Hi, I’ve been reading a lot of literature on the keto diet. Whenever I find a good article I like to check the references which brings me to the studies that were done. My question is about the stats that are used.
I keep seeing this format.
-0.81 (95% CI: 0.62, 1.05; P = 0.11)
Can someone break this down and let me know what these numbers stand for and what they actually mean. I think the first part is the risk ratio but after that I’m not sure.
Usually the first number is the estimated effect size, so for example a parameter decreased by 0.81 units. The CI is confidence interval. A 95% CI is typical. So the authors are 95% confident the real parameter value is between 0.62 and 1.05. P = 0.11 is your p-value. P-value is the percent likelihood of detecting an effect/change/relationship if none was truly present. So there is a 11% chance of detecting that parameter as important if it really wasn’t important… If that makes sense. 
Ok thanks, So what do these numbers mean to a layperson. I found some info on the p value. Basically it has to be equal to or less than 0.05 or it’s not statistically significant. But what about the other numbers? I’ve googled away on this but I can’t find anything I really understand
In ecology we often use p-values of 0.10 because there’s a lot of variables uncontrolled for. In my opinion, a p-value of .11 is pretty good unless the sample size was massive.
If you can copy and paste some of the results on here and I can help you interpret them. What were the parameters they were testing?
Ok Thanks. here is one of the ones I was looking at. Same stats as the one I first posted but with the rest of the results
“ During 5-23 y of follow-up of 347,747 subjects, 11,006 developed CHD or stroke. Intake of saturated fat was not associated with an increased risk of CHD, stroke, or CVD. The pooled relative risk estimates that compared extreme quantiles of saturated fat intake were 1.07 (95% CI: 0.96, 1.19; P = 0.22) for CHD, 0.81 (95% CI: 0.62, 1.05; P = 0.11) for stroke, and 1.00 (95% CI: 0.89, 1.11; P = 0.95) for CVD. Consideration of age, sex, and study quality did not change the results.”
Ok here is how I would interpret this to a layperson: A risk estimate of 1 means no or little difference in risk.
Saturated fat risk of CHD: No risk observed.
High p-value and the risk estimate is close to 1, 95% confident its between 0.96 to 1.19.
Saturated fat risk of stroke: No risk observed.
P-value is high for such a large sample size, but there’s also a lot of variable not controlled for. However, estimated risk is less than 1 and between .62 and 1.05.
Saturated fat risk of CVD: No risk observed.
Extremely high p-value and the estimated risk estimate is 1.
Does this help?