The following commentary appeared a little while ago.
Research data everywhere and not a drop to drink
By gshaw
Created Jan 25 2012 - 1:48pm
Physicians like numbers. Data, double-blind studies, peer-reviewed journal articles, evidence. And they clamor for scientific proof whether the issue is prescribing statins [1] to patients at risk for heart disease or whether the debate at hand is the value of e [2]lectronic health records systems [2], the pros and cons of email communication [3] between docs and patients, the benefits of e-prescribing [4], or the impact of m-health technologies on patient outcomes.
Show me any IT initiative that will affect a physician's workflow, schedule, paycheck, or liability risk and I'll show you a doctor who's calling for evidence that the rewards outweigh the risks.
And since m-health, e-health, connected health, telehealth and data-driven health (et al) are pretty much dead in the water without physician support, researchers are scrambling to deliver it.
The Journal of the American Medical Informatics Association recently published a flurry of such studies, including one that found using an automatic alert system in providers' EHR systems significantly increases the documentation of previously unknown patient problems [5]. Another found that poor EHR implementation can skew quality measures. A third found that some EHRs are lacking in adverse drug event detection [6]. And yet another said they're a good tool for identifying preventative services in order to avoid unnecessary procedures.
But wait, there's more: On any given day you can find a new study that proves this or that about EHRs and other health IT tools. Web-based tools aren't effective for diabetes management [7]. EHRs improve hospital nursing care [8]. EHRs reduce racial disparities [9]. It goes on and on.
You see the problem, here, right? There are thousands--if not hundreds of thousands--of questions about electronic health data and the various tools physicians can use to harness it. By the time researchers finish slicing and dicing data in incremental studies such as these, the EHRs of today will be sitting on a shelf in the Smithsonian ... and we still probably won't be any closer to reaching a consensus about their overall benefits or efficacy.
Lots more here:
There are two threads here that I think we need to tease out. The first regards the need for evidence based practice and the second is the need for real clarity as to just what a particular piece of evidence means and just how far the findings of even a very conclusive and well conducted study can be extrapolated into drawing more general conclusions.
On the first issue we need first to accept that common sense and evidence are not the same thing! We also need to acknowledge that with the costs and risks of health care services rising we need not to be adopting and doing things that just seem right and make apparent sense. We need to devote increasingly scarce resources to those activities that can be shown to really make a difference to health outcomes.
If ever there was an example of the sort of woolly thinking that leads to bad decisions in action it is the thought bubble that has led us to the PCEHR. Never been tried, never been tested but off we go! What nonsense!
On the same point we also cannot go on pretending that e-Health is harm and risk free. There is now a lot of evidence that says that is simply not the case!
On the second issue I would be the first to admit that right now the evidence base for sweeping generalisations in Health IT looks very shaky indeed. There are all sorts of questions around whether an apparently conclusive study conducted at a major referral centre means much for the larger world of more normal facilities and whether what is found in premium implementations is actually seen to be generalizable. It is all too easy to say well it worked there, it makes sense that it does so the outcome will always be true in all situations. This is just not so - as the drug makers will testify. In careful clinical trials a particular therapy is excellent - but put into the hands of the ordinary clinician and a less motivated patient and the outcome can be very different!
I believe this is fundamentally important material that deserves more thought and reflection. It is just not good enough to hope something will work - as I keep being told elsewhere ‘hope is not a strategy’. We need evidence and proof!
David.
p.s.
As a coincidence - just as I finished this blog article this appeared:
http://blogs.crikey.com.au/croakey/2012/01/31/mental-health-funding-well-targeted-or-just-well-meant/
Guess what? The same problem of Government health policy being implemented in an evidence free way!
p.s.
As a coincidence - just as I finished this blog article this appeared:
http://blogs.crikey.com.au/croakey/2012/01/31/mental-health-funding-well-targeted-or-just-well-meant/
Guess what? The same problem of Government health policy being implemented in an evidence free way!
