progress-monitoringheuristic

This is reflection on howyou are managing a case– This heuristicprobably kicks in whenhill climbing is just notgetting you far enoughtowards a finaldiagnosis, so you decideto use another strategy,or try going in a differentdirection – The most time-intensive of the category2 heuristics used asreasoning strategies in System 2 thinking

< Good use: Reflection on the results from an anaemic cat suggests that your initial thoughtsare not going to explain the clinical picture. The anaemia is non-regenerative, this has notchanged over 3 days (so it wasn’t just pre-regenerative), the cat is FIV and FeLV negative, thePCR for Mycoplasma haemofelis is negative and the ferritin level is normal. So you change tack,and embrace the possibility that the cause of anaemia is primary bone marrow disease and elect to do a bone marrow aspirate for cytological assessment. (This example also fits well withan anchoring and adjustment heuristic – ie, the reasoning strategy heuristics are not mutuallyexclusive)< Bad use (misleading): You are not getting far investigating a cat with anaemia and acolleague suggests that you focus more on the nasal philtrum ulceration that is also evident. You take a biopsy, but before the results come back the animal dies due to intra-abdominalhaemorrhage. The biopsy of the nasal lesion is consistent with photosensitization. Histologicalexamination of a liver biopsy collected at necropsy demonstrates hepatic amyloidosis. (This example probably starts off as hill climbing and turns into progress monitoring)

Table 1 (continued from page 419)

in contrast, anchoring and adjustment,means-end, hill-climbing and progress-monitoring heuristics are not cognitive biases,but general-purpose strategies for approach-ing complex problems, including but notrestricted to reasoning processes. These aregenerally used in System 2 thinking for diffi-cult diagnoses and veterinarians will be awareof their use, if not their names!

Availability heuristicThe availability heuristic relies on estimatingthe frequencies of events on the basis of how easy it is to recall them from one’s memory of past experiences.5 it is probablythe origin of the old adage ‘common thingsoccur commonly‘.

The availability heuristic can work well aslong as certain variables are controlled; for


example, if types of disease remain constantin the local population or when the frequencyof occurrence does not differ between geo-graphic regions. But what, say, if a UK veteri-narian undertaking a locum placement inAustralia is presented with a cat with dilatedpupils, muscle weakness and labouredbreathing. Snake bite as a possible cause may not be part of the availability heuristic.Hopefully a colleague or an astute veterinarynurse may consider this diagnostic possibility,because for them it would be a commonoccurrence and easy to recall.

As this example suggests, the availabilityheuristic can influence our decisions whendeciding on a diagnosis, or even drawing up alist of diagnostic possibilities, based on a perceived pattern of signs or more completecase data. Note, however, that as discussed in Article 2,6 there is the risk of availabilitybias leading to an incorrect answer if a mem-ory is recalled not on the basis of it being acommon occurrence but due to its unusual-ness or some other circumstances that left astrong impression.

Representative heuristicThe representativeness heuristic relies on theassumption that there is a high probability ofcommonality between objects of similarappearance – in other words, a ‘what does thisremind me of?’ approach.5 in diagnostic rea-soning, this is a useful heuristic (ie, gives thecorrect answer more often than not) because,while the presentations of many diseases canvary substantially between patients or overtime in one patient, most presentations of any one disease share at least some featuresderived from a combination of history, signal-ment, clinical signs or physical findings.

The clinical signs of erythema, crusting skinwith early bleeding and ulceration at the tipsof the ears in a white cat might be a cogentexample of the use of pattern recognition inarriving at a presumptive diagno-sis of actinic keratosis progress-ing to squamous cell carcinoma(especially in sunny Australia),but it is not an example of the use of the representative heuris-tic. However, if some of the casedata did not classically fit withactinic keratosis (eg, the cat rarelywent outside) and other case dataraised the possibility of other dis-eases (eg, the cat had been both-ered by flies recently, or had ahabit of rubbing its ears againstan armchair that had just beensent off to be ‘cleaned’), then therepresentative heuristic will comeinto play in deciding between the


alternatives. Thus, the representativenessheuristic draws on patterns of features andsometimes whole illness scripts that are in theclinician’s long-term memory to choosebetween diseases with overlapping clinical features.

Familiarity/recognition heuristicThe third heuristic of note in System 1 think-ing is the familiarity heuristic (familiar thingsare regarded as being more ‘important’ orhaving more ‘value’ than unfamiliar things, sowe place more emphasis on diseases or clini-cal signs that we are familiar with). The recog-nition heuristic can be regarded as a morerestricted, more ‘extreme’ version of the famil-iarity heuristic (recognisable things are more‘important’ or have more ‘value’ thanunrecognisable things, so we concentrate onthe recognised more than the unrecognised;may also be referred to as ‘take the best,ignore the rest’).

The familiarity/recognition heuristic allowsfocus on key features of a condition – a pre-sumptive diagnosis can be made more quick-ly, so that the clinician can rapidly move forward with confirmatory testing.Apparently spurious information is ignored,which is both the strength and weakness ofthis heuristic. Certainly this short cut canwork well, as long as something really impor-tant hasn’t been ignored that might have suggested a viable alternative diagnosis or apertinent comorbidity (concurrent disease).

Anchoring and adjustment heuristicThere are versions of the anchoring andadjustment heuristic in System 1 thinking(where this heuristic influences the way people intuitively assess probabilities7) and in System 2 thinking as a reasoning strategy (ie, as one of our second category of heuristics).As a reasoning strategy, veterinarians startwith an implicitly suggested reference point

(‘anchor’) and make adjustmentsto it to reach the diagnosis. Theanchor itself may arise fromSystem 1 or System 2 thinking,and may have the capacity tolead to correct or incorrect deci-sions (‘false anchors’, see Figure 1and also box on page 422)

By consciously utilising theanchor in System 2 thinking, it canbe assessed for its soundness and‘adjustments’ introduced sequen-tially and logically. Thus, theanchoring and adjustment heuris-tic is often an example of usingSystems 1 and 2 thinking in tan-dem (see Article 2), which can savemuch time in clinical reasoning.

Figure 1 Is this a case ofmissing an obvious anchor – or is the owner super-unobservant? Whatever thereason, starting with thewrong anchor can lead todisaster in diagnosis



‘False anchors’

Means-end and hill-climbing heuristicsin the means-ends heuristic a complex reason-ing problem is divided into smaller ‘subprob-lems’ in the knowledge, or hope, that solvingall of the smaller problems will result in solving the larger problem. This reasoningstrategy is commonly used in the analytical,problem-oriented approach by veterinarians.in fact a great deal of our System 2 thinkingfor diagnosis utilises this heuristic or its closecousin, the hill-climbing heuristic.

The hill-climbing heuristic differs in thatthere is greater uncertainty about whether thechoice of a smaller subproblem might startyou on your way to solving the larger prob-lem; in other words you do not know whetheror not solving that problem will actually helptowards solving the larger problem, but youare willing to give it a go. We’re sure some ofyou will recognise deliberately using thisstrategy to help reach a diagnosis. it could beargued that the hill-climbing heuristic, andother reasoning strategy heuristics used inSystem 2 thinking, form the central basis forproblem-oriented medicine: the resolution ofnumerous disparate problems into a smallernumber of key problems being the corner-stone of achieving a definitive final diagnosis(or diagnoses when multiple disease process-es coexist in the same patient).

These heuristics are used as reasoning strate-gies because they allow the problem of diagno-sis to be redefined as something that is simplerto achieve.8 With hill climbing you realise it is along way to the top for that elusive diagnosis,but at least you are going up the hill by break-ing the climb into segments; thus progress isbeing made – or you think it is! For example, a cat presented with signs of central nervoussystem (CNS) disease may require a complexseries of investigations (neurological examina-tion, blood tests, cross-sectional imaging, sero-logical tests, cerebrospinal fluid [CSF] collec-tion and analysis, etc) to determine the cause ofthe problem. Your practice policy may be thatsuch cases will be referred to a neurologist.Thus even by making a diagnosis of ‘CNS dis-ease’, you are achieving something and ‘mov-ing up the hill’. Perhaps you could instead electthat, in all such cases, you will send blood offfor non-invasive testing to rule in/out sometreatable neurological diseases (lead poisoning,thiamine deficiency, cryptococcosis) and thenoffer referral for advanced imaging and CSFcollection if there is no response to a 10 dayhigh-dosage course of clindamycin (to coverthe two further treatable possibilities of ascend-ing middle ear infections and toxoplasmosis).You may not reach a definitive diagnosis, butyou are moving forward (definitely hill climb-

Figure 2 Lateral radiograph of the left thoracic limb of an old Abyssinian cat (a). Note themultiple punched out osteolyticlesions. The presumptive diagnosis of the author who was sent theradiograph was multiple myeloma,even though no further lesions couldbe detected in full body radiographsof the cat. On examination of theactual patient (not its radiographs!),there was clear evidence ofsuppurative inflammation, with pusdraining from an open wound (b).The cat improved markedly afterstarting amoxicillin clavulanatetreatment, with healing of the lyticbone lesions (c). Images courtesy of Emma Hughes

Heuristics,

or ‘rules

of thumb’,

are used

by everyone,

every day,

sometimes

knowingly

but often

unknowingly.

A misleading outcome might occur if the anchor is based onan ‘urban myth’ that you were exposed to earlier in life (maybefrom a university lecturer or more experienced colleague who‘got it wrong’). ‘False anchors’ can also occur if you hear anincomplete history or are not told of key physical or laboratoryfindings. In our experience this is not uncommon when casesare referred on to a specialist practice or even ‘handed over’to another veterinarian in the same practice.For example, you might be told to take over management

of the ‘cat fight abscess in cage 4’, only to find later it isn’t atypical polymicrobial anaerobic bite-associated abscess (ie, with canine tooth puncture marks over the lesion), but an unusual infection with a saprophytic pathogen (such asNocardia species). They didn’t tell you the lesion wasn’taccompanied by bite marks, didn’t respond to lancing anddidn’t have a strong foetid anaerobic odour! Another fertile ground for false anchors is radiological

interpretation in isolation (ie, without history or physicalexamination). For example, one of us was sent a radiograph(Figure 2a) that demonstrated multiple lytic lesions, reminis-cent of multiple myeloma. When it was eventually possible toexamine the patient, there was a single large draining sinustract from the affected region (Figure 2b); and (jumping ahead)the cat was cured following a long course of amoxicillinclavulanate (Figure 2c). With the benefit of hindsight, the radiological (and clinical) diagnosis was multifocal bacterialosteomyelitis.

a

c

b



ing!). on the other hand, ‘hillclimbing’ could be misleading ifyour subgoal does not get youcloser to the real diagnosis.

Sometimes hill climbing mayjust be a form of displacementactivity (Figure 3), risking delay-ing a definitive test or interven-tion that might provide a tissueor microbiological diagnosis, ora surgical option for therapy. Forexample, many people resort totesting for feline immunodefi-ciency virus (FiV) in a sick catwhen they are ‘diagnosticallydestitute’, even though it is gen-erally not possible to determinewhether the cat’s FiV status hasany impact on the current dis-ease status. (This, interestingly, is in contrast to the situation with HiV/AidS, where Cd4counts and viral load measurement can bevery informative.)

Progress-monitoring heuristicis there a reasoning strategy heuristic thatcomes into play when you are not making satisfactory progress in solving a case? There is,and it is the so-called progress-monitoringheuristic.9 This heuristic probably kicks in whenhill climbing is just not giving you enoughadvancement, so you decide to use anotherstrategy or to change direction; for example,you consider involvement of another body sys-tem or a new infectious agent, you try a differ-ent imaging modality, or look at the chestinstead of the abdomen (even though signs initially pointed to abdominal involvement).

This can be a very useful strategy when youare using an analytical problem-orientedapproach as it provides a ‘reality check’ forprogress. However, beware the impact of pastexperiences and your current emotionalstate – they can misdirect you as you assessthe state of play (especially if owners areapplying pressure for a rapid diagnosis!). Thedanger is that you may select a new strategytoo early and not base it on an objective eval-uation of the current strategy in use.

Progress monitoring can also come into playwhen a lack of response to treatment makesyou question the presumptive diagnosis.

How do heuristics relateto patterns, illnessscripts, intuition anddiagnostic algorithms?

So we come to see that it is System1’s unconscious use of someheuristics – drawing on patterns ofclinical features and more com-plete illness scripts stored in long-term memory – that underliesmuch of what is described as ‘intuitive’ diagnosis. Intuition isconventionally defined as ‘the abil-ity to acquire knowledge or under-standing without the conscioususe of reason’. This definition mayseem perplexing – how do youknow something, without know-

ing how you know it? It is for this reason thatthe use of intuition in the field of medical rea-soning has long been looked on in a ratherderogatory fashion by some physicians.However, it is now becoming accepted that

experienced and expert clinicians improve theircapacity for diagnosis through intuition bydrawing on information held in long-termmemory in a meaningful and often unconsciousway.10 Information in long-term memory cancome from lectures, published texts, personalexperience, conversations with mentors/colleagues, conferences and seminars. We canperhaps now better understand why expertsoften find it difficult to explain how they cameto a diagnosis, particularly if this involvesseemingly astonishing cognitive leaps. Inessence, it is down to them having numerouspast experiences with that type of disease andcataloguing these effectively in memory, so thatthey can be recalled quickly through certaintriggers, such as specific information or keyclinical or historical findings. There is no doubtthat certain people have a flair for this type ofdiagnostic reasoning, and it helps greatly tohave a retentive memory – and to think like adetective! The great Sherlock Holmes acceptedthe importance of intuition and the fact that itappeared to happen without conscious thought.

Sherlock Holmes on use of intuition

‘It was easier to know it than toexplain why I know it.’ – Sir ArthurConan Doyle (A Study in Scarlet)

‘From long habit the train of thoughts ran soswiftly through my mind that I arrived at theconclusion without being conscious ofintermediate steps.’– Sir Arthur Conan Doyle(The Memoirs of Sherlock Holmes)

Figure 3 When all is chaosand confusion, it can betempting to find something‘worthwhile’ to do. The hill-climbing heuristic employedin a bad way!

On the path to self-discoveryYou have probably come to realise that you do indeed use heuristics indiagnostic reasoning – even though it may not always be clear what type ofheuristics; if you are using them as cognitive biases or reasoning strategies;or when to trust them and when to question them. What is certain is that asyou become more experienced in veterinary practice your use of them willincrease, whether for diagnosing disease through System 1 thinking orhelping to diagnose through strategies employed in System 2 thinking.



Because intuitive diagnoses commonlyinvolve heuristics as cognitive biases, it isimportant consciously to put some safeguardsin place through trained System 2 thinking.These were discussed in Article 2, and aresummarised on the right.

Moreover, after some practice and reflectionon these safeguards, you might be able to better explain to a junior colleague or recentgraduate (or even yourself!) how you came to acertain diagnosis. Hopefully, your junior col-league will quickly realise that you are some-one who has seen lots of cases, learned fromyour mistakes as well as your triumphs, andalso from colleagues/mentors and lifelonglearning strategies involving journals, semi-nars, meetings and case discussions. Followingon from this, it may even be possible to deliber-ately utilise generic illness scripts stored aslong-term memories employed in intuitivediagnoses to teach ‘the art’ of clinical diagnosis.

So-called ‘script theory’, which has beenaround since the 1970s, was used in the lexiconof cognitive psychology to refer to the way inwhich people understood real-world events,usually in an effortless way.11 The medical fra-ternity saw an opportunity to adapt this con-cept to case-based clinical reasoning and giveit some credibility for teaching deliberatelyconstructed generic illness scripts to under-graduates. Schmidt et al divided script-basedintuitive clinical reasoning into generic andinstance illness scripts (see boxes).12

Because of the complexity of many veteri-nary cases, common usage of part or completeillness scripts in System 1 thinking speeds upthe diagnostic process by ‘separating thewheat from the chaff’. The question is, howcan you be sure that it was the (spurious) chaffthat you discarded, rather than the (critical)wheat? To safeguard this process, it is vital tolink illness scripts with working memory viaeffortful System 2 thinking. This will make illness scripts and the recognition of patternsmore effective cognitive devices by ensuringthey are less prone to the omission of data germane to the case.

Safeguards< Slow down and review your reasoning< Ask yourself if you have missed anything< Ask yourself whether the diagnosis could be wrong, and try to consider

other possible diagnoses< Search for evidence that will confirm or refute the plausible diagnosis

Instance illness scripts

Instance (instantiated) illness scripts focus on the retrieval of information aboutindividual patients held in long-term memory. For example, you readily recalla striking case early in your career of a cat being investigated for syncope.After extensive investigation, the final diagnosis turned out to be Dirofiliaimmitis infection. The inference is not that all cats with heartworm disease(dirofilarosis) demonstrate syncope or presyncope, nor that all cats that fainthave heartworm disease. Rather, the instantaneous illness script that comesto mind serves to remind you to ‘check fainting cats for heartworm infection’with a blood antigen test or high resolution echocardiography of the rightventricular outflow tract and main pulmonary arteries.

Generic illness scripts

Generic illness scripts refer to the rapid and often unconsciousretrieval of collective and integrated long-term memories andexperiences of a specific disease. By their very nature, generic illness scriptsstereotype disease conditions and do nothave defined borders, as certain informationmight be shared by several scripts.13

Moreover, generic illness scripts, whenrecalled, really require validation throughengagement of System 2 thinking. Importantquestions to ask are: does this specific casehave the key features of the stereotypicalcase; and, if not, what tests are required toidentify or support those features? For example, some or all of the following

features comprise the generic illness scriptfor insect-bite hypersensitivity (usuallymosquitoes or midges): punctate lesions onor near the nasal planum (papules andpustules initially, but ulcers after the pustulesburst) (Figure 4); similar lesions (generally

less marked) on the pinnae or toes (hence the old nickname ‘ears, nose and toes syndrome’); inflammation (predominantly

eosinophilic) on biopsy or exfoliativecytology; and rapid resolution with or withouttherapy in hospital. Occasionally herpeticdermatitis can present in a similar fashion,affecting comparable anatomical regions (in particular the naso-ocular region) andinvolving eosinophilic inflammation;importantly, lesions do not resolvespontaneously, but do so after the adminis -tration of famciclovir. (A skin biopsy, immuno -histology and PCR analysis will differentiatebetween these possibilities, to the greatsatisfaction of System 2 advocates, thoughat some considerable expense.)By deliberately remembering and

constructing generic illness scripts originallyused in System 1 thinking, they can be used as valuable teaching aids for thenovice practitioner.

Figure 4 Cat with mosquito-bitehypersensitivity. Historically, lesions such asthese were often attributed to autoimmunedisease or the eosinophilic graulomaspectrum of conditions. But following twoseminal publications,14,15 the generic illnessscript made this a commonplace diagnosis inAustralia and other places with warm humidconditions and an abundance of midges andmosquitoes. Whether a black hair coat is partof the generic illness script or an urban mythis unresolved! Courtesy of Ildiko Plaganyi



Another way to try to avoid bias and omis-sion of pertinent data is to employ diagnosticalgorithms. A medical algorithm is any compu-tation, formula, statistical survey, nomogram(graphical representation of relationships) orlook-up table which might be useful in diagnos-ing or managing a disease. diagnostic or clini-cal algorithms, especially in the veterinaryarena, commonly take the form of decisiontrees. They rely on preliminary clinical signs orphysical findings, or groups of these (perhapssome derived from deliberate constructs ofgeneric illness scripts) to further guide investi-gation in order to detect, support or refute adiagnosis.16 in this way, diagnostic algorithmsfollow the problem-oriented, forward-thinkingapproach to diagnosis. Heuristics superficiallyappear to have little role in the construction ofdiagnostic algorithms, although hill-climbingand means-end heuristics are often used as gen-eral-purpose reasoning strategies in algorithms.

diagnostic algorithms, by their very nature,are designed more for complex cases requir-ing comprehensive investigations. Whenmoney is no object, use of such algorithmsusually leads to a definitive answer. Many in

the medical sphere include probabilities as ameasure of uncertainty about a specific diag-nosis. However, it could be argued that algo-rithms are only as good as the premises orhypotheses upon which they are based. if oneof the premises is flawed, then it is possiblethat incorrect diagnoses will be reached. Sinceknowledge is never certain, and basic premis-es or hypotheses are continually being chal-lenged, algorithms require constant updating.Therefore, it is prudent always to check whenthe algorithm was developed!

Veterinary diagnostic algorithms tend todeal in black and white answers, since evi-dence-based probabilities for diagnostic testshave rarely been developed. So, by their verynature, they tend to ignore the ‘shades of grey’that often exist for biological systemsresponding to perturbation (ie, disease). inother words, while algorithms strive for objectivity and accuracy, they do not allow foropen-mindedness and scepticism. This is theshortcoming of algorithms, as these four cornerstones of scientific enquiry work best in concert. Nonetheless, while not suggestingthat clinical algorithms should be used for allcomplex cases or difficult diagnoses, we dourge you to use them when appropriate.Especially when they have been developed byexpert clinicians using the application of evi-dence-based veterinary medicine tempered bytheir own clinical experiences, because thiscan reduce the risk of error and lead to successrates comparable with those achieved by intu-itive veterinary experts.17

Outline of clinical reasoning

Engage in trained System 2 thinking and consider: (i) evidence required to confirm diagnosis, (ii) alternativediagnoses and (iii) whether yourdecisons were influenced by bias

CLINICAL REASONING

SYSTEM 1 THINKING(eg, pattern recognition)

SYSTEM 2 THINKING(eg, analytical, problem-oriented)

Cognitive error needs to be managed

Unconscious use ofcognitive bias heuristics,and patterns and illnessscripts from long-termmemory for intuitivereasoning

Cognitive error needs to be managedConscious use of

working memory,reasoning strategyheuristics and diagnostic algorithms

Consider and confirm hypotheses or premises used in reaching diagnosis(evidence-based). Did you recogniseand manage any cognitive bias inreaching your decisions?

Diagnosticalgorithmsfollow theproblem-oriented,forward-thinking

approach todiagnosis.

Sherlock Holmes on the illness (crime) script

‘As a rule, when I have heard some slight indication of the course of events, I am able to guide myself by

the thousands of other similar cases which occur to mymemory.’ – Sir Arthur Conan Doyle (The Red-Headed League)



The intuitive diagnostician!As veterinarians spend time inpractice, gain experience through adiverse case load and learn ‘tricks’from mentors and colleagues, theones with the right skill set candevelop into intuitive clinicians. Thereally good ones move effortlesslybetween Systems 1 and 2 thinking.Most will use System 1 first (defaultmode). If that doesn’t work, or if an alarm bell goes off in their headwhile working up a difficult case,they then move to System 2, andsometimes back again!

System 1 thinkingdefault system

If no pattern or illness script recognised ...

use System 2

If pattern or illness scriptrecognised ...

safeguard against error by using System 2 to test

Final comments and conclusions

In this article we have suggested that heuristicsare powerful tools in medical reasoning, espe-cially for the experienced clinician. Basically,they have roles in reasoning in both System 1and System 2 thinking: in the former by draw-ing on long-term memories, including patternsand theoretical illness scripts, to assist in intu-itive diagnosis; and in the latter by assisting inreasoning strategies as well as drawing onlong-term memories. It is important to acceptthat heuristics used in System 1 thinking cansometimes mislead, and that some effortfulSystem 2 thinking is required to put checksand balances in place. Even expert clinicianscheck to ensure that they have sufficient evi-dence for their definitive diagnosis and havethought of (and often excluded) possible alter-natives. Moreover, they commonly regard theultimate test (check) of a diagnosis as being theresponse to medical or surgical therapy.If we have any summary advice to impart, it

would be to learn to trust your intuition but at the same time put in place safeguards todiminish the impact of bias and misguidedlogic, whatever their origin. Accept that yourbrain can mislead you! We have an importantadmission to make: we still get cases wrongand we still get confused about how we think.does that alarm us? only when we are not

cognisant of making those mistakes. The start-ing point to being the best clinician you canstems from that awareness.

The thought processes in diagnosis are stillpoorly understood. For that reason, it seemshard work at times. But accepting the uncertain-ty of the process can mean that you canapproach diagnosing difficult cases with a senseof purpose – and at least with some optimism.Remember Sherlock Holmes referring to the‘thrill of the chase’ when you are next mullingover a difficult case – and ensure that you use allyour faculties to sniff out the answer!

AcknowledgementsRichard Malik is supported by the Valentine Charlton Bequestadministered by the Centre for Veterinary Education of theUniversity of Sydney.

FundingThe authors received no financial support for the research, author-ship and/or publication of this article.

Conflict of interestThe authors declared no potential conflicts of interest with respectto the research, authorship and/or publication of this article.

References1 Gigerenzer G and Goldstein dG. Reasoning the fast and frugal

way: models of bounded rationality. Psychol Rev 1996; 103:650–669.

2 Mithen SJ. Thoughtful foragers – a study of prehistoric decisionmaking. Cambridge: Cambridge University Press, 1990, pp 1–289.

3 Eva KW and Norman GR. Heuristics and biases – a biased

perspective on clinical reasoning. Med Educ 2005; 39: 870–872.4 Mcdonald CJ. Medical heuristics: the silent adjudicators of

clinical practice. Ann Intern Med 1996; 124: 56–62.5 Harvey N. Use of heuristics: insights from forecasting

research. Think Reasoning 2007; 13: 5–24.6 Canfield PJ, Whitehead ML, Johnson R, et al. Case-based clinical

reasoning in feline medicine. 2: Managing cognitive error.

J Feline Med Surg 2016; 18: 240–247.7 Epley N and Gilovich T. Putting adjustment back into the

anchoring and adjustment heuristic: differential processing

of self-generated and experimenter-provided anchors. Psychol

Sci 2001; 12: 391–396.8 Eysenck MW and Keane MT. Thinking and reasoning. in:

Cognitive psychology – a student’s handbook. East Sussex:Psychology Press, 2010, pp 470–473.

9 MacGregor JN, ormerod TC and Chronicle EP. Information pro-

cessing and insight: A process model of performance on the nine-

dot and related problems. J Exp Psychol Learn 2001; 27: 176–201.10 Greenhalgh T. Intuition and evidence – uneasy bedfellows.

Brit J Gen Pract 2002; 52: 395–400.11 Gardner H. The mind’s new science: a history of cognitive

revolution. New York: Basic Books, 1987, pp 165–170.12 Schmidt HG, Norman GR and Boshuizen HPA. A cognitive

perspective on medical expertise: theory and implication.

Acad Med 1990; 65: 611–621.13 Charlin B, Boshuizen HPA, Custers EJ, et al. Scripts and clini-

cal reasoning. Med Educ 2007; 41: 1178–1184.14 Wilkinson GT and Bates MJ. A possible further clinical mani-

festation of the feline eosinophilic granuloma complex. J Am

Anim Hosp Assoc 1984; 20: 325–331.15 Mason KV and Evans AG. Mosquito bite-caused eosinophilic

dermatitis in cats. J Am Vet Med Assoc 1991; 198: 2086–2088.16 davies C and Shell L. Common small animal medical diag-

noses: an algorithmic approach. Philadelphia: WB Saunders,2002, pp 1–261.

17 McKenzie BA. Commentary – veterinary clinical decision-

making: cognitive biases, external constraints, and strategies

for improvement. J Am Vet Med Assoc 2014; 244: 271–276.

Available online at jfms.com Reprints and permission: sagepub.co.uk/journalsPermissions.nav

For reuse of images only, contact the corresponding author

case-based clinical reasoning in feline medicine 3: …...418 jfms clinical practice special article...

Documents