PCCR I: Clin Reasoning

Clinical Reasoning

“puzzle” to a pt

central pt of physician competence (heart of clinical practice) “complex, multidimensional”

Dual Reasoning Process

Syst 1: Intuitive (recog!)

• mental shortcuts/ automatic!!

Syst 2: Hypothetico-Deductive (Rational/New)

• analytical approach via deliberate&controlled through process

Anchoring Bias

tend to perceptually lock onto salient features in pt initial presentation too early in diag process and failure to adjust in light of later info
locks onto singular sx when other red flags raised

Availability Heuristic

Assump that diag is more likely, or more freq occurring (if comes readily to mind)
sev pt w/ appendicitis → consider same for nxt w/ acute right lower quad abd pain

Confirmation Bias

seek supportive evidence for diag at exclusion of more persuasive info refuting it
makes presumptive diag of URI in well-appearing pt w/ cough, rhinorrhea, fever despite asymm chest wall

Diagnostic Momentum

Prioritizing diag made by prior clinicians, discounting evidence of alt explanations

don’t conside MI in pt recently diag w/ acid reflux in setting of sim sx

Framing Effect

Interpretation of info is influenced heavily by way in which info is presented (framed)

pt freq visits for asthma exacerbation, clinician fails to explore structure forces that drive med adherence and explore alt causes of current exacerbation

Representation Error

Failure to take prevalence into account when esti probability of diag

see more of geriatric places diverticular bleed high on differential when eval rectal bleed for adol pt

Visceral Bias

Visceral arousal (neg&pos feelings toward pt) leads to poor diag decisions

assumes unhoused pt no able to manage complicated tx plan, prescribes simpler/less optimal w/o discussing options

Common biases

Availability (easily remembered diag)

Base rate neglect (pursuing “zebras)

Representativeness (ignore atyp features inconsistent w/ favored diag)

Confirmation bias (seeking data to confirm)

Premature closure (stopping diag process too soon)

Rules for Good Decision Making

Slow Down

Aware of base rate of disease for items on differential

Consider what data is relevant

Actively seek alt diagnoses

Ask Q to disprove

Remember you are often wrong!

Differential Diagnosis List

appli of clinical reasoning to sort through findings and create list of possible causes for pt issue

• likely diag/critical!

Working Diagnosis

diag that best matches your pt;s issue at that time

• defined as precisely as data allow

• may be leading diag OR still hypothesis

• foundation for devising tx plan

Dev Differential Framework

Anatomic (Upper abd pain)

Organ/syst (SOB→ what cardiac causes… head→toe)

Physiologic: (cough → what patho cause?)

take key pts ot narrow (opposing descriptors to compare and contrast)

Tom G. Prince, MD, Psychiatrist, General Hospital

Toxin/Trauma

Musculoskeletal/rheumatologic

GI

Pulm

Renal

infectious

Neuro

Cardiovasc

Endocrine

Metabolic/genetic

Derm

Psych

GU/Gyn

Hemtatologic

VINDICATE (differential)

Vascular

Infectious

Neoplastic

Drug related

Inflm/Idopth/Iatrogenic

Congential

Autoimm/Allergic

Trauma/Toxic

Endocrine/metabolic

Ranking Diffferential Diag

Possibilitstic (#1): consider all known causes equally likely and simul test for all (not useful)

Probabilistic: consider disorders more likely (highest pretest prob)

Prognostic: consider most serious 1st

Pragmatic: consider diag most responsive to tx 1st

Prevalence

total burden of existing cases of disease in a population during a specific pd of time

• “how many ppl have disease rn”

• Chronic, incurable diseases have high prevalence vs. short-lived diseases have low prevalence

Incidence

rate of new, newly occurring cases of a disease in a population during a specific pd of time

• “how many new cases are happening?”

• helpful for understanding disease causation and outbreaks

Sensitivity

percentage of pt who have true positive

• HIGH= low chance of false neg

  • Neg likely true neg

• SnOUT

Specificity

percentage of pt who have a true neg

• HIGH= low change false pos

  • POS likely true pos

• SpPIN

Probability

prob that disease is present b4 further testing is done

• Clinical Decision Rule (CDR): use of docu tool (typ list of criteria)

Test Threshold

Probability below which diag is so unlikely you exclude it w/o req further testing

Treatment Threshold

Probability above which the diag is so likely you treat the pt w/o further testing

Posttest pobability

when a test is so helpful it pushes you past one of the thresholds

Validity

does the test accurately identify whether a pt has a disease?

• gold standard (reference) test → best measure of whether pt has a disease

Positive Predictive Test (PPV)

probability that person with a pos test has the disease

Negative Predictive Test (NPV)

probability that a person with a neg test doesn’t have the disease

Likelihood Ratio

likelihood that a given test result would occur in a pt w/ the disease

VS the likelihood that the same result would occur in a pt w/o the disease

• “is it really useful”

Plan

Disposition

• outpt?

• inpt? (ICU, step down, tele, med surg)

Further work-up

Consult/referrals

Treatment

• non-pharmaceutical (therapeutic/suppoortive)

• pharmacological measures

Follow-up

Patient edu/health promotion

Medical Decision Making (MDM)

written record of thought processes and actions a clinician take when diag and tx a pt (summary of clinical reasoning)

• explains thinking → down to cause of prob

• mor detailed summary statement (+clin reasoning)

Critically Appraising Clinical Evidence

EBM in PPAP

3 Basic Q

• Results valid?

• What are results?

• How can you apply to results to patient care (generalizable)