Electronic healthcare records (EHRs) and public policy

The Obama transition team has asked for comments about health policy. Their stated focus is on "ordinary folks" questions like “What would you most like to see fixed?” and “How would you like us to communicate with you about health policy?” Those are important, and in another post I've focused on a specific group of those, namely the needs of contractors and entrepreneurs who don't have ordinary company jobs.

In this post I'd like to extend the discussion a little further. Most especially, I'd like to discuss IT-related areas about which we might be able to offer some insight, whether or not each of us has much experience directly dealing with healthcare apps.

One subject stands out. The Obama team seems committed to accelerating the adoption of electronic healthcare records, which surely can be very good things. A discussion seems called for on at least three fronts:

  1. What benefits should an EHR initiative strive to achieve?
  2. What pitfalls should an EHR initiative seek to avoid?
  3. What specifically should the government do to promote the benefits of EHRs while avoiding the pitfalls?

Let's start with some benefits (the first two points are quoted from and current as of a post I made back in 2005 on my old Computerworld blog):

  1. The cost savings from electronic health records can be huge. The US government estimate is $300 billion annually.
  2. So can life savings. Those are harder to estimate. Studies of deaths due to medical error tend to focus on things that health records wouldn't necessarily avert. Still, estimates in the tens of thousands of lives saved annually in the US alone are defensible. In fact, I agree with them. Numbers in that range for the entire world are an obvious slam dunk.
  3. Where there are life savings, there surely are also great improvements in the quality of care beyond life and death. Indeed, the benefits might be even greater in "softer" quality of life issues such as pain management, where a more "holistic" view of the patient can be needed to balance comfort vs. addiction.
  4. Improvements in patient convenience could also be immense. How would you like to be able to get a vaccine, or see a new specialist, or get while your usual doctor is unavailable (whether due to your vacation or hers), without memorizing and rewriting your whole medical history?? I sure would.
  5. It's hard to conceive of a successful transition to personalized medicine without there being electronic health records to help measure its effects.

The potential pitfalls are immense as well. EHRs represent one of the greatest data integration challenges ever. And the privacy risks are daunting, as are more general questions of data ownership. My recommendations therefore include:

  1. Careful attention is needed to privacy legislation. In particular, the current focus on possession of information needs to be supplemented by one that limits discriminatory or other harmful information use when data exchange is necessary or desirable. I outlined some ideas on this point a while back, but they almost seem to have gotten more attention outside the US (e.g., at the BBC) than domestically.
  2. Government should centralize its own involvement with EHR technology. The Veterans Administration and every service of the Defense Department operate EHRs. Insurance and research arms of the Federal Government also are involved with EHRs. It's easy to over-design an EHR system, so an abstract imposition of standards could very easily backfire. But consolidating EHR acquisitions in one place, so that at least there's some sensitivity to meeting broader future-proofing goals? That seems like a wise compromise, between the extremes of strategic over-engineering on the one hand, and tactical chaos on the other.
  3. EHR-based medical research should be aggressively pursued. I occasionally hear of some interesting-sounding project to, for example, text mine health records in a quest for insight. RemedyMD is building a whole business plan around using EHRs intelligently to improve health outcomes. But this is all scattershot. If technical, regulatory, and organizational barriers could be overcome, vastly more could surely be done. And by the way -- thinking about what research to pursue will help inform decisions about what EHR system designs to put in place.

Related links

What do you think about health care in the US? Please post your comments here, or submit them directly to http://www.change.gov! More ideas about how to participate in the discussion are outlined in my overall healthcare discussion introduction post.

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.

Copyright © 2008 IDG Communications, Inc.