Back to dashboard
Randy Neil Biostats 17: USMLE Step 1 - Viewer Requested Biostats Questions #4
Episode Notes
USMLE purpose: Convert diagnostic test word problems into a 2×2 table and calculate PPV/NPV, sensitivity, or specificity.
How to use this page
- Draw the box immediately.
- Put the real disease status across the top and test result down the side.
- Fill totals first, then calculate missing cells.
- For “test positive and actually has disease,” think positive predictive value .
Episode metadata
| Field | Details |
| Episode | Randy Neil Biostats 17 |
| Topic | Viewer requested questions #4 |
| Runtime | 7 min |
| Published | 2021-03-10 |
| Source | Open YouTube video |
One-liner
If a patient tests positive and the question asks the probability they actually have disease , the metric is PPV = TP / (TP + FP) .
2×2 Diagnostic Box
| Disease + | Disease - | |
| Test + | True positive | False positive |
| Test - | False negative | True negative |
Formula table
| Metric | Formula | Question wording |
| Sensitivity | TP / (TP + FN) | Among disease+, how many test positive? |
| Specificity | TN / (TN + FP) | Among disease-, how many test negative? |
| PPV | TP / (TP + FP) | Positive test → truly diseased? |
| NPV | TN / (TN + FN) | Negative test → truly disease-free? |
Worked example
If 100 patients have HSV2 and 90 test positive, then:
- TP = 90
- FN = 10
- If 400 do not have HSV2 and 20 test positive, then:
- FP = 20
- TN = 380
PPV = 90 / (90 + 20) = 90 / 110 ≈ 82%
Anki-style rapid recall
What does PPV ask?
If the test is positive, what is the probability the patient truly has disease?
What does NPV ask?
If the test is negative, what is the probability the patient truly does not have disease?
What is the denominator for PPV?
All positive tests: TP + FP.
Practice Questions
Question 1
A test is positive in 90 patients with disease and 20 patients without disease. What is the PPV?
- A) 90 / 100
- B) 90 / 110
- C) 380 / 400
- D) 380 / 390
Reveal answer & explanation
Answer: B) 90 / 110
PPV = true positives / all positive tests = TP / (TP + FP).
Quick Reference Summary
| If question says... | Think | Formula |
| Positive test, actually sick? | PPV | TP / TP+FP |
| Negative test, actually well? | NPV | TN / TN+FN |
| Disease present, test catches it? | Sensitivity | TP / TP+FN |
| Disease absent, test rules out? | Specificity | TN / TN+FP |