OBJ - Introduction to Clinical Reasoning

Question 1

List different strategies that physicians use to solve clinical problems

Answer 1

1. Hypothetico-deductive
   - Arrive at the diagnosis by considering each potential Dx in isolation without comparing/contrasting with other potential Dx
   - Inefficient, redundant, but can be accurate
   - Used by novice learners
2. Algorithm or schema
   - Arrive at diagnosis through a series of prioritized decision points or steps
   - Work through a series of “yes”/“no” ?s with branch points
3. Schema “plus”
   - You create + tie concepts together–linking words (or meaningful connections)
   - Causes, manifested by, resulting in, is treated by, such as
   - Studies have shown helps organize thoughts, understanding
4. Key finding (key feature or key clue)
   - Limited differential (splinter hemorrhage)
5. Heuristics (cognitive shortcuts)
   - Consider CNS infection in all patients who present with fever and mental status changes
   - Perform arthrocentesis in all patients presenting with acute monoarticular arthritis
6. Probabilistic (Bayesian)
   - Applying statistical principals to patient encounters
   - Experts usually do
7. Pattern recognition - Diagnosis “at a glance”

Question 2

Describe how to construct a problem list

Answer 2

1) Prioritize & interpret various abnormal symptoms
2) Be as specific as possible to help refine differential Dx
3) 2 Strategies:

Splitting Splits/lists each symptom & relevant other historical information
- Characteristically complete
- Gives max opportunity to identify all the key findings and correct diagnosis (House)
- Can get lost in the trees (problems) and
miss the forest (dx)
- Great for: complicated presentations & unfamiliar/rare scenerios

Lumping lumps symptoms, exam findings, and lab abnormalities into a single problem (Occam’s razor)

• Emphasizes pathophysiological grouping of problems & creativity
• Premature dx/over simplification of a case
• May omit key details
• Great for: familiar/classic presentations, unifying dx hypotheses generated

Question 3

Define sensitivity and specificity

Answer 3

Sensitivity:

• Not influenced by disease prevalence
• Proportion of patients with the disease who have a positive test
• Math = true positives/all patients with disease (true positives + false negatives)
• SNOUT -> Rule Out

Specificity:

• Not influenced by disease prevalence
• Proportion of patients in whom the disease is absent who have a negative test
• Math = true negatives/all patients without disease (true negatives + false positives)
• Can use to calculate false positive rate = 1-specificity
• SPIN-> Rule in

Question 4

Calculate positive and negative predictive values

Answer 4

Positive predictive values IS influenced by disease prevalence
• The proportion of patients with a positive test who have the disease
• Math = true positives/all positive tests (True Positives + False Positives)
• Number without false positives

Negative predictive values IS influenced by disease prevalence
• The proportion of patients with a negative test who do not have the disease
• Math = true negatives/all negative tests ((True Negatives + False Negatives)
• Number without false negatives