01 May Being a Critical Consumer of Medical Research: Understanding Association Versus Causation
You hear the rooster crow and then you see the sun rise. The next day, you hear the rooster crow again and the sun rises again. Days go by and this happens again and again. A young child may assume that the rooster is causing the sun to rise. As adults, and savvy students of science, we know better. The rooster crowing did not cause the sun to rise. The rooster crowing is associated with the sun rising. This is an old and often cited story to illustrate the difference between association and causation. And it’s a crucial topic to revisit when reviewing new research.
Most SUDC research studies to date are considered observational, which means they look at the information gathered from children who died unexplained deaths and compare them to other groups. While this information is crucial to our learning about SUDC, the information gathered can only show potential associations, not causation.
Association (or correlation) is when researchers find a relationship between two or more variables. They may be coincidental, effected by other unknown variable(s), and/or with further research be proven false or causative./p>
Causation, in contrast, means one variable caused the direct effect in the other. Proving causation is a high bar to achieve and must be able to be reproduced to be proven as fact.
The best type of research to attempt to prove causation are randomized clinical trials where a group of similar individuals are randomly assigned to two groups where one group is given an intervention while the other gets nothing. The outcomes of both groups are then compared. For example, a group of similar people with the same disease are divided into two groups, with one being given a medication being researched while the other is given a placebo pill. Neither group would be aware of which group they were in – the one being given the medication or the one who was not.
By nature, we all hold biases that are both conscious and subconscious to us. We can also hold deeply emotional beliefs based in fact, fiction, our personal experience or a combination of these. By learning to be critical consumers of science and more conscious of those biases, we can help ourselves to more accurately understand the natural world we live in.
Here are some tips:
First, be careful not to conclude the meaning and quality of any research publication by the abstract alone. This is the brief summary of the article that is listed at the beginning of the paper. Read the full article before forming your opinion. Research publications usually follow the same general format – abstract, background, methods, results, and then discussion. All sections are important.
Second, be a critical reader of research by channeling your “inner detective.” Recognize your biases and try to put them aside. Does the article prove to you that their approach was reasonable and their methods based in sound science?
Be skeptical. Statistics can be misleading. Watch Sanne Blauw’s Tedtalk linked below for an eye-opening discussion. Also, consider the journal and whether it includes a stringent review process in order for the research to be published. Contact the author directly if you have a follow-up question. Their contact information is usually listed in the article.
You can locate published scientific articles by searching by the author, article and/or journal on pubmed.com “PubMed comprises over 29 million citations for biomedical literature from MEDLINE, life science journals, and online books. PubMed citations and abstracts include the fields of biomedicine and health, covering portions of the life sciences, behavioral sciences, chemical sciences, and bioengineering. PubMed also provides access to additional relevant web sites and links to the other NCBI molecular biology resources. PubMed is a free resource that is developed and maintained by the National Center for Biotechnology Information (NCBI), at the U.S. National Library of Medicine (NLM), located at the National Institutes of Health (NIH).” (Pubmed.com)
To date, there are over 11,000 published references on SIDS and infant death but less than 30 on SUDC.
Third, consider if the methods make sense? The discussion can never consider every question a reader might ask, but is there anything else in the data that is not discussed? And why might that be?
Fourth, consider the limitations of the paper. The discussion should also describe those limitations, which is one of the most important elements. Review them carefully as they usually put the data in a more realistic light. We do not live in an ideal world and barriers to research into SUDC are many.
And finally, if you have questions about SUDC research, email [email protected] for assistance.
Want to learn more? The below videos give great overviews of these concepts- try them out!
Online Educational Resources:
Critical Appraisal of Evidence (Good info on Bias)
How to defend yourself against misleading statistics in the news | Sanne Blauw | TEDxMaastricht
The danger of mixing up causality and correlation: Ionica Smeets at TEDxDelft
https://www.students4bestevidence.net/association-is-not-the-same-as-causation/