"Your Test Came Back Positive": How Worried Should You Really Be?

A test that is "99% accurate" comes back positive, and anxiety spikes. The natural conclusion is that you are 99% likely to have the condition. That conclusion is usually wrong, and understanding why is among the most valuable ideas in everyday statistics.

Imagine 100,000 people

Take a hypothetical condition affecting 1 in 1,000 people, and a test that catches 99% of true cases (sensitivity) and correctly clears 95% of healthy people (specificity). Among 100,000 people, 100 are sick and 99 of them test positive. Of the 99,900 healthy people, 5% test positive anyway, which is 4,995 false alarms.

Now ask the right question: among everyone who tests positive, how many are sick? That is 99 out of 5,094, or about 1.9%. The result is positive, yet the chance of truly having the condition is closer to 2% than 99%. The two events probability calculator and the Venn diagram tool reproduce this by treating "sick" and "positive" as overlapping events.

Base rates change the story

Same test, different population. If the condition affects 10% of those tested, say because they have symptoms or high-risk exposure, then 9,900 of 10,000 sick people test positive, and so do 4,500 of 90,000 healthy people. A positive result now means about a 69% chance of being truly sick.

Nothing changed about the test. The base rate changed.

Odds give another view. The positive likelihood ratio here is 0.99 divided by 0.05, about 19.8. Multiply it by the prior odds of 1 to 999 and the posterior odds are roughly 0.02, a probability near 2%. An odds ratio calculator helps summarize how strongly a result and a condition are tied together in a two-by-two table.

Why a second test helps

Repeating the test can move the needle dramatically. If errors were independent, a second positive result would lift the probability from about 2% to about 28%. Real tests often share errors, which is why doctors usually confirm with a different method rather than repeating the same one.

This explains why screening everybody for rare conditions generates many false alarms, and why guidelines weigh age and risk factors. It also explains why a negative result for a rare condition is highly reassuring, since the base rate is already low. That is also why public-health programs often begin with higher-risk groups, where the base rate is higher and each positive result means more.

Reading the numbers on the page

Test reports quote sensitivity and specificity. Sensitivity asks: if you are sick, how likely is a positive? Specificity asks: if you are healthy, how likely is a negative? Neither answers the question you actually care about after a result, which is the predictive value, and that depends on how common the condition is. When you read a performance claim, look for who was tested to produce it.

Statisticians organize everything in a two-by-two table of test result against true condition. Our first example fills it with 99 true positives, 4,995 false positives, 1 missed case and 94,905 true negatives. Once the table exists, every probability is a simple division. The chi-square test of independence or the Fisher exact test can then check whether the result and the condition are linked.

The same table shows the reassuring side. Only 1 of 94,906 negative results is a miss, so a negative result here is about 99.999% reliable.

A positive result is a reason to investigate, not a diagnosis. This article is general education, not medical advice, and real results belong in a conversation with your doctor.