Archive for November, 2005

Criteria for choosing a community Health Information system

Wednesday, November 9th, 2005

Welcome to my eHealth blog. I will be using this space to talk about issues related to converting the U.S. health care system from paper to electronic records. Since you’re reading this, you undoubtedly know that making complete electronic health information and decision support available whenever/wherever needed would save lives, improve care, and save money.

My objective is to help expedite the transition to electronic records, while protecting (maybe even enhancing) privacy.

I am a physician (MD) and also have a PhD in computer science. My entire professional career (now over 3 decades) has been spent working on the effective utilization of computers in health care. I currently am an independent consultant helping organizations and communities to develop & implement health information systems.

Most recently, I was the Senior Advisor, National Health Information Infrastructure, at the U.S. Department of Health and Human Services (2002-5). While there, I succeeded in focusing the attention of the nation and the President on this issue, leading to the creation and funding of the Office of the National Coordinator for Health Information Technology.

In the mid-1990s, I built the successful (and still operational) statewide immunization registry in Oregon. I’ve posted a separate message (below) with more on my background for anyone who’s interested.

Today I’d like to discuss a key issue facing communities wishing to move ahead with health information systems. The idea of such systems is to make all the medical records of a patient in that community available for care anytime. These records are now mostly paper, and every location where a patient gets care only has a piece of the “total” record. Many communities are trying to decide how to move forward with developing these systems, so I’m going to suggest a set of evaluation criteria to help guide the decision-making process.

There are three categories of evaluation criteria that communities should consider in evaluating a potential health information infrastructure:

  • Requirements – does the system do what is needed?
  • Feasibility – can the system be built and operated?
  • Financial – is there a way to pay for the system and sustain it?
  • Let’s start with requirements. To answer this question, the functions of the system (at least at a high level) must be defined. Folks experienced with the development of information systems know that there must be a statement of requirements before you start or you’ll never know when you’re done. In plain English, a requirement is something that you will be able to do AFTER you have implemented a system that you cannot do now.

    There are at least five requirements in this context:

  • 1) privacy
  • 2) transition from paper records
  • 3) access to information
  • 4) incremental steps
  • 5) universal availability
  • Privacy is absolutely critical. Medical information is perhaps the most sensitive personal information there is – everyone should be able to keep this information private and only have it used by clinicians when needed to take care of them. In the context of community health information systems, this means that patients/consumers must control ALL access to their complete records. In my view, this is the only sure way to maintain the trust needed to establish and operate such a system.

    Any electronic records system must provide a reasonable transition from paper records, which are still used widely. At present, only 10-15% of physicians are using electronic health records (or EHRs) in their offices (more on this in a later post). Any new information system must recognize and accommodate to the existence of these paper records. The decision may be to ignore them (which I don’t think is best), but they have to be considered.

    The central requirement for any community health information system is to provide a way to make all the information about a patient available for care when and where needed. To do that, there must be a way to collect and deliver all the accumulated records from the various places where the patient has been (including sites outside the community).

    A community health information system is a large undertaking. Like any other large information system, it is best done in small incremental steps. The “finished system” takes time and it’s best if the gradual development of it results in benefits for more and more people as time goes on.

    Finally, such a system should be available to everyone. I believe participation in such systems should be voluntary. However, no one should be precluded from participating because of the way the system is designed. For example, access to the system via the Internet makes sense, recognizing that 99+% of public libraries provide such access at no charge. Therefore, anyone in the community can utilize the system via the Internet even if they do not have access to it at home. However, for those who cannot use the Internet at home, there may need to be emergency access to certain critical functions by phone.

    The next major evaluation category is feasibility. First, there is user acceptance – will the system satisfy them enough to be used? Next is stakeholder acceptance – closely related to user acceptance. All the stakeholders need to be happy about the system (even if they are not users). For example, while employers may not directly use such a system (except in their individual roles as consumers), they should be comfortable that it is meeting their needs – for example, for improved health care and better accountability. Finally, there is technical feasibility. Has such a system been built before? Does it require new technology? How complex is it (simpler is better)?

    Lastly, financial issues should be considered. How much will it cost to build? Where will the funds come from? Perhaps more importantly, who will pay for the system once it’s built? These questions should be answered very specifically. It is very risky to build a community health information system with a financial sustainability plan that says “it will benefit everyone, so everyone will pay according to the benefits they receive.” While this is a nice principle (and hard to argue with), it is hard to apply after-the-fact. So there’s a need to identify beforehand where the benefits are and who will pay what (and when).

    So if your community is thinking about implementing a health information system, consider using these criteria in the evaluation process as you look at various alternatives. I’ll look forward to your questions and comments about these.

    Do you think there are other questions that should be added to our discussion?

    Next time – Consumer attitudes about electronic health records