No research today… I read some articles yesterday but lacked the patience to write about them yesterday. Life got a little busier than usual this past week. I think I’ll be reserving Fridays for some commentary on what I’ve read, which I’m sure will include the articles you’ve seen me review as well as others that didn’t deserve a post.
So here’s the deal – most people don’t read research. I mean really read it. So, to make things a little more realistic and give a small idea as to how I do it (which isn’t the “right” way, it’s just what I do to save time and to get to the truth as quickly as I can), I’ll explain my process.
Step 1 – pubmed or PLoS search for articles based on key words I’m interested in that day… simplicity is nice. “Gluten”, “Autoimmunity”, “Insulin sensitivity”, and so on.
Step 2 – Choose a sexy title to read about
Step 3 – Skim the Abstract for something interesting
Step 4 – Find the full text. Skip the abstract and intro.
Step 5 – Read the methods. They’re boring but they let you know what the researchers ACTUALLY DID… If we’re being smart, and it’s an RCT or other experiment – then we know the authors should ONLY be commenting on what the results of their study present.
Step 6 – Read the results. See what happened with their experiment.
Step 7 – Make your own conclusions. Skip the discussion and conclusion. This is where most get lazy. This is where most check the discussion, or probably more realistically, just the conclusion. They usually contain bias and give away the authors hopes and biases. “While our study did not support our hypothesis, the current literature suggests that blah blah is still suggested. Greater sample sizes and longer follow up is suggested to come to more accurate conclusions.” Or some other BS. Don’t read it – it’s there to trick you. Read the methods and results and make your own conclusions.
Step 8 – Hope that it was a good experiment. Use logic and intelligence to put the conclusions into context.
Step 9 – Find a new article to read after that.
If authors are abandoning their conclusions and bringing up poor references (Youdo check their references right?), then their word isn’t to be trusted, but you CAN still take the information their experiment created as it was usually designed with the hopes of supporting their hypothesis.
Anyway, I suppose I’m a little frustrated with the run-around you get when reading studies. It would be nice if all the published were the methods and results, but they don’t. Oh well… I suppose it’s the nature of humans.
Stay healthy, Friends!
Dr. Adam Ball