# Is the correlation in betwen childhood trauma and addiction stronger than the correlation in between diabetes and obesity?

I came across this image on the internet, which makes the claim, "There is a stronger link between childhood trauma and addiction, than there is between obesity and diabetes. Two thirds of addicts report being abused as children."

Doing some research to try and establish notability I found an interview from Johann Hari:

for every traumatic event that happens to a child, they’re two to four times more likely to grow up to be an injecting drug user—so, stronger link between childhood trauma and addiction than there is between obesity and diabetes. This is really powerful.

He cites an "Adverse Childhood Experiences Study."

So, the claims that have been made are:

1) There is a correlation inbetween childhood trauma and addiction

2) The correlation in betwen childhood trauma and addiction is stronger than the correlation in between diabetes and obesity

3) For every traumatic event that happens to a child, they're two to four times more likely to grow up to be an injecting drug user

4) Two-thirds of addicts report being abused as children

• Is there a correlation? Sure. Is it a signal that stands out from the noise: most people are abused in some way as children, a large proportion of adults abuse some drug or other thing like sex, gambling, etc? Claim #3 would seem to be a mathematical impossibility: after a few doublings or quadruplings, the chance would be greater than 100%. The only way to prove causation would be to do a controlled experiment, right?
– user29285
Dec 15, 2015 at 0:46
• @nocomprende Yep, claim 2 has to be carefully defined, as in: Addicts are x times more likely to have been abused than the gen. pop., while people with diabetes are y times more likely to be obese than the gen. pop., where x > y. Claim 3, I was aware of that, I assumed he meant the upper limit was close to 100%, which seems unlikely, but... possible? I'm still interested how he got those numbers, is he misquoting some stats from somewhere or did he just make them up? I was hoping someone more familiar with the info than I am could answer. Dec 15, 2015 at 17:23
• This article seems pretty related to this topic even if it doesn't directly address the various claims: med.stanford.edu/news/all-news/2011/06/…
– user35545
Sep 1, 2016 at 1:13
• There is a stronger link between A and B than between C and D? Who cares about apples versus oranges? The statement is only relevant and provable/falsifiable if there are also links between (A or B) and (C or D).
– user22865
Sep 1, 2016 at 14:40