Portable Electronic Medical Records

I have been meaning to do a long post on EMRs for a while. After reading a recent post on Ars Technica on this issue, I decided this is as good of a time as any.

Although Mr. Gitlin readily admits that EMRs are unlikely to solve the problems of “inefficiency” in the medical community within the US, he claims it will decrease cost in the end. He disdainfully alludes to “high cost” in the US system without ever questioning why the costs are so high besides noting that our focus is generally on emergent and not preventive care. Now we can argue this point to death, but if you’d like an enlightened point of view on this subject, I suggest you read any number of PandaBear, M.D.‘s blog posts. I’ll leave this one lie.

My next issue was the fact that Gitlin is actually making two very distinct arguments for EMRs without differentiating the two. First, he argues that EMRs are a perfect way to organize an office. They free the workplace of excess paperwork and can make an efficient way to keep track of “billable” services, something very important to physicians. This much I can definitely support. There are considerations, but I think in the long run, physicians will be better off doing more on computers and less on paper. At the same time he alludes to EMR portability, which is a whole separate can of worms.

The long and short of it is, people need to decide what they want. Currently HIPAA dictates that medical records are on lock-down, and the only people who have access to them are are you and those whom you designate. These “others” could be someone like a spouse or a parent or another physician. I think preserving this privacy is (at least mildly) important from a patient’s perspective. After all, it’s not everyone’s business what your latest test results are or what diseases you’ve been diagnosed with or what medications you’re on. However, from a health care professional’s perspective, this is a frustrating impediment. It does matter to them what your test results were last Monday in the E.R. It’s helping dictate your treatment. Acting like electronic medical records are going to magically make that red tape junkie HIPAA disappear is naive. They might make it easier to transport the information once approval has been given, though.

If people are treating their medical records with that kind of security, then the security vulnerabilities exposed by making things electronic are significant. Even the most secure computer systems are vulnerable to attacks, and considering that many of these systems will be running on Microsoft platforms, there is an increased risk to any data on entire computer networks. If the secretary opens a bad email attachment, is it going to worm its way into the EMR database server and start uploading all the records to someone else’s computer? My point is not that the type of security necessary to run such a system is impossible. Just that it will be a lot of work for everyone, including patients. If a patient needs to “grant access” to their health care providers, they are going to be the gateway into administering their electronic medical record. This means strong passwords, which will probably need to be changed frequently. Are they going to want to do that? How about health care providers? Sure most hospitals have a (probably sub-standard) IT department, but what about your average medical practice? Are they going to be able to employ an IT professional (or pay for the temporary services of one) to set up and maintain these records? All of this is simply going to add cost and overhead to a community that is already overly criticized for how “inefficiently” it works.

The business model that will probably end up working is one where large companies are in charge of the records and medical practices pay fees for the use of their services. In other words, Dr. Jones pays $X every month to have his patients’ health records stored online by a company. When he needs to view the patient’s records, he simply logs in from his office computers (or even from home) and downloads the information. If he has new test results, they would be uploaded. This actually could be a fairly profitable market. It’s basically what’s being set up by Google. For this to be successful, they would still need cooperation from the patient, however, and there would still be security issues. While there would be IT professionals in charge of keeping the data secure, the centralization of data would make it more of a target for criminals. There probably aren’t very many people who would try to break in to Dr. Jones’ patient EMR database since it’s only for a small number of patients. If a large company was hosting hundreds of thousands of patient records, it’s more of a target.

My biggest issue with Gitlin’s argument is that he claims portable EMRs will eliminate (or even significantly reduce) the amount of duplicate testing that is performed and thus save everyone money. See this is an issue that people have to deal with concerning health care in the US. Doctors are skilled professionals. If someone comes to them (especially a specialist) with a problem, they’re going to want their own x-rays and CT scans, not ones from another facility. That’s not to say that they shouldn’t be using them, but considering there’s no incentive for them to work any differently, I highly doubt whether implementing portable EMRs is going to change things. After all, today doctors can get x-rays and CT scans from other institutions if their patients say it’s ok. They just don’t use them.

Gitlin did address my main concern with EMR portability, which is a less concrete aspect: the standards. See right now, there are no standards set in the US for electronic medical records. There’s no “right way” for information to be stored in EMRs, so if you want to transmit information from one doctor’s office to another, they probably need to be using the same program on their end as you are on your end. This is a really bad way to deal with any sort of information. The government has laid some loose guidelines, but nothing really telling people what should be done. They want the market to sort it out. The problem is businesses are all going to try to come up with a proprietary format that will gain a significant market share. This way, everyone has to use their software and their databases. In other words, all the businesses are competing to see who’s going to “win” this race to control everyone’s medical records, and by “winning” the market, they will control the standard and thus the future of the industry. This is a “great” way to form an industry from a business perspective (lock everyone into your format) but a horrible way from an end user’s perspective (doctors and patients). It would be really great if we could nip this issue in the bud now, before any one company has a market share. This way there won’t be a fight to get one big company to relinquish it’s power. Plus, it would be great if EMRs used an open standard so we can continue to have cross-platform competition in the future. We can learn a lot from the mistakes of the past on issues like this.

Look for a future post on FOSS operating systems and EMR programs to tackle some of the issues I brought up here.

8 thoughts on “Portable Electronic Medical Records

  1. Pingback: ICMCC Articles » Blog Archive » Portable Electronic Medical Records

  2. Do you think doctors (even specialists) would be more inclined to use other facilities’ records if those records were available as part of a pervasive, user-friendly computer system in their own facility? If bringing up yesterday’s X-ray from another facility is so easy and painless and the quality is approximately as good, would specialists still want their own?

    And what do you think is the best way to encourage the development and adoption of an open standard for medical records (which, I agree, would be great). Who has the knowledge, both technical and medical, to put something like that together?

    Great site, found you on the trackback list for the Google Health post, next to my own blog.

  3. To answer your first question, I think that was kind of the point I was getting at (in a circuitous manner). I think the amount of technical knowledge and responsibility for a physician to keep up his/her own records in a publicly accessible format is astronomical. It’s much too big of a task for your average physician’s office to take on, even as part of a large practice. In other words, yes, they would want to use other companies services, but more importantly they’d HAVE to. This leaves the responsibility in the hands of companies, whose biggest interests may not be that of the public good. Things like open standards will likely take a back seat to proprietary formats unless a good company is doing the developing. Of course, to be completely pessimistic about the issue, it really doesn’t matter if an “open standard” is developed first. That’ll just mean a proprietary format will have to market itself to large hospitals more aggressively and convince a large number of people to use their product so that their format can be a standard too. I’m alluding to the current debate concerning ODF and OOXML, an “open standards” not unlike what we’re talking about.

    Do I think that all (or most) specialists would use “good” tests from other facilities? I can’t say for sure. Probably not. The simple reasoning is likely because they require more information to treat the disease than to diagnose it. But more important is the threat of legal action in the medical community by not covering your butt on every possible aspect of treatment coupled with the fact that there is no incentive for them to cut costs. If these 2 factors were mitigated, then, yes, I do believe doctors would want to decrease costs for their patients.

    My other thought is what types of records would be exchanged between physicians? I think portable EMRs would be very beneficial for providing specialists with a baseline of information about their patients, including a thorough history, pertinent prescription information, and loads of other “background”. But as far as treating a current condition, there probably isn’t going to be a whole lot of test results worth sharing beyond informing a primary care physician of a specialists’ (or ER physicians’) findings. When any physician is treating a patient, they’re going to want “current” results on their patient, and that definition can vary depending on the patients’ condition. Thus, any results they get are likely to be out-of-date, and they will need new ones.

    What do I think is the best way to encourage development of open standards? Know who’s running your business. There are plenty of folks who would be superb candidates to help shape the future of this industry. Take the folks at GNUMed and DebianMed, for example. You need people who are familiar with open standards and who understand why they’re important. Otherwise, to corporate executives, they just seem like a way to throw profits down the drain. There seem to be two bottom lines in this debate: creating well organized, cross-platform, compatible standards and lining the pockets of you or your shareholders. Basically you need a good group of software engineers, physicians, and other health care providers to team up in this effort. And of course looking to the government for support wouldn’t be a bad idea, especially considering the public good :)

    I hope I’ve answered all your questions. Let me know if you have more or if I’ve missed the point on something.

  4. Thanks for the reply! You did answer my questions – the first one was somewhat intended to be a clarifying/can-you-elaborate? type of question anyway.

    And those are interesting links you posted – it seems like they are having trouble getting a foothold though. Does that sound accurate to you?

    Also enjoyed Naked CSS day. :-)

  5. I think that sounds accurate, but since I posted, I found a few more. It appears Red Hat (leaders of the Enterprise Linux field) has a dedicated Healthcare division. IBM also has an open-source health care framework called Eclipse. I had read about this a few months ago but had forgotten about it.

    A major problem with adoption is the same as that of adopting other free software for business purposes. GNUmed and Debian-Med are free, open-souce, community-supported software. Companies are leery to adopt community-supported software because they don’t have anyone to complain to if something doesn’t work the way they want it to. Since only larger hospitals with good profit margins can afford things like EMR services, enterprise companies (either free or proprietary) are far more likely to win support from these facilities. Unfortunately, proprietary companies will almost always win out. Free software, either community-supported or enterprise, is a better choice for everyone: large hospitals, small community clinics, and even private doctor’s offices.

    A business model like Canonical and Ubuntu would be very beneficial to many low-income hospitals as a way to improve their hospital systems, since the software is free of charge, and enterprise-level support is optional. Larger income facilities subscribing to enterprise-level support would be helping improve the software for everyone in the health care field. Low income hospitals like Stroger (Cook County Hospital) here in Chicago, would be able to implement EMR’s with little or no cost. This idea lies at the heart of free software. Once the software is written, there’s no real reason to charge for the software itself because its already done. Companies can charge for support or to create a custom installation tailored to the facility’s specific needs, but when a “new and improved” edition comes out, they won’t be “stuck” in the spot of having to budget in money for new software licenses. Meanwhile, any facility is able to utilize the latest and greatest software available to the health care community.

    Unfortunately, for the large hospitals this “level playing field” is not desirable. They want to be the only hospital with the latest and greatest, so you have to come to them. Unfortunately this is detrimental to the public hospitals and the people who go to them.

  6. Pingback: Encephalosponge » Now with Identicons

Leave a Reply