Application of Knowledge for Osteopathic Medical Practice
Core concepts
Application of knowledge questions integrate basic and clinical science into clinical reasoning. Build illness scripts: risk factors, pathophysiology, clinical findings, diagnosis, and management. Use pretest probability to choose tests and interpret results. Sensitivity rules out disease when negative (SnNOut); specificity rules in disease when positive (SpPIn). Know absolute versus relative risk, attributable risk, NNT, odds ratios, and confidence intervals. A p-value is not the probability that the null hypothesis is true; it reflects compatibility with the null under assumptions. Case-control studies estimate odds ratios, cohort studies estimate relative risk, and randomized trials support causation better than observational designs. For therapy, compare absolute risk reduction and NNT rather than relative risk alone. Distinguish efficacy from effectiveness. Traps include association versus causation, confusing incidence with prevalence, screening bias, lead-time bias, and choosing a test because it is newer rather than because it changes management.
Worked example
A test has sensitivity 90% and specificity 80%. In a low-prevalence population, many positive results are false positives despite good sensitivity. Rule of thumb: use sensitivity to rule out and specificity to rule in, but always update with pretest probability.