Screening Test

1. See Also
   1. Contingency Grid or Cross Tab (includes Statistics Example)
   2. Bayes Theorem (Bayesian Statistics)
   3. Fagan Nomogram
   4. Experimental Error (Experimental Bias)
   5. Lead-Time Bias
   6. Length Bias
   7. Selection Bias
   8. Likelihood Ratio (Positive Likelihood Ratio, Negative Likelihood Ratio)
   9. Number Needed to Screen (Number Needed to Treat, Absolute Risk Reduction, Relative Risk Reduction)
   10. Negative Predictive Value
   11. Positive Predictive Value
   12. Pre-Test Odds or Post-Test Odds
   13. Receiver Operating Characteristic
   14. Test Sensitivity (False Negative Rate)
   15. Test Specificity (False Positive Rate)
   16. U.S. Preventive Services Task Force Recommendations
2. Definitions
   1. Factors affecting Positive Predictive Value
      1. Prevalence of disease
      2. Accuracy of test
3. Types: Bias
   1. Screening Bias
   2. Lead-Time Bias
   3. Length-Time Bias
4. Criteria: Ideal Screening Test
   1. Disease Features
      1. Disease significantly impacts public health
      2. Intermediate probability of disease
      3. Detection occurs before a critical point
         1. Critical point occurs before clinical diagnosis
            1. Screened patient is still asymptomatic
            2. Diagnosis would not otherwise occur this early
            3. Endometrial Cancer's critical point is too late
         2. Critical point occurs in time to affect outcome
            1. Disease must be detected early enough for cure
            2. Lung Cancer's critical point occurs too early
   2. Test Features
      1. Screening Test tolerated by patients
      2. High Test Sensitivity to detect asymptomatic disease
         1. Best quality for Screening Test
         2. Test criterion or threshold is set low to minimize false negatives (at the expense of increased false positives)
      3. High Test Specificity
         1. Best quality for confirmatory test
         2. Test criterion or threshold is set high to minimize false positives (at the expense of increased false negatives)
   3. Screened Population Features
      1. Disease has high enough Prevalence to allow screening
      2. Medical care available if Screening Test positive
      3. Patient willing to undergo further evaluation
5. Evaluation
   1. Assume you know the disease state
      1. Test Sensitivity
         1. Given the patient has the disease,
         2. What is the probability of a true positive test
      2. Test Specificity
         1. Given the patient does not have the disease,
         2. What is the probability of a true negative test
      3. Probability of test given disease
         1. P(test | disease) = Test Sensitivity / Test Specificity
         2. Where P (A | B) = Probability of A given B
   2. Assume you know the test result
      1. Positive Predictive Value (PPV)
         1. Given a positive test,
         2. What is the probability that the patient has the disease
      2. Negative Predictive Value (NPV)
         1. Given a negative test
         2. What is the probability that the patient does not have the disease
      3. Probability of disease given test
         1. P(disease | positive test) = PPV / NPV
         2. P(disease | positive test) via Bayes Theorem = P(positive test | disease) * P(disease) / P(positive test)
            1. where P(positive test | disease) = Test Sensitivity
            2. where P(disease) = Prevalence of disease in the tested cohort
            3. where P(positive test) = Probability of positive test in the tested cohort
6. References
   1. Desai (2014) AMIA Board Review, Clinical Decision Making
   2. Gates (2001) Am Fam Physician 63(3):513-22 [PubMed]
   3. MacLean (1996) Med Clin North Am 80(1):1-14 [PubMed]
   4. Nielsen (1999) Med Clin North Am 83(6):1323-37 [PubMed]