Still, I've continued to support medical blogging as a useful academic endeavor, hoping that someday this support would be borne out. When sites like Sermo and Facebook came along, I despaired that more physician opinions were going to be hidden behind walled gardens, available only to select colleagues or friends.
Then, last week, some revelations -- I discovered a member of my department's leadership was blogging, or at least, had commented on a blog. How about that!
The other revelation? Facebook may be the last great hope for academic discussions to flourish on blogs.
This all arose from a pretty academic question, about emergency department implementation of electronic medical records. Does the degree of implementation (full, partial, or none) impact patient wait times in the emergency department?
This is a good question, and has been asked many times by doctors, administrators, vendors, and researchers. And naturally, the answer depends. Some implementations of ED electronic systems have been very successful -- here's a nice example that I helped write. Of course, skeptics can point to a lot of disastrous implementations (Over at EPMonthly last year, Kevin Klauer summarized the skeptic's viewpoints very well, and I've covered some stylistic objections to EMR in the past).
While ED adoption of electronic medical records was always an expensive question, with last year's ARRA's provisions for stimulating health IT spending by tying medicare reimbursements to meaningful use of electronic medical records, this has become a political issue, as well.
Enter the medical blogosphere. Specifically, the WSJ blog post covering this new paper by Furakawa on electronic information systems and emergency department wait times.
Furukawa's retrospective analysis of ED length-of-stay and other metrics was based on a 2006 database of 364 hospital EDs, and a separate assessment of degree of EMR implementation (how many functions had been turned over to electronic systems). He found that the 1.7% of EDs that had the most robust implementation of EMR had 22.4% shorter length-of-stays than those with no or minimal electronic records. This was statistically significant.
Interestingly, he also found a statistically significant increase in time-to-physician (which he calls wait times) in EDs with basic (not full, not minimal) implementation of electronic medical records. This effect was more pronounced among lower acuity patients, but it was present among even the very sick patients.
Got that? Both are interesting findings, inviting a lot of questions. The academic in me wants to dive into how time-to-physician was measured, how variations between sites were normalized, what wait times were like at these centers before partial EMR implementation, how these changes in length-of-stay compare to other interventions or factors in ED crowding, etc. I would love for this to be what discussion centers around, though that kind of discussion remains rare in the blogosphere.
What we got instead was this:
The reports on adversity from HIT are being stifled by the HIT industry. They are not being posted fast enough to convince the White House that it is a collossal waste of money and time to conduct an unregulated experiment using unconsented patients at the behest of the HIT industry and vendors. Tuskegee Project was similar, except for the race factor. They took the downtrodden and subjected them to disease.
With EMR HIT, they are taking the downtrodden (sick patients) and subjecting them and their care to HIT medical devices that have no proof of safety and effectiveness.Wow, huh? If you had told me Tuskegee would come up in a discussion of medical records, I would have assumed the comparison was between staying with paper charting in 2010 to leaving syphilis untreated in 1972. And I would've said that's an awful, hyperbolic comparison that doesn't advance the debate.
But to hear Tuskegee invoked against electronic records? Well, I don't know where to begin, other than to note Godwin's Law states that any web debate will eventually degenerate into one side comparing the other's to Nazism. So I guess we have that to look forward to, and Tuskegee is just a waypoint on that journey. Or maybe this is a calculated attempt to move the Overton window, so that an idea that seems quite reasonable is suddenly perceived as unthinkable...
Another element to this "debate" is that other commenters let anonymous posts like the one above slide without criticism, but when an ED informaticist chimes in with their real name and opinion, they're immediately pressed to disclose conflicts of interest, and cite their sources. That's all well and good, but where's the parity? These demands seem like they're coming from other physicians but oddly enough, when I google these commenters' names I only find... other WSJ blog posts.
Anyway, the poor quality of blog comments has been lamented for years, on many forums. It's just the first time I've seen the ugliness come to my area of research.
It's particularly discouraging, because if trolls and narrow-minded partisans can hijack and debase a blog discussion (and, presumably, discourage participation from more sources) in one of the few areas I've actually studied, what hope is there for learning from the comments on other topics? And has this stifled the development and legitimacy of academic blogging?
I find myself wondering, for the first time, if networks like Facebook might actually be helping foster online discussion, by tying people's online comments elsewhere back to their account.
Facebook founder and CEO Mark Zuckerberg was derided this spring when he described having different online personas as "a lack of integrity." And I understand the critic's perspectives -- there's real value in keeping some characteristics or interests private, and in keeping one's public comments anonymous, or at least, separate from other realms. And I've always liked that it's the discretion of the blogger or publisher to allow comments at all, to let them stay anonymous, or to force commenters to use a real email address or name.
From a different perspective, however, I think it's generally assumed that, if commenters were forced to use real names (or tie their comments to, say, a Facebook account), that the quantity of comments would plummet, but the overall quality would probably rise. My question is: Does this rise in quality offset the decline in quantity? In the past I would've said no, but now I'm not so sure.