How Being Human Messes Up Medical Research

Click here to read an original op-ed from the TED speaker who inspired this post and watch the TEDTalk below.

Sometimes, as I'm talking to patients and their families, I realize that I have no idea what I'm talking about. Really.

The big catchword in medicine these days is "evidence-based." We are supposed to make our decisions about diagnoses and treatments based on research studies; we shouldn't do anything unless there is "evidence" to back us up.

But here's the really big problem: lots and lots of research never gets published. And not because it isn't good and important research, either. It doesn't get published because, well, we are human.

We like good news -- like that Medicine X cures Disease Y. That makes us feel happy and hopeful and powerful. But even more than that, we like it when research shows that something does something. Reading that Medicine X does nothing for Disease Y, well, that sounds boring, sort of why-bother? But it's not boring and we should bother -- because if the people who find that Medicine X works publish their studies, and the ones who find that it doesn't don't, we end up with a really skewed view of Medicine X.

And, as Ben Goldacre eloquently describes in his TEDTalk, that happens all the time. Half of all clinical trials don't get published, he says -- and ones with "positive findings" (the ones that show something does something) are twice as likely to get published. So when I'm using evidence to make my medical decisions, well, I'm missing a whole bunch of it.

It's not like people aren't aware of this problem. They are. That's why they've created, a place where everybody who even starts a clinical trial is supposed to register. And journals theoretically shouldn't publish any trial that wasn't registered. But this doesn't work... because we are human. We don't always do as we're told, researchers and journals alike. And even if they've registered their study, some researchers don't bother to even try to publish studies in which nothing happened -- they figure nobody will be interested. Not cool enough.

I suppose I could go to and find out about all the research that got started and do the math and ask the important questions... but I don't have time for that. Some days, I barely have time to get to the bathroom.

If we are going to fix this problem -- and it is a real problem, one that could and does cost lives -- we are going to need real change in how we publish and think about medical research.

In this new digital age, we don't need paper journals -- and the culture of journals is part of the problem. There is the competition for cachet, the way that editors choose which of the many submissions they will publish, the role of the media and our love of grabby all makes it more likely that the "positive" studies will be chosen and publicized.

What if we just had databases? As studies were done, they could be peer-reviewed, and if deemed to be good (i.e. appropriately done) studies, they could be edited for clarity and entered into a free, searchable online database. All those people who used to work for the journals could perhaps create reports, ones that anyone could subscribe to via RSS feed, that had information like: "As of April 1, Z percent of studies show that Medicine X helps Disease Y, Q percent show that it makes it worse, and H percent show no effect." Then we'd have real evidence to work with.

There are, of course, two possibly insurmountable problems with this idea. First, someone would have to pay for it. Second, there would be much less in the way of glory for individual researchers. Might just be a non-starter, even if it could save lives.

Changing how we think about medical research is easier. We could start that now, by understanding that the people who do and publish research are fallible. They are human. Evidence is crucial, and we must never stop our quest for it, but we can't forget that it's impossible to know everything, that there are always more sides to the story, and that by definition, medical research is ever-evolving.

We doctors wouldn't be able to speak with the same confidence and certainty, of course. And that would not only be hard for doctors, but for patients, too. When it comes to our health, to matters of life and death, we want certainty... because we are human. But certainty is something we are never going to get.

To get the most out of medical research, it's going to take honesty, humility and a willingness to create systems that work for the greater good. Maybe we can do that -- and maybe we can't. After all, we're human.

Ideas are not set in stone. When exposed to thoughtful people, they morph and adapt into their most potent form. TEDWeekends will highlight some of today's most intriguing ideas and allow them to develop in real time through your voice! Tweet #TEDWeekends to share your perspective or email to learn about future weekend's ideas to contribute as a writer.