Positive predictive value—the role of specificity

In this short video, you’ll learn how sensitivity and specificity influence the positive predictive value and why specificity has a much greater impact.

If you haven’t done so, please watch our first video on predictive values first!

[00:00:00] Apart from the prevalence, there's also another variable that affects the positive predictive value, in a major way. And that variable is called specificity. Here's an example, let's take our population of 1000 individuals again and this time, let's say the prevalence is 20%. So, 200 have the disease and 800 don't. Let's say the test sensitivity is 70% and the specificity is also 70%. So,

[00:00:30] out of the 200 diseased, 70% or 140 will be picked up by the test, where 60 will be missed. Similarly, 70% or 560 of non-diseased will be picked up, whereas 240 will be falsely classified as diseased. Overall, we have 380 folks who tested positive. 140 true

[00:01:00] positives and 240 false positives. Now, let's calculate the positive predictive value, that's 140 true positives, divided by 380 who tested positive, times 100, equals 37%. So, pretty bad. Now, let's see what happens to the positive predictive value if we change sensitivity to let's say 90%. So, now out of the 200 diseased, we're going to pick up 180 people and we'll miss 20. Nothing changes on the right side since specificity stays the same at 70%. So, overall, we end up with 420 people who test positive.

[00:01:30] What's the positive predictive value? Maybe you want to give it a try and calculate it yourself. Pause the video and come back when you're done. So, the PPV is calculated as 180, divided by 420, times 100, which equals 43%. So, not a big improvement from our initial 37%, right? Now, let's see what happens if we take our initial numbers and change the specificity from 70% to 90%. So, in this

[00:02:00] case, the left side stays the same since sensitivity is left unchanged at our initial 70%. Now, we're correctly diagnosing 90% of non-diseased or 720 people and we're going to get 80 false positives. So, overall, there are 210 people who test positive. The positive predictive value thus 140, divided by 210, times 100, which equals 67%. So, much better than our initial 37%, right?

[00:02:30] And why does specificity have so much more influence on the positive predictive value than sensitivity? Well, because there are many more people in the non-diseased group. Therefore, a 1% change on the right side of the vertical line or the specificity has a much bigger impact than a 1% change on the left side or the sensitivity.