Obama Proposes Digital Health Records 563
An anonymous reader writes "'President-elect Barack Obama, as part of the effort to revive the economy, has proposed a massive effort to modernize health care by making all health records standardized and electronic.' The plan includes having all conventional records converted to digital within 5 years. Independent studies are fixing this cost somewhere in the range of $75 to $100 Billion, with most of the money going to paying and training technical staff to work on the conversion. Early government estimates are showing 212,000 jobs could be created by this plan."
Creating jobs? (Score:3, Informative)
> Early government estimates are showing 212,000 jobs could be created by this plan."
Uhhh, what? Adding computer automation generally _decreases_ employment - that's the idea anyway.
Part time jobs maybe? If so, that seems pretty important to mention.
Maury
Re:stupid question but..... (Score:2, Informative)
They are doing this. Sort of. Electronic Medical Records are being developed all over the place. The thing is, this being private industry, just about everybody is doing it their own way. There is no standard. There are existing standards for data interchange, but there are no standards for electronic records.
If Obama's plan pans out, all these companies that have spent all this money on all these different systems are going to have to spend a bunch more money switching to the standard. They're get Gov't assistance, of course, but they still need to train their people on the new stuff so soon after they were trained on the proprietary systems.
Re:stupid question but..... (Score:5, Informative)
The idea is sound, but very difficult to properly implement. You'd think with all the benefits, healthcare providors would be clamoring to make the switch (some already have), but there are a lot of hidden problems associated with digital records.
I'm a healthcare technician in the USAF, where the DoD has already implemented a system called AHLTA. When a patient presents for a doctor's visit, all the screening, labs, tests, orders, prescriptions, and physician's notes are entered into the system, where they can be referred to easily for future visits. No need to store thousands of paper records, or train records techs to pull them and locate the appropriate exams.
Problems we experience: Privacy. It's difficult to ascertain just what records which type of doctor should be seeing, so right now we basically have a system where any variety of doctor or technician can see any variety of a particular patient's records (except Mental Health and STD visits, which are accessible only by password by default). This situation makes some patients rather uncomfortable.
Data load. That's a lot of records. And it requires a lot of trained technicians to keep track of it. And it requires frequent audits to ensure the information is current, and has not been illegally accessed. And the system has to communicate and exchange information with several other (often outdated) systems.
Server outages. When we have one, the clinic is virtually paralyzed. We can't refer to the patient's paper record for reliable case history, because the system was implemented five years ago. There won't be any recent records to refer to.
Good Old Boys. The transition has been difficult chiefly because, let's face it: doctors just don't wanna have to go to all the trouble of learning a whole new computer system. It's easier for them to scribble some notes on a sheet of paper, and stick it away in a paper record and be done with it.
There are clear advantages, but it just seems like we're not quite far enough along to handle such a system for just DoD personnel, let alone every single man, woman, and child in the US. A five-year plan is just not feasible from where I'm standing.
24% (Score:5, Informative)
Re:stupid question but..... (Score:4, Informative)
About two years ago I came down with pneumonia. I thought it was the flu and so did the doctor at the urgent care clinic. After two weeks when I didn't get better they took X-rays and found that I was really in bad shape. They sent me to the hospital but they didn't have "privileges" at the hospital so they had to send me to the ER. They gave me a DVD with my X-Rays. When I got there they didn't know what to do with the DVD!. Well since I didn't look that bad and my ekg was good I waited about 10 hours!
I was in the hospital for a week. Buy they time I got in they where wondering if I should go into ICU since one lung was completely shutdown and only had half of the other one working.
I have to wonder if they could have just popped in the DVD if I would have gotten in sooner.
Oh and I do have a real doctor. Since I have always been so healthy in the past when I tried to get in they told me it would be three weeks.
The poor guy felt so bad that he office now has standing orders that if I say I really need to get in that I get in.
Re:stupid question but..... (Score:4, Informative)
If this can save so much money why isn't the health care industry already doing it? Are they really that stupid or are all the promises of big savings not likely to pan out?
Electronic Medical Records (EMR) are great, but there's a significant startup cost. Next time you go to a doctor's office, look at the cabinets full of charts. Now imagine buying a system, hardware, and training. Then there's a transition period where you're entering data into the new system and the old paper-based system (unless you're 100.0% positive that the new system will be completely reliable and that you didn't buy the wrong one and need to start over). Also consider huge storage costs, not for the EMR database(s) alone, but for scans of all the extra paperwork - insurance card, signed HIPAA paperwork, fax from the family doctor, X-rays, etc. Finally, consider the enormous workload of converting old records. It's one thing to start entering new data, but what do you do with the old stuff? Do you pay someone to do thousands of hours of data entry, or do you just scan every scrap of paper in and call it good?
We've already bought an EMR system for my wife's practice, but haven't gotten far into deployment for all of the above. We want to go electronic for all the reasons you could imagine, but it's not like you can flip a switch any more than a Fortune 500 company could decide that they're going to switch from Windows to Linux one day. There's a huge amount of preparation and migration involved unless you're making a clean start.
By the way, "not wanting to make it easy for patients to switch" is not an argument against EMR, at least for my wife. If a patient wants to see another doctor, it's mainly for one of three reasons. First, we want people to get a second opinion on my wife's advice if they choose. If the other doctor agrees, then the patient feels good about their treatment plan. If the other doctor disagrees, then the patient makes an informed choice about their treatment. Either way, they've become an active participant in their care which is a good thing. Second, if they're in another town and need emergency care, we'd love to be able to fire off an encrypted email (or FTP or whatever the standard becomes) to their treating physician. Third, if they want to switch doctors permanently, an electronic transfer is far easier than making copies of their entire record (since we're legally obligated to maintain 7 years of records on our own and can't just pass along the original copies of everything).
So we're onboard with the general sentiment. If there are any "flag days" for conversion, though, this better not be Yet Another Unfunded Mandate or a lot of older practices will simply close their doors. A doctor near retirement with tens of thousands of charts won't willingly spring for an expensive system plus all the labor involved.
Good luck, didn't work in UK (Score:5, Informative)
We have a similar "project" or rather it should be characterised as a "permanently stalled horror story." There are only 70 million or so people in the UK so nobody understood why the initial budget was in the billions. Now it's in the tens of billions and no end in sight. Google NHS IT if you really want to spoil your day.
KP already does (Score:3, Informative)
Just casual observations as a Kaiser Permanente member...
It looks like most of their records are digital already. I suppose the biggest roadblocks are patient confidentiality and government privacy regulations. So I'm assuming Obama's plan would/should focus on security more than anything else.
There are some interesting details with how KP handles things:
* All email correspondence goes through their own secure webmail servers. They only send you notices like "You have new email on our servers, log in to our secure servers to read it" to unencrypted mail on external email accounts.
* The data appears to be tied to a particular medical centers, so when we moved from one center to another, they had to transfer our record over. So apparently things are compartmentalized, so any random doctor can't look up information on every KP patient, just ones assigned to their medical center (and maybe their department). I'm not sure if that means each center has its own database server, but I'd assume not... it'd make more sense if they had two or more data centers in different cities with some redundancy and mirroring between them.
* Accounts for my spouse and kids can be linked to mine so I can make appointment requests and stuff for them, but it seems like it's still possible for them to hide their own medical appointments and records from me if they wanted to, I think. At least until I get a bill in the mail for things not covered by insurance :P
* There are not really any useful medical records available via the online interface, other than your email trail with doctors/nurses. It would be neat to be able to log in and download the kids' growth records and ultrasound pics. But if you really want stuff like that, you still need to get it from the doctor during a visit.
* You memorize your MRN (medical record number) real fast, because just about everyone you talk to (whether in person or on the phone) asks for it. They don't seem to "cache" it so they can start talking to you by name, nor do they transfer your MRN to the next person in the chain. OK, I guess the doctor, when you finally get to one, talks to you by name.
* Doesn't seem to have sped up any part of the process... it typically takes about 3 hours to do a visit, between checking in with registration, seeing the nurse, seeing the doctor, checking into the pharmacist, and then picking up a prescription.
Overall, I'm actually pretty happy with the service, because my family mostly tries to avoid going to the doctor so it doesn't bother me that they mostly avoid seeing me. But it could stand to be a bit more efficient. Having digital records doesn't seem to have help or hurt much in either respect.
modernizing (Score:2, Informative)
Data Mining (Score:4, Informative)
I understand the potential problems with security, cost, screwups and stuff, but part of me wonders how much of this data could be used for diagnostic analysis by looking at symptoms, vital signs, treatments and outcomes over a very large population.
Re:stupid question but..... (Score:3, Informative)
Because no one knows what the standard should be about. There is no standard electronic medical record system. At the moment, everyone interprets it differently. Some scan paper and say that's enough for them. Others just type full text in. Few have them more structured. What the format will contain will depend on the intended features of the medical record and there is no clear agreement on that.
Believe it or not, health care is far more complex than what software nerds think it is and making a data format standardized without understanding its implications is just asking for trouble. There is enough health informatics literature to suggest that we don't fully understand technology insinuations in a clinical setting. Doctors will happily adopt technology that helps them and their patients. So far many remain unconvinced at the choices they have.
It is also more realistic to focus on standardizing communication between systems rather than the data formats within the systems, at least for now. There are already methods and standards that have had partial success (see HL7, IHE).
Regardless, we eventually need a standard. 5 years is simply too short a time frame for that to happen, largely because US healthcare is too diverse and judging by the pace of research in the field so far. This might be easier to do within more homogenous health systems such as those in Europe (or at least, so I hear). It is good that the funding is beginning but the goals must be realistic. Can we force something in 5 years? Sure. Just not certain that it would give the bang for the buck.
Re:Doublespeak time! (Score:3, Informative)
The problems of privacy were there back then - if anything, they're worse now in a more-connected world. They still need to be solved. Just because Obama has touched it doesn't make it turn to gold, I think most everyone could agree on that.
Can you link to specific instances - a slashdot article or comments to same - where McCain was bashed for his EMR plans?
Other issues (Score:5, Informative)
In terms of privacy, we audit all access to our medical records and have a team of auditors who monitor access. I've been responsible for writing exception reports and such. It's far easier to tell who's accessing your medical records than paper copies laying around.
Data load is a big deal. We have our main EMR and multiple data repositories where we can do reporting and other non-operational work. Lots of people support all that infrastructure. On the plus side, that infrastructure lets us do things that saves money and lives.
Server outages have been pretty bad, but we have assorted downtime procedures and downtime systems. An example is a downtime database located in the various medical record office that are constantly updated with patient allergies.
Our staff has gotten quite used to working with an EMR, but there are still cultural issues. For example, what a physician writes in a medical record may be visible to the patient. They have concerns about speaking plainly (e.g. describing a patient as alcoholic). At the same time, patients have a right to know how they're being treated.
My biggest worry about the new plan to convert paper records is that there are so many EMR systems. Will they pick one of the existing vendors? Build a new one? The ideal for the patient is to have a single nation-wide EMR that they can take wherever they go. This has a huge impact on existing EMR vendors and installs, though, so I doubt people will take that approach.
Re:That estimate seems really high (Score:3, Informative)
Well here in the UK we have the exact same thing for approx 60 million people. The estimated cost? 18 billion dollars and probably more. What's the population of the U.S.A.?
Re:stupid question but..... (Score:3, Informative)
Because each member of the health care team needs to be able to chart in different ways and have access to different pieces of information without exposing them to more than required to do their job (part of HIPAA) (doctors, CNAs/PCTs, lab tech, RTs, OTs, PTs, nurses, pharmacists and pharm techs, social workers...),it's a huge chore to get all of this up and working like its supposed to be, because each member has their own portion of the software specifically designed for them. As someone who works day to day with the handwritten disaster that is a normal chart and can compare it fairly to the amazing repository that is an electronic chart?
I can tell you that there simply IS no comparison.
As for why hasn't this been done before? It simply comes down to cost of implementation. It's not just that you need to lay out for the software and the equipment to run it properly, you need IT staff to run it, maintain it, update it. You need IT/medical staff to teach people how to use it, (paying for a couple of days of training for everyone in the system isn't exactly a small financial burden either) and then theres time lost in the learning curve as well.
There's actually a TON of stuff that our system can do, its not just charting; its being able to pull up labs and other diagnostics at a moments notice, its having all of that information not only available in one place, but charted in such a manner that its very easy to spot trends. It will most certainly improve health care in the long run.
Re:stupid question but..... (Score:1, Informative)
Excellent argument, except for the lack thereof.
As someone who works IT in the health industry, let me tell you about a REAL boondoggle or two.
How about this boondoggle: I worked on the specs for a database of carrier records last year that included six - count them, six - project managers, none of whom talked to one another, none of whom fully understood the project, and none of whom had the same goals. Ultimately, the project was scrapped because the six musketeers spent so much time fighting with one another and redrawing each others parts of the specifications that the project was eventually deemed so complex that it was no longer cost-justified and they bought out the remainder of our contract and killed the entire project.
My company "earned" over $1.2 million for producing exactly NOTHING under contract for this group. Not ONE of our coders ever lifted a finger because nobody was ever able to procedure a final spec for them. And how much more money was spent in salary on the six executive level clowns while they did nothing except torpedo their own project?
Perhaps you'd like to discuss instead the group that was running three different databases that were connected solely by sneakernet: one for payments, one for social security and medicare processing, and one that duplicated each of these other two functionalities because it was supposed to replace and combine them, but the internal processes and job descriptions were so poorly defined that they had to keep "adding on" to it until they eventually gave up and just resorted to using THREE systems instead of two. We politely declined to contract with them at all.
Who do you think the "sugar daddies" are that footed the bill for these two goobers?
Creating a single standard for health records has already been done, now there just needs to be an efficient and effective way of tying all the different data stores together and allowing fast, secure communication. The VA already did it in the U.S., as have some entire countries, or portions of countries.
Anybody who opposes this either:
1. Stands to lose something from it - i.e. you're one of the millions of useless pieces of flotsam in the industry that's just sucking up money and producing nothing in return
2. Has a political axe to grind (my guess as to you'e commentary on the matter)
3. Has never actually seen just how dysfunctional the proprietary systems of insurance companies, carriers, and resellers are
Re:stupid question but..... (Score:4, Informative)
The fact that every other major industrialized nation has universal healthcare, provides outcomes comparable to or better than the US, and does so at lower expense (measured either per capita or as a share of GDP) than the US.
The first already is taxed higher, in part because of the health consequences, and proposals on the latter have been made independently of universal healthcare.
Re:stupid question but..... (Score:4, Informative)
Speaking as the son of the owner of a body shop, you clearly have no idea the hell insurance companies put people through. The DMV can make you wake for 5 hours on a bad day, but the auto-insurance companies will spend months, and I've even seen years, denying or delaying payment. My dad spends just as much time dealing with the insurance companies as he does running the entire shop. Not to mention that they always want to get the cheap, less durable parts, or crappy after market parts, if it saves them any money, regardless of the impact it'd have on safety, or the vehicle for that matter.
And we already have a nanny state. You can't do most drugs, you already get taxed highly on cigarettes, smoking is already banned in public in many municipalities, etc. You realize that there are many industrialized nations which already HAVE universal health care right? It's not like this'd be some grand experiment for us. If anything, we're behind the curve on this.
On a side note, this is NOT Universal Health Care, as you seem to assume. This is standardization. This means everyone'd have the same information and be capable of sending it to other facilities and physicians without absurd hassles and delays which could cost a patient their life.
Re:stupid question but..... (Score:4, Informative)
It's not just the format that kills you...it's the dataset, too.
Fat lot of good it does for the government to list a set of specs, when at the same time they don't list a means for actually encoding the content of those specs. Something simple, like drugs in a drug database, can cause all sorts of havoc when you realize that the only freely-available standardized coding scheme (NDC) doesn't so much identify drugs as it does specify an exact drug, manufacturer, package type, and quantity. In other words, aspirin isn't aspirin, it's "Easprin Tablets, 975mg, tablets, 100-count, bottle, oral, NDC 10802-9757-*1". Yes, that's a correct spelling of "Easprin", and that's an "asterisk 1", just to make things interesting. Not quite the same as (somehow) identifying that the patient is taking "aspirin".
Now, there are multiple PROPRIETARY databases out there to identify these drugs, but fat lot of good it does to try to communicate those identifiers to someone else if they don't use the exact-same PROPRIETARY database that you do. Or if the wording of the drug description in THEIR proprietary database doesn't exactly match the wording description in the one that you're using. Or whatever.
In fact, some companies (SureScripts being one) have acknowledged the shortcoming by asking providers to provide a "representative NDC" in order to identify the med that you're trying to communicate in an e-script. So, you write a script for "generic drug foo", but you send a "representative NDC" that specifies drug "bar", which is an equivalent of "foo", and which has a specific NDC. The pharmacy fills the script using the "representative NDC" to help them determine which actual packaged drug (and therefore which NDC) they are going to dispense--let's call that "meh". Then, when they eventually ask for a refill, they ask for a refill of "meh", which you sent as "bar", when what you really wanted to give the patient was simply "foo". Confused yet?
So, it's not simply a matter of defining a data exchange format...we still have to figure out how to commonly define what "it" is that we're actually exchanging in the first place.
Add in the difficulties in trying to determine what "problems" a patient may actually have (is one diagnosis of hypertension indicative of an ongoing problem? Was it secondary to another acute condition? Was it a mistake?), the dilemma of possible mis-use of the information by employers, insurance companies, or even government agencies, and the somewhat-fundamental problem of needing some way to globally identify the patient in the first place (by law we can't use SSN--and heck, in a college town, many of the students/patients don't even HAVE a Social Security Number), a national heathcare ID number does not yet exist, and even if it DID exist there would be a huge bureaucracy around securing this number and the information that it points to...
Nope, just having an agreed-upon format for sending records back&forth doesn't really solve the entire problem. It's a valid first step, agreed, but once the format has been loosely defined, determining what goes in the fields of those formats...THAT is when things can get "interesting".
Re:There is a pitfall though. (Score:4, Informative)
Pardon me for asking, but what statistics are you using to cite your first "fact"? I find that claim quite hard to believe.
Actually there have been quite a few such studies recently, mostly as eaurpean countries attempt to figure out what laws make sense with their healthcare systems. The first one to show up in Google for me was:
van Baal PHM, Polder JJ, de Wit GA, Hoogenveen RT, Feenstra TL, et al. (2008) Lifetime medical costs of obesity: Prevention no cure for increasing health expenditure. PLoS Med 5(2): e29. doi:10.1371/journal.pmed.0050029
On the second point, optimizing system means reducing unnecessary demands on the system also.
I never argued that, only that laws and regulations tend to ignore what actually reduces demand on the system in favor of punishing what people dislike (obesity) while ignoring any evidence. Regardless of if smoking or obesity reduces the cost on the system for everyone, most people will favor rules banning it and claim justification using the healthcare systems because the desire to punish is stronger than the desire to make the system cheaper.
Wow, that's the first time I've heard that study applied to socializing medicine. I've always heard of it being used to illustrate the immense hate of "rich" people and justify confiscatory taxes on said "rich".
Don't understand the logic of such an application. How does one argue such a study justifies progressive taxes? How does hate justify taxes?
I think the real argument has nothing to do with that, though. If everyone is expected to share the costs of something, everyone should share the responsibility of keeping the costs down. Poor diet and the ensuing health issues are not something that deserves sympathy.
Sympathy? I have two concerns and neither has anything to do with sympathy. The first is reducing the amount of taxes that need to be spent on the socialized portion of the healthcare system. The second is personal freedom, where said freedom does not significantly impact others. My problem is others are inclined to remove freedom and increase costs because they want to punish people they feel are doing something wrong (overeating or smoking in their home). Personally, I'm medically underweight and don't smoke, but I'm a strong advocate for personal freedoms and I don't like my taxes wasted on regulations that just increase costs to me while reducing the freedoms of others. If people want to overeat or smoke, you'd better have some really convincing evidence that it is costing healthcare a lot more than it is saving before you will get my support on restricting their freedom to choose.
If you pig out on Doritos and Big Macs and end up a diabetic because of your poor choices, why the hell should anyone else have to pay for your lifestyle?
If you exercise all the time and eat really well and as a result live twenty years longer why should anyone else have to pay for your lifestyle? Oh yeah, because paying for everyone's lifestyle saves money overall as well as bring numerous other societal benefits like reduced crime and a more stable economy.
They end up having to wait in line (especially when you start seeing the inevitable rationing that comes from socialized medicine) for the bums that chose to live poorly and have health issues because of it.
Sorry, these scare tactics don't work on me. I spent years waiting in lines in the good ole USofA when I developed a serious medical condition. I came within months of marrying a friend and moving to Canada just for the healthcare. Objective reviews of healthcare systems around the world don't exactly paint the US's system as the top of the heap, especially considering how much more we pay. Investing the same amount in a socialized healthcare system would not inevitably lead to any longer wait times for the average person than we have now. They would pro
Re:exatly (Score:5, Informative)
Having worked in Radiology IT, I'd point out that the human eye is actually a limiting factor in our ability to see things. So, yes, you are correct in that an analog X-Ray had more diagnostic information encoded into it than a digital one. However, the human eye can only see so much data. The human eye isn't a microscope. So you don't need to scan an image to microscopic precision. You only need to scan it to a level of precision that the human eye can't detect a difference. And it turns out that to do a scan of an X-Ray at that level of details results in a file that's approximately 10.5MB in size. It's a very well known metric, actually.
And if you have a 600 bed hospital which sees 150,000 patients per year in the Department of Radiology, and 50% of those are X-Rays images, with 2 images (PA, Lateral) per exam, that results in, oh... about 75,000 x 0.5 x 2 x 10.5MB = 787 GB of data per year. Add some extra space for buffering, swap space, and so forth, and you're talking about a Terrabyte of data per year for the X-Rays at a large community hospital. MRI, CT, Ultrasound, Angiography, and all the rest will add more, of course.
The hospital I used to work at stored about 2.5 terabytes of images per year.
Re:Why not study America's EMR? (Score:3, Informative)
Because while CPRS is a wonderful organizational success story, Britain's EMR is a national success story. And national scale is what we are looking at right now.
In US, EMRs came early and were great in academic centers and federally funded facilities. They did not work out so well for smaller practices.
For one, CPRS does not have billing and made adoption difficult in the past. Not to mention the fact that they kept rewriting it several times. The clinicians love it because it close fit they way they worked in VA. Not surprising since a ton of money was spent perfecting it in that way. But the problem is with moving it to a non-VA setting.
BTW, there are already efforts to make CPRS more generally applicable (http://worldvista.org/). They are not a major player yet however.