How Do We Get Better Research Results?

We learn about diseases and illnesses from research studies. Some focus on treatment (clinical trials), while others study groups of people (observational studies). Unfortunately, many studies are unclear, wrong, or can’t be easily transferred into everyday medical practice.

Study designs actually matter. They determine whether a research study helps real patients or just asks esoteric questions to further careers or make profits.

There are many ways to improve research studies. Let’s start with observational studies.

Observational Studies

There is one very simple concept to remember about observational studies:

Correlation does not mean causation!

In other words, just because A and B fit together (are related, associated, or correlated with each other), it doesn’t mean that A caused B to happen.

HealthNewReview.org posted a great article about this, called Observational Studies – Does The Language Fit The Evidence? – Association Versus Causation.

There are several types of observational studies. This chart explains the pros and cons of each type.

F. Perry Wilson suggested a (better?) way to design observational studies in a recent MedPage Today post. A regular problem with these studies is that real life includes many things (factors) that researchers don’t include in studies:

“There are always other confounding factors that we didn’t think about, or we didn’t measure.” – F. Perry Wilson, MD

He suggests a method that helps build in factors that traditional study designs often overlook. Wilson did a great job of explaining both of these concepts in his article:

• Randomized clinical trials (RCTs), known as the gold standard of clinical research, and
• Instrumental variable analysis (IVs), which has not been popular, even though it has been around for over 80 years.

Study designs need to answer questions that apply to real people. This is why some of us work with study sponsors to infuse the patient perspective into study design and implementation. In case you are interested, here are some examples in articles and presentations. Just let me know if you’d like to join us.

There are lots of issues in trial design – this is just one. Future post material – aren’t you excited?! Please share your thoughts or add resources on observational studies in the comments. Thanks!

All content © 2016 by Deborah Collyar unless otherwise specified. All rights reserved. Permission is granted to use short quotes provided a link back to this page and proper attribution is given to me as the original author.