Is Wikipedia a High Quality Evidence-Based Resource?
The cop-out answer to that question is "sometimes." Wikipedia is a nice resource, but beware, the hidden dangers of unreliability are likely on the upswing.
Pinsker J. The covert world of people trying to edit Wikipedia - for pay. The Atlantic, August 11, 2015. Accessed from http://www.theatlantic.com/business/archive/2015/08/wikipedia-editors-for-pay/393926/.
I came across the above article in the Atlantic about a year ago, and it's taken me all this time to come up with an opening to include it in my blog. Folks who read Evidence eMended regularly know that I will frequently put in a link to Wikipedia as a way to give interested readers an avenue to more detailed explanations. Why would someone who preaches the value of evidence-based medicine also refer to a web site that anyone (and it really is anyone) can edit?
Pinsker's essay really caught my eye; he very convincingly described instances where hired guns attempted to alter medical content in Wikipedia to present products in a more favorable light. The primary example he used involved the orthopedic procedures of kyphoplasty and vertebroplasty, likely equivalent to placebo in efficacy for treatment of vertebral compression fractures associated with osteoporosis according to the best data available. In 2013, the medical editor for this Wikipedia page noticed that someone had wanted to edit the comment that the procedures' effectiveness are "controversial," replacing that word with the phrase "well documented and studied." That's a big change in the meaning of the sentence.
Long story short, some dogged detective work by the editor revealed that the source of the edit was an employee of an orthopedic hardware company that sells devices used for kyphoplasties. This employee subsequently complained to the editor that the Wikipedia posting was "scaring prospective patients and insurance companies;" although not explicitly stated by the employee, of course this bad publicity jeopardizes the company's profit margin.
Wikipedia has sought to prevent such hired gun tactics by changing its terms of use to discourage edits by individuals with serious conflict of interest, but the extent to which this has decreased the practice is unknown. My own (previously unwritten) policy in use of Wikipedia in Evidence eMended is that, for medical content, I'll check out the veracity of the post on at least one other high-quality web site, such as a professional society practice guideline or via original articles in the scientific literature.
In the past few weeks, I deliberately tried to use Wikipedia a bit more often than I usually do, so I'd have some illustrations for this posting.
For example, on August 9 I linked to a page about comparative effectiveness research. That's a topic I try to follow fairly closely, so it didn't require any extra PubMed searching. I thought the Wikipedia post was concise, accurate, and didn't require a doctorate in statistics to understand the main points. On August 16 I linked to Wikipedia's take on fee-for-service pricing, but this time I did check with a few other sources for accuracy. Again, the page was concise and didn't require an MBA to understand.
My favorite recent Wikipedia link, however, was on July 26. This illustrates my other use of Wikipedia, for the multitude of nonscientific, opinion discussions of pop culture. This one in particular was to see if anyone actually pays attention to the phrases I use (in this case, plagiarize) from popular culture. I'm sure all the Beach Boys aficionados got the reference without my Wikipedia link. Sometimes guys just wanna have fun while they blog.
So, use Wikipedia if you want, but, like any other information source, be sure to verify the accuracy if the content is important to you. That's a key principle of evidence-based medicine.
