# Thread: Relative risk

1. ## Relative risk

The following comes from "Applied Medical Statistics Using SAS" 114-115. In it:

p1 occurs when a disease is present when a risk factor is present
p2 occurs when a disease is present when a risk factor is not present
p3 occurs when disease is absent and risk factor is present
p4 occurs when diesase is absent and risk factor is absent.

They define relative risk as (p1/(p2+p4))/(p2/(p1+p3))

It seems to me that it should be (p1/(p1+p3))/(p2/(p2+p4))

Which is what I found in other sources. Anyone know?

2. ## Re: Relative risk

Originally Posted by noetsi
The following comes from "Applied Medical Statistics Using SAS" 114-115. In it:

p1 occurs when a disease is present when a risk factor is present
p2 occurs when a disease is present when a risk factor is not present
p3 occurs when disease is absent and risk factor is present
p4 occurs when diesase is absent and risk factor is absent.

They define relative risk as (p1/(p2+p4))/(p2/(p1+p3))

It seems to me that it should be (p1/(p1+p3))/(p2/(p2+p4))

Which is what I found in other sources. Anyone know?
I believe you are right and the text is mistaken. It doesn't make sense to have completely different terms in the numerators and denominators of the proportions.

3. ## Re: Relative risk

I found several links that agree with my understanding. I am hoping that someone might actually have the book I am citing and explain that I am misunderstanding the authors But it looks like a mistake in the text. One that really baffled me at first.

4. ## Re: Relative risk

It looks like a simple typo, an easy mistake to make when writing an expression involving a lot of similar terms like that.

5. ## Re: Relative risk

yeah but given that it is presumably being read by medical researchers (and given the mistakes that have been identified in existing medical research) it is a bit concerning.

"Yes several studies have clearly shown cyanide is positively associated with lower cancer rates"

Well that could be the case....

6. ## Re: Relative risk

that is a great topic ....thank uuuuuuuuuuuuu

7. ## Re: Relative risk

Don't forget those people that like to hang out in caves in colorado that have low levels of radioactivity.

Yeah, the respective p should be in the denominator along with its counterpart. I always remember incidence/prevalence given knowledge of temporality are the part / part + counter part, then odds equal part / counterpart.

8. ## Re: Relative risk

Does anyone know how I could send the scan here? I wanted to let others see if I am simply misreading what they are saying or there is reall a mistake in the text.

9. ## Re: Relative risk

Upload the scanned image to a site like imgur.com or imageshack.com, and then share that link here, either in IMG tags or by just pasting the URL.

10. ## Re: Relative risk

You can also attach it to the post directly.

11. ## Re: Relative risk

Perhaps they meant to say RR = P1(P2+P4)/P2(P1+P3), and just divided by accident.

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts