Den här sidan finns på svenska. Byt till svenska
Definition — what does that mean?
Of the people who have the condition, the share the test catches.
In Swedish: Sensitivitet
Se
The share of the people who have the condition that the test catches: true positives divided by everyone with the condition, true positives plus false negatives.
Se=0.9
Se equals 90 divided by (90 plus 10), equals 90 divided by 100, equals 0.9
You are given
TP, the true positives, are the people with the condition whom the test flags; FN, the false negatives, are the people with the condition it misses. The denominator is everyone who has the condition, and the sensitivity is the share of them the test finds.
Of 100 people with a condition, a test flags 90 and misses 10: Se = 90 / (90 + 10) = 0.90, or 90%. Turned round, a test with a sensitivity of 0.80 that missed 10 people found 0.80 · 10 / 0.20 = 40, as the card above works out.
| Condition present | Condition absent | |
|---|---|---|
| Test positive | TP | FP |
| Test negative | FN | TN |
Sensitivity reads down the first column. Specificity reads down the second: TN / (TN + FP), the share of healthy people the test clears. The two describe the test and do not change with how common the condition is.
A sensitivity of 90% says the test rarely misses a case. It says nothing about what a positive result means for the person holding it: that is the positive predictive value, TP / (TP + FP), which reads across the first row and depends on how many of the people tested had the condition. A test that finds 90% of cases can still be wrong in most of its positives when the condition is rare.
What is a good sensitivity?
High enough that a missed case is rare for the purpose: a screening test for a serious disease is set
for sensitivity, at the cost of more false alarms.
Sensitivity or recall?
The same number. Machine learning calls it recall, or the true positive rate.
Where do I get the interval?
The sensitivity and specificity calculator gives a 95% Wilson interval from the counts.
| Solve for | Rearranged | With the numbers | Comes to |
|---|---|---|---|
| Se | TPTP | 9090 | 0.9 |
| TP | SeFN1 | 0.9·101 | 90 |
| FN | TP(1 | 90(1 | 10 |
Last reviewed October 20, 2026 ·How pages are reviewed
Statistics and probabilityAll formulas
Copy the code into your page where the formula card should appear. It loads from lovetocalc.com, so it stays current on its own — all we ask is that the credit line comes with it.