Increased Disease Prevalence Increases Pretest Probability
The correct answer is D: increased disease prevalence would increase the pretest probability. This is a fundamental principle of Bayesian clinical reasoning where disease prevalence directly determines the pretest probability before any diagnostic testing is performed 1.
Understanding the Relationship Between Prevalence and Diagnostic Test Characteristics
Likelihood Ratios Are Independent of Prevalence
- Likelihood ratios (LRs) remain stable across different disease prevalences, which is precisely what makes them superior to predictive values for assessing test performance 2.
- LRs combine sensitivity and specificity to provide an omnibus index of test performance that is numerically stable regardless of population prevalence 2.
- The equation: Pretest odds × LR = Posttest odds demonstrates that LRs serve as multipliers independent of the starting prevalence 3, 2.
Important caveat: While theoretically independent, empirical studies have shown that sensitivity, specificity, and consequently LRs can vary with prevalence for tests based on continuous underlying traits, due to differences in trait distribution relative to diagnostic cutpoints 4. However, this is a measurement artifact rather than a mathematical property.
Predictive Values Are Highly Dependent on Prevalence
Positive predictive value (PPV) increases with higher disease prevalence 5:
- PPV reflects the proportion of positive test results that are true positives
- As prevalence rises, more positive tests represent true disease, increasing PPV
- Therefore, statement C (claiming increased prevalence reduces PPV) is incorrect
Negative predictive value (NPV) decreases with higher disease prevalence 5:
- NPV reflects the proportion of negative test results that are true negatives
- As prevalence increases, more negative tests represent false negatives (missed cases), decreasing NPV
- Therefore, statement A (claiming decreased prevalence reduces NPV) is backwards—decreased prevalence actually increases NPV
Prevalence Equals Pretest Probability
Disease prevalence in the population directly determines pretest probability 1, 3:
- Pretest probability represents the likelihood of disease before testing
- In Bayesian analysis, this prior probability p(θ) incorporates existing knowledge about disease frequency 1
- When applying tests to individual patients, clinicians adjust this population prevalence based on patient-specific characteristics 3