Don’t you love it when people who don’t know better think that they know better, and then they end up
Some people will tell you with a straight face that simple statistical analyses are better than complex ones that account for all sorts of different possible confounders. Not only that, but they’ll tell you that you should adopt public health policy based on simple comparisons between two groups. Let me tell you why this is dangerous… Why you should consider the universe surrounding your data.
One of the big lessons that I learned first while getting my master’s degree in epidemiology and now that I’m
Remember Dr. Brian S. Hooker? The PhD who published a seriously flawed — and now retracted — study in which he said that the MMR vaccine caused autism in African American male boys? Yeah, that one. Well, it appears that he has been busy trying to form a case for the legal action he has pending in the vaccine court. This time, he and a group of friends went on one heck of a fishing expedition into the Vaccine Safety Datalink project to see what would come up. What did come up was yet another seriously flawed “study.” I put the word “study” in quotes because, as an epidemiologist, I cannot call this a study. I really can’t. So let’s analyze “A Dose-Response Relationship between Organic Mercury Exposure from Thimerosal-Containing Vaccines and Neurodevelopmental Disorders” by Geier et al (including Hooker), published in the International Journal of Environmental Research and Public Health. Like Jack the Ripper would, let’s take this apart
Epidemiologists have to make sense out of the noise, out of the seemingly random. But what is random?
Most studies carried out in the research of diseases and medications is carried out at the population level. But the findings of such studies may not always apply to the individual in front of you. So you need to take a few things into account.
Doing research for this biostats paper has taught me a few surprising things about suicide.
It’s almost formulaic, isn’t it? In almost every narrative of heroism, the hero finds him or herself in a very
If you’ll indulge me, I’d like to write down some things I learned from my biostats exam in an attempt
My lovely wife and I delivered a presentation (more like a chat) today at the annual conference of the American Mental Health Counselors Association. It was a one-hour presentation on the use (and abuse) of research studies in mental health settings. Coming from an infectious disease background, I felt a little like a fish (not out of water, but) in a different kind of water.
So I did my best to explain some basic biostatistics stuff and how research studies are designed and conducted and why some studies are better than others. As we all learned from the Wakefield fraud, a case series is not necessarily a good design to draw conclusions about causality. Because the practice of mental health counseling is moving more and more towards demanding that all interventions (or as many as possible) be “evidence-based,” I thought it was important to present to the participants what we mean by “evidence” and where that evidence comes from.
There were a few attendees at the beginning, but people trickled down as time went by. This was definitely a different kind of audience than what I’m used to. You’ll see that there was a lot of back-and-forth with a few of the participants, and there were plenty of interruptions. I kind of liked it, actually.
So why “tag-team”? Because my lovely wife introduced the talk and also contributed to the presentation with her perspective on the subject. She is finishing up her master’s degree in mental health counseling. She’s one smart cookie, my wife… Mostly because she married me.
The sound recording of the presentation follows, and you can download a PDF of the presentation by clicking here.