11a – Clinical Diagnosis I Flashcards
What are the 2 ways to define normal?
- Abnormal as unusual
- Abnormal as associated with disease
Abnormal as unusual: 2 ‘types’
- Gaussian: mean +/- 2SD
- Percentile: lower or upper 95%=normal
- *BOTH assume all diseases have same prevalence
- *Leads to ‘diagnosis of non-disease’
“Diagnosis of non-disease”
- 95% of normal subjects fall within the reference range for the test
- 5% of normal do NOT
o 2.5% on either end - *the only normal animals are the ones who haven’t been tested enough yet
Abnormal as associated with disease: 1 ‘type’
- Diagnostic: comparison with a GOLD STANDARD
- *need clinical judgement
What are most diagnoses based on a combination of? (3)
- Signs
- Symptoms
- Tests
- *ruling in and out disease becomes an assessment of PROBABILITIES
What are the 3 possible actions you will eventually have to choose from (in regards to probability of disease)?
- Do nothing
- Get more info (test or response to treatment)
- Treat without obtaining more information
*Generally depends on PROBABILITY OF DISEASE
What might a diagnostic test include?
- Any technique that differentiates healthy from diseased individuals OR between different diseases
- Ex. x-rays, stethoscope
What is accuracy?
- Degree of AGREEMENT between ESTIMATED value and the TRUE value
- QUALITY of test
o Validity (lack of bias)
o Reproducibility (precision or repeatability)
What is the ‘equation’ for accuracy?
- Accuracy = validity + reliability
What is validity?
- Ability to measure what it is SUPPOSED to measure without being influenced by other sources of SYSTEMATIC ERRORS
- *if valid=unbiased
o Does NO ensure accuracy - Not always repeatable
What is reliability?
- Tendency to give the SAME results on REPEATED measures of the same sample
When does a reliable test gives repeatable results?
- Over time
- Between locations
- Between populations
What are some sources of false positives and negative results? (KNOW!)
- Laboratory error
- Improper sample handling
- Recording errors
What are some laboratory errors?
- Depends on both analytical accuracy and precision
- Can vary between labs or within labs
- Does the lab have a recognized quality assurance/control in place?
Doing nasal swaps to detect BRD, importance of timing between sample and culturing
- If more than 3 days after collection=VERY POOR CULTURE RESULTS
What are some sources of false negative results? (KNOW!)
- Improper timing of test
- Wrong sample
- Natural or induced tolerance (ex. BVD)
- Non-specific inhibitors
What are some sources of false positive results? (KNOW!)
- Group cross-reactions
- Cross contamination (ex. Johne’s disease)
How should the accuracy of any diagnostic test be established?
- ‘BLIND’ comparison to an independent and valid criterion for infection or disease status
- *GOLD STANDARD
What are examples of diagnostic tests that are accurate (‘gold standard’)?
- CULTURE of organism
- POST-MORTEM EXAMINATION
- Biopsy
- Long-term follow-up
What are pathognomonic tests?
- Absolute predictor of disease or disease agent
- Can have FALSE negatives
What are some examples of a pathognomonic test?
- Culture of MAP (Johne’s disease bacteria)
- Culture of T. foetus
- Culture of Salmonella
What are surrogate tests?
- Detect SECONDARY changes that will HOPEFULLY predict the presence or absence of disease or the disease agent
- Can have FALSE NEGATIVES and FALSE POSITIVES
What are some examples of surrogate tests?
- Serology
- Serum chemistry
What are the 2 pieces of info when selecting and interpreting diagnostic tests?
- Diagnostic VALIDITY
o SENSITIVITY
o SPECIFICITY - Need to understand our test subject
o Prevalence of the disease in source population OR pre-test probability that patient has disease
What are the sources of validity?
- Laboratory
- Test manufacturer
What are the sources to understand our test subject?
- Signalment
- History
- Clinical examination
- Published literature
- Clinical judgement
How do you make a 2x2 to use for diagnostic test interpretation?
- Top: Disease present (A, C), disease absent (B, D)
- Left: Positive (A, B), negative (C, D)
What is the cutoff point in the lab?
- Most tests/machines now give us a quantitative value
o We need to divide them into positives and negatives (tend to OVERLAP) - *use the cutoff: ‘positive’ vs. ‘negative’
o You will get some false positives and false negatives
What is sensitivity?
- Proportion of subjects with the disease who have a POSITVE test
- Indicates how good a test is at detecting disease (positives)
- =1-false negative
- SnNout
SnNout
- When using tests with VERY HIGH SENSITIVITY, negative results help to RULE-OUT DISEASES
- *more sensitive=less false negatives
What is the equation of sensitivity using a 2x2 table?
- =true positive divided by (true positives + false negatives)
- *=true positives/all WITH disease
What is specificity?
- Proportion of subjects WITHOUT the disease who have a negative test
- *indicates how good the test is at identifying the non-disease
- =1-false positive rate
SpPin
- When using test with VERY HIGH SPECIFICITY, positive results help to RULE-IN DISEASE
- *more specificity=less false positives
What is the equation of specificity using a 2x2 table?
- =true negatives divided by (true negatives + false positives)
- *=true negatives divided by all WITHOUT disease
What will the cut-off determine?
- SENSITIVITY and SPECIFICITY of the diagnostic test
o Usually assume them to be CONSTANT
How are sensitivity and specificity related?
- INVERSELY RELATED
Usually sensitivity and specificity are constant attribute of a diagnostic test, when might they change?
- stage of the disease OR
- physiological status of the animal
- Ex. Johne’s disease (slowly progresses over time)
Johne’s disease as an example of specificity and sensitivity changing
- Lower serological cutoffs in younger
- *sensitivity changes
- Specificity does NOT change
What is true prevalence?
- Proportion of population who have the infection under study at one point in time
- *don’t usually know it
What is true prevalence equation from a 2x2 table?
- DISEASE POSTIVE animals (a+c) divided by ALL ANIMALS (n)
How do you figure pretest probability?
- Start with disease prevalence
- Refine to local population
- Refine to population your serve
- Refine according to patients presentation
- Add in results of history and exam
- Also consider you OWN THRESHOLD FOR TESTING
What is apparent prevalence and what is it’s equation?
- *what we normally have (get it from test results)
- All TEST POSTIVE (a+b) divided by ALL ANIMALS (n)
What are the 3 aspects we can use to determine the PREDICTIVE VALUE of our screening test?
- Sensitivity
- Specificity
- Prevalence
What is positive predictive value? (EXAM)
- More useful in clinic
- Proportion of patient with positive test results who HAVE the target disorder
- Affected by sensitivity, specificity, and prevalence
What is the positive predictive value (PPV) equation from a 2x2 table?
- PPV = true positives (a) divided by those who TESTED POSITIVE (a+b)
What is negative predictive value?
- Proportion with negative test results who DON’T have the target disorder
- Affected by sensitivity, specific and prevalence
What is the negative predictive value (NPV) equation from a 2x2 table?
- NPV = true negatives (d) divided by those who TESTED NEGATIVE (c+d)
When looking at the sensitivity and specificity on a 2x2 table, which ‘direction’ are you going?
- GOING DOWN (left: sensitivity) AND UP (right: specificity)
When looking at PPV and NPV on a 2x2 table which ‘direction are you going?
- ACROSS THE TABLE (left to right)
