In a study, ROC (Receiver Operating Characteristic) curve analysis of two diagnostic tests yields AUC values of 0.92 and 0.67 respectively. What can be correctly inferred?
- A Test 1 is more specific than Test 2
- B Test 1 has better overall discriminatory ability than Test 2 ✓
- C Test 1 has higher sensitivity at every cut-off compared to Test 2
- D An AUC of 0.67 indicates the test performs worse than chance
Explanation
The Area Under the ROC Curve (AUC) is an overall measure of a diagnostic test's discriminatory ability across all possible cut-off thresholds. An AUC of 0.92 indicates excellent discrimination (near 1.0), while 0.67 is poor to fair (AUC 0.5 = no better than chance, not 0.67). AUC does not directly indicate sensitivity or specificity at any single threshold. Test 1 overall performs better, though at specific cut-offs, test 2 could theoretically be more sensitive or specific.
Reference: Park's Textbook of Preventive and Social Medicine, 27th ed.
High-yield for: NEET PGINI-CETNExTFMGEUSMLEPLABMRCP
Written and medically reviewed by the StethoPrep medical team.