Picky Details

This week I took a look at the article Effects of Cholesterol and Inflammation-Sensitive Plasma Proteins on Incidence of Mycardial Infarction and Stroke in Men (downloadable here http://circ.ahajournals.org/).

All too often I find myself reading articles similar to this one, who find a correlation between two things and think to myself "le awesome so glad we figured this out" (or if it isn't totally figured out at least we know a little more than we did before). Then I start to think about the results a little more and the possibilites for outside factors and errors start piling up in my mind. Particularly in health related studies, I find myself asking well what if these people smoke versus people who don't smoke, or work out 7 days a week versus watching CSI 7 days a week, or eat McDonald's for every meal like the guy on Super Size Me? More often than not I find that researchers aren't necessarily ignoring these details but they just don't mention them in their results and I find myself asking well do these relate to a soccer playing 16 year old or a 35 year old code monkey?

Personally I'd say that 7/10 of the articles leave me unhappy with just a little bit of the population details missing. By details I mean I want to know more about the people who were in the study, are we sure that these statements can be applied across the board, or does it only work for qualified individuals with certain risk factors etc. This article in Circulation, did a great job, imho, of separating risk factor groups and determining just how universal their results were. By dividing their subjects in to categories from smokers to non-smokers, (even a subset of how many cigarettes per day) level of physical activity, pre-existing conditions, and current physical health the study can more accurately declare their results, stating that elevated ISP levels, not hypercholesterolemia alone can give men a higher risk of ischemic stroke. This slight differentiation in detail could lead to much more efficient and successful follow up work.

Do you guys think that when looking at studies such as this one there should be a certain number of characteristic groups defined and studied before publishing results, particularly with studies that look at inflammatory factors, their origins, and effects? Is it possible that this would just be another hoop researchers would have to go through in order to get possible life saving results out in to the world or do you think it's more of a precautionary measure that would save a lot of time and trouble from unnecessary treatments?

I think it could be argued that it needs to be thought of on a case to case basis but for the most part definitive guidelines of different characteristics represented per group would usually be more helpful. Especially when you consider, like many inflammatory diseases, that there are so many risk factors and characteristics that can make one person more prone to susceptibility than others.