Slashdot Mirror

← Back to Stories (view on slashdot.org)

Study Says E-prescription Systems Would Save At Least 50k Lives a Year

Posted by samzenpus on from the take-a-green-pill dept.

First time accepted submitter shirleylopez1177 writes "Approximately 50,000–100,000 people die in America because of preventable adverse events (PAE). These PAEs or medical errors are among the leading causes of death, ranking higher than breast cancer, AIDS and motor vehicle accidents in terms of the number of fatalities caused. As a response to the problem of medication errors, e-prescription systems have emerged. Few studies have looked at how e-prescribing systems compare to traditional systems in their potential to reduce medical errors. However, a study from Australia published two weeks ago in PLoS Medicine examined the impact of e-prescription systems on medication errors in the inpatient setting and demonstrated that these systems are indeed effective."

32 of 134 comments (clear)

  1. 10 years ago... by goathumper · · Score: 5, Interesting

    I worked on a hospital system 11 years ago that would provide this sort of cross-referencing functionality. It always baffled me why their use wasn't widespread. Back then there were (evidently) no smartphones, etc, so the whole idea of having barcodes on patients' wrists was revolutionary, as was the concept of having computer systems perform the drug-to-pathology matching and medication interactions analyses.

    From what I learned working on that project, this sort of system can lower the costs of operation, staffing, and evidently lower risk inside a hospital. Does anyone out there know why they've not seen widespread adoption (besides the "obvious" tin-foil hat doctor-nurse-conspiracy theories)?

    1. Re:10 years ago... by SteelKidney · · Score: 2, Informative

      I expect that reading the Daily WTF ought to answer your question. Or Diebold's attempts to use whatever legal maneuvers they could in order to cover up the fact that they were selling extremely poor-quality software. Or the fact that Sony got so thoroughly and completely pwned over the past couple of years that it's not unreasonable to assume that anything more complicated than "Hello World" written by a Sony team is yet another hack waiting to happen.

    2. Re:10 years ago... by Anonymous Coward · · Score: 4, Insightful

      inertia.

      No - cost.

      Hospitals have strict budgets and have to penny pinch. The software vendors charge a ludicrous amount for their software - so much that the hospital admins cringe and have a very hard time finding the money. And with these hard times, hospital revenues are in a huge slump - all those unemployed people have lost their health insurance and therefore can't pay their hospital bills - which the hospitals eat much of it. (COBRA is obscenely expensive and if you have a "preexisting" condition, you can't get cheaper insurance or any insurance for that matter; so millions of people go without even when they can afford health insurance.)

      To head off the "software vendors have to worry about lawsuits and that's why they charge so much!"

      No they don't. They have no more product liability costs than any other company and as far as FDA requirements, they've actually reduced some of the regulation.

    3. Re:10 years ago... by AngryDeuce · · Score: 4, Interesting

      Very much this. Doctors are notorious for being stuck in their ways, especially as concerns administration and computerization. My step-mother actually just quit her administration job at a small practice a few days ago because they were still doing everything on paper; she said she hadn't worked in an office with that minimal level of technology in almost 20 years.

      Especially now as doctor's "margins" are getting thinner due to Medicare cutbacks and such, I'm sure this trend will continue. New tech costs money, and medical tech, even on the administration end, is ridiculously expensive.

    4. Re:10 years ago... by goathumper · · Score: 2

      The interesting thing is that the whole system had been proposed and led by doctors. They knew the benefits and seemed to actively want them. Perhaps most crucially: the system didn't take doctors out of the loop - humans could still override the computer's warnings/indications/whatnot as necessary (obviously this would be well-audited).

      I agree that the risk of replacing humans with technology is still there. And yes - hacks are always possible as long as humans are in the mix of creating the computerized system. However, even if it lowers the number of fatalities due to PAEs by half, it would be a huge win money-wise for insurance companies, etc. (which begs the question: why hasn't it been done on that basis alone? We all know ca$h makes the world go round...) - despite the risk of hacks or tampering.

      Just sayin'... maybe we should build a F/LOSS platform for this so that it can be widely audited and its quality can be more transparently verified... volunteers?

    5. Re:10 years ago... by timeOday · · Score: 3, Insightful

      Especially now as doctor's "margins" are getting thinner due to Medicare cutbacks and such, I'm sure this trend will continue. New tech costs money, and medical tech, even on the administration end, is ridiculously expensive.

      I think the opposite: private practices are being driven out of business by large hospitals that work closely with insurers (including digital records), and more doctors are becoming employees instead of small business owners. In other words, price pressure is asserting itself and forcing consolidation, like with every other industry. Good or bad? I'm not entirely sure. We certainly do need to cut costs. There won't be many mom-and-pop shops that refuse to move to computer records any more.

    6. Re:10 years ago... by genjix · · Score: 3, Interesting

      3 years ago I damaged my elbow. I went to see the hospital, and the nurse being too busy to hear my full story hurried me along telling me it was sprained. I knew what a sprained elbow felt like and this wasn't it, but I shrugged my shoulders and assumed it would get better. It's been aching on and off over the last few years.

      A physician on the bitcoin forums was offering medical advice for a bitcoin. I typed up my full story and sent it to him. He wrote me back a long response that quite literally scared the crap out of me into seeing a doctor. I took his write-up to my General Practioner and she right away knew what was wrong and referred me to all the relevant specialists.

      That guy on the bitcoin forums literally saved me from crippling injury in a few years time. Had I not spoke to him, it may have been too late before I got it checked out. I always kept putting it off since I'm so busy and it didn't seem like a big deal.

      Thank you bitcoin forum guy.

    7. Re:10 years ago... by demonlapin · · Score: 4, Insightful

      I'm a physician whose hospital just tried to push all orders onto electronic order entry - not just medications but diet orders, PT/OT/nursing orders, everything. It got massive pushback. Why?

      Most doctors see patients at more than one hospital. Many use an electronic system at their clinic. They have to remember five or six usernames, passwords, and different ways of doing things, any one of which is likely to change at any time due to an upgrade, and some of which they may not use for months (as an example, many surgeons maintain privileges at a wide variety of hospitals to be able to suit patients - but they may not operate at a given one for two or three months at a time). The interface is often clunky. And they're SLOW. Paper is FAST.

      Great example from a committee meeting last week: one endocrinologist is part of a group that has taken over management of difficult diabetic inpatients. Most of them have Medicare, or Medicaid, or nothing at all. From his perspective, he's getting paid very little for his work. On paper, he can check blood sugars, write an order, and move on to the next patient in about two minutes. On computer, the same process takes about five minutes. Thirty patients an hour versus twelve... and so he said that if he's forced to do electronic, he will just stop doing the difficult diabetic management. It's no longer worth his time.

      And, as others have said, these systems are fantastically expensive, and so while there are some savings to be reaped they are mostly taken by the vendor and the increased IT expenses. And then your vendor decides to EOL your software... what do you do then? Buy their replacement product, because it's a lot cheaper to stay with the same vendor? Buy a new whole-hospital system from another vendor? We're wrestling with that now.

    8. Re:10 years ago... by nbauman · · Score: 2

      I used to evaluate medical office-based systems about 15 years ago, and I kept an eye on the field ever since.

      They made wonderful predictions, about half of which came true. (I made predictions. Mea culpa.)

      The billing systems worked very well. When they went to Medicare/Medicaid billing, the investment paid for itself in about 6 months.

      The clinical systems didn't always work so well.

      Transmitting lab reports worked very well. They substituted a standard paper format for a standard electronic format.

      Keeping patient notes is difficult. In the New England Journal of Medicine a year ago, a doctor was complaining about EMRs that automatically insert boilerplate, winding up with a hundred pages that nobody has time to go through. In contrast, the old handwritten notes forced doctors to think it out beforehand, be concise, get to the point, and emphasize what was relevant to clinical decision-making.

      There's a tradeoff between a narrative and a structured description. Some systems encourage (or force) doctors to structure their notes. The problem then is that in some systems doctors had to go through pages and pages of menus, many of them with irrelevant choices.

      Few doctors are anti-technology, but they don't want an electronic system that takes more time than paper, unless it has significant advantages. (Remember all the Ask Slashdot articles about, "What's the best way to take notes in class?" Lots of techies use wire-bound notebooks.)

      Electronic prescribing worked reasonably well. But even so, doctors say that it takes longer. They could quickly write a prescription on paper, but the electronic systems force them to go through pages of menus. In one study, the death rate in a pediatric ICU increased with an electronic prescribing system, because it took so much longer to write a prescription. It would seem that it should reduce drug interactions, but in real-world studies it doesn't do that. (Doctors had better understand what drugs they're giving the patients and how they interact with the other drugs a patient is taking. If they don't, they've got problems that computers won't solve.)

      Unexpected problems come up. How do you keep a keyboard sterile?

      The fundamental problem (which the Obama administration is perpetuating) is that it requires a lot of testing and evaluation, with controlled studies at every step, to get these systems working right. And there are a lot of steps.

      Doctors aren't willing to get a beta release that works right most of the time, but still has a few bugs in it. They're right.

      A system analyst can define a scientific problem, or a financial operation, in great detail, and then pass it off to the programmers to write the code. But health care systems, even a small medical office, are much more complicated and unpredictable, with lots of exceptions to the routine. And humans are really complicated.

      I believe in computerized medical records. I think all these problems can be solved. I think they'll be completely digital some day. But medical systems are more difficult to develop than their enthusiasts thought.

      We have to try things out in small systems, in small steps, and then large systems, in controlled trials, and make sure they do what you want them to do (save lives), not just what the programmer is able to do (save data), at every stage. If your computer doesn't save lives or improve patient outcomes, doctors won't use it. And rightly so.

      Then of course they have to be part of an integrated system. It's bad enough to learn one new system; a doctor doesn't want to learn 20 new systems. They could be reading JAMA, or even seeing patients, in that time.

      If you want to find out why computerization is going slowly, talk to some doctors and ask them.

      I knew a medical secretary at a major academic medical center. She used to keep index cards for every patient. While she was waiting for the system to deliver their electronic charts, she had a 3x5 index card with the essential information on it. She was actually faster than the younger secretaries, who had never used index cards.

  2. NHS e-Prescribing by Anonymous Coward · · Score: 2, Insightful

    Here in the UK, system like this are in use in both General Practice and in Hospitals. I worked for a company for seven years that supplied software that did precisely this to NHS and private hospitals both here an abroad. I wonder how the stats compare between the UK and the USA in this regard?

  3. You heathen technocrats! by fph+il+quozientatore · · Score: 3, Insightful

    How dare you replace a competent, well-trained, warm-hearted human with an emotionless machine?

    --
    My first program:

    Hell Segmentation fault

  4. The begin of the article misleads... by Troyusrex · · Score: 5, Informative

    by implying that drug errors are causing 50,000 to 100,00 deaths a year when, in fact, drug issues are a very small portion preventable adverse events (PAE). Things like falls and catheter infections are far more common. The article mentions that drug allergies and cross drug reactions are already extremely low and unaffected by implementing e-prescription (probably because the computers in the pharmacy already alert to this). The only thing effected are illegible prescriptions. I think e-prescriptions are a fine idea but this article is misleading as to how much benefit it would have in terms of lives saved.

    1. Re:The begin of the article misleads... by Another,+completely · · Score: 2

      I saw an article a few years ago that gave a great comparison. Sorry I can't find the reference, but at the time it said your chance in a hospital of getting the wrong medication ("wrong" defined as not what you were prescribed; never mind unnoticed conflicts and so on) was higher than the chance on a commercial flight of having your luggage lost. Some of those are certainly from illegible prescriptions or poorly labelled units, but I bet more are from procedural mistakes.

      Still, electronic prescriptions sound like a good idea for everyone concerned.

    2. Re:The begin of the article misleads... by vlm · · Score: 2

      An elderly patient may have mentioned a decade ago that they were "allergic" to some medication because they got a headache after they took it, but once that allergy is on the drug allergy list, no one is going to put themselves on the line and delete it. As a result, the lists of drug allergies tend to accumulate junk over time and may prevent physicians from using the most appropriate medication.

      Amoxocillian makes me puke, at least it did once 30 years ago. Or maybe I puked after amoxocillian because I was home from school and ate nothing but junk food because I was sick and miserable. Fast forward 30 years and horrible ear infection from my ear infected kids, go to doc, amox worked great on the kids but I can't have it. Doc suggests something and warned me of horrific side effects (was it cipro ?). I talked him off the ledge and we agreed zithromycin would be safer and more appropriate. 4 hours later the fever was gone, feeling better, etc. Even azithromycin is not harmless. The "best" answer probably would have been amox and don't eat any taco bell or other upsetting substances, but that is not possible for insurance reasons, etc.

      You don't want to get in a situation where you have a relatively minor headache, but aspirin gives you a slight tummy ache, so they "have to" do exploratory brain surgery instead. I can imagine an old person being "allergic" to everything and therefore getting crazy treatment plans that are much riskier than a minor reaction.

      When I was in the army my Drill Sergent "forced" everyone with a red allergy dog tag to find out what their reaction was, not just that they were allergic as a simple binary yes/no. He had some story about being in central america with a buddy with a minor leg infection and the corpsman only had antibiotics on hand that his buddy was allergic to, so they were contemplating cutting his leg off vs how bad would the allergy reaction be. Supposedly option 3 medivac saved both his life and leg...

      --
      "Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
    3. Re:The begin of the article misleads... by Rich0 · · Score: 2

      Agreed. A big problem is that often there is only a binary allergic/not-allergic list.

      My wife has been to the hospital numerous times and I end up going through the allergy list when she is unable to do so. Half the stuff on the list raises eyebrows because they are medications that she regularly takes. I explain to them that she isn't allergic to them, but that she does have sensitivities that should be considered (lower does, extended-release, avoid if possible, etc). They end up leaving them on the allergy list since they don't know what else to do with them. I try to talk to the doctors often so that they're in the loop, and usually the stuff on the list is more for chronic treatment so it isn't as big a deal.

      They really need to have lists that include what actually happens. If a drug makes you really sleepy or nauseous it is a completely different situation than if it causes anaphylaxis. However, I've seen doctors try to get my wife to take a drug in the hospital that we know makes her nauseous when she is already nauseous, and we already know that an extended-release formulation works better for her (but the hospital didn't have it handy). Things like that make me tend to micro-manage the nurses and account for every pill she gets, so that red flags like that can be escalated (especially when it just involves me running home to grab a bottle of pills and have the pharmacy ID it).

      I've also spotted cases where nurses try to administer drugs that doctors had intended to stop, despite having electronic everything already implemented (obviously the doctor forgot to update the orders). Again, being present I can have them bug the doctor and get it straightened out.

      There has to be some way to cut down on odd mistakes like these. Often they don't turn out to be serious, but they do often prolong a stay and add expense. Plus, you're far more likely to develop some complication from a 5 day stay than a 2 day one (I've had to deal with hospital-induced issues like heart failure, anemia (from thinners), and general loss of sleep/etc). Delays get compounded when a missed order doesn't get caught until you end up waiting another day (patient took a pill that had to be stopped pre-procedure, or some test is booked up, or whatever). In fact, I'd say that 90% of the time I've seen in the hospital amounts to ordering tests at 8AM one day, and then reviewing results and ordering more tests at 8AM the next day. If they just checked the results when they were available you'd cut out half-days of latency all over the place.

  5. Re:The solution, according to the summary? by gmack · · Score: 4, Interesting

    If the doctor could log in and select the medication and have the pharmacy read the prescription it would, on it's own, prevent a lot of errors that happen from misreading prescriptions. On top of that, if there is something wrong that requires a specialist then the patient is in a fun place where no one doctor knows what all medications are prescribed so a system that did any sort of automated conflict checking could save a lot of lives.

    The current system is far from perfect, I once almost lost my job because some pharmacist misread my prescription for Singulair (Asthma med) and gave me an antipsychotic instead and for a week I couldn't be motivated to do anything.

  6. Inadequate summary. Sigh. by Ronin441 · · Score: 2

    The summary (mostly) included one of the two key facts:

    each year approximately 50,000–100,000 people die in America because of [...] medical errors

    But not the other:

    implementation of e-prescription systems resulted in an approximately 60 percent reduction in total medication-error rates, and a 44 percent decrease in serious medical errors

    So the expected improvement is 22k to 44k less deaths per year in America.

  7. Re:No by 19thNervousBreakdown · · Score: 2

    Approximately 50,000–100,000 people die in America

    WHICH America? North or South?

    --
    "You can always count on Americans to do the right thing - after they've tried everything else." - W. Churchill

    WHICH America? North or South?

    --
    <xml><I><am><so><damn>Web 2.0</damn></so></am></I></xml>
  8. Just yesterday... by orphiuchus · · Score: 2

    I got some shit advice from the medical staff at my university. I'm taking a drug called celexa and got a cold, not wanting any adverse interactions I called them up and asked what medicine it was OK for me to take. Coricidin Cold and Cough they said, was the safe choice.

    I Googled it before I went to the store and found a major interaction via drugs.com. A potentially fatal interaction. Super.

    1. Re:Just yesterday... by vlm · · Score: 2

      Coricidin Cold and Cough they said, was the safe choice.

      The new stuff made from chlorpheniramine or the old stuff made from psudephedrine?

      Thats the "killer" with brand names.

      --
      "Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
  9. Re:Too Optimistic by Geraden · · Score: 2

    Not true!

    While human error like you describe above certainly exists, these systems can also catch drug allergy interactions, drug-to-drug interactions, and even food-drug interactions. Along with the already-existing systems in most pharmacies, these systems provide another layer of protection for patients. They also provide doctors with real-time best-cost analyses, allowing them to prescribe the most effective, least expensive drugs based on a patient's particular drug coverage. This may help to lower the overall price of healthcare and insurance coverage.

    Protection from the errors you describe isn't technologically insurmountable, either. Robotic systems that are linked to the prescription and automatically fill prescriptions eliminate the pharmacy errors, and EMRs that provide diagnosis/drug checking are likely right around the corner. Doctors don't like the latter much, however, because they are perceived as taking too much of the medical process out of their hands.

  10. We have this in Estonia by Reigo+Reinmets · · Score: 2

    We have this e-prescription system in Estonia for over 2 years now, nation wide. The good - It's easier to get some recurring prescriptions that you have to take all the time, you just calle the doctor and say you are running low on the meds, he checks your previous prescriptions and can easily see that yes, you should have only a few left... (to detect you are not attempting to scam extra medicines for black market or something). Also another good thing is that combined with an electronic pharmacy database, you can check online exactly which pharmacies currently have this medicine in stock in the right quantity(My partner takes meds that come in 10, 20, 30, 50 and 100mg forms, She only wants the 100mg ones). The bad - Initially they had performance problems because they forgot one basic simple thing when calculating peak usage - All elderly / pension receiving people(They are also the ones who require a lot of prescription medicines.) get their pension on the same day in the beginning of the month. This gives them a reason to leave the house and go to the city, and most of them also buy their medicines within the 2 days period following it, causing a massive performance bottle-necks for that moment(This problem was later fixed by adding more servers + optimizing).

  11. My Doctor's System Hates My Pharmacy's System by Jonah+Hex · · Score: 2

    And I hate them both! I have tried to make use of the CVS pharmacy automated refill system, but from what I can tell the "automation" goes into a blackhole and requires manual intervention. The system the doctor pushes is to fill out a form on their website, but from what I can tell it just generates a phone call from them to the pharmacy... sometimes. For me it is a major hassle, especially since my drugs are not considered "maintenance" prescriptions and have limits on getting insurance to cover mutli-month supplies. - HEX

    --
    Horror & SciFi Erotic Nudes
    1. Re:My Doctor's System Hates My Pharmacy's System by zindorsky · · Score: 3, Funny

      And I hate them both! I have tried to make use of the CVS pharmacy automated refill system

      You should try the SVN or HG systems instead.

      --
      If the geiger counter does not click, the coffee, she is not thick.
    2. Re:My Doctor's System Hates My Pharmacy's System by sjames · · Score: 2

      That's probably because were written by the sort of developers who derive a PatternFactoryFactory to create Pattern Factories that spit out HIJKLMNOP generators to instantiate the blargle!

      If instead, they just specified a simple tag:value record in plain old text, it would probably inter-operate just fine.

  12. Re:Are all deaths equal? by neapolitan · · Score: 2

    You two have a good understanding of the tradeoffs involved with decision-making. Unfortunately, many people do not and see suboptimal outcomes as "errors" in a very black-and-white world. I think the IOM report fed into many fears.

    I am continuously annoyed about the IOM report -- as other posters have said, it is now out of date, and sensationalist IMO in the way it counted mistakes and deaths / errors. An "error" that had no effect in a critically ill patient who died 3 days later was counted as a fatal outcome. On the other hand, the sensationalism at the time might have been a bit warranted -- doctors are often very complacent and perhaps the attention was needed / desired to get large scale action. However, it had the side effect of the erosion in trust in those that work very hard, diligently, and conscientiously every day.

    I very, very rarely use handwritten prescriptions. Certainly as inpatient (patients who are currently in the hospital) essentially all major medical systems have computer order entry as of 2012. In my outpatient clinic (people just coming for a doctor appointment) it is 100% computer medical scripts with automatic interaction and allergy checking. All of my hospital system is this way.

    I can't remember ever having ANY medication or dosing error. Obviously I can't know about it if I don't catch it, but computer order entry, automatic checking, and the many layers of check from doctor, nurse practitioner, pharmacist, and nurse, (and patient!) does provide a safety net.

    Can we do more? Well, banning handwritten prescriptions would be a pretty bad idea (if I'm in a community clinic wanting to give a patient some antibiotics for an ear infection, I think I should be allowed.) There are side effects to every initiative. Encouraging computer use is indeed being done, but limited by cost concerns.

    --
    Slashdotter, ID #101. UIDs are in binary, right?
  13. Before Windows Vista there was... by tepples · · Score: 3, Informative

    maybe we should build a F/LOSS platform for this so that it can be widely audited and its quality can be more transparently verified

    Can you code in MUMPS?

  14. Re:No by hrvatska · · Score: 2

    WHICH America? North or South?

    The America that people around the world generically use to refer to the United States of America. It might have something to do with America being the largest word in the country's name. When Iranians chant 'death to America', they're referring to the USA. Not Canada. Not Brazil. Not Mexico. Just the USA. Everyone gets this reference except people who have to ask 'WHICH America? North or South?' They're so fucking dense they go around wondering if Iranians want death for all countries in the Americas or just in North or South America. Maybe if Canada, Mexico, Brazil, Argentina, or any other major country in the Americas had America in its name there might be some confusion. But they don't, and there isn't.

  15. Good open sources software exists by sgent · · Score: 2

    in the public domain. VISTA is the Veteran Administration's EMR which has generally gotten very good reviews by physicians. However, it is an unbelievably archaic on the back-end (uses M, predates relational databases, etc.). In addition there is no emphasis on charge capture, so it often is useless for billing purposes.

  16. Medication errors != deaths by nbauman · · Score: 2

    Just because they made an error, that doesn't mean a death resulted from the error. A patient's blood pressure may have shot up or down for a day, but (unacceptable though it is) they might have caught it and it might not have harmed him.

  17. Bitcoin whore by nairnr · · Score: 3, Funny

    Thank you for your bitcoin advertisement. Now to return back to reality.

  18. Technology hubris by nbauman · · Score: 2

    Because non-tech people usually don't understand the "computer is a universal tool" thing and have problems stretching the limits of their imagination ("I had no idea a computer could do that for me...")

    Because IT evangelists think that every problem in the world can be reduced to computer code, and they create medical systems without understanding how medical practice works.

    Then the doctors try the system out, it runs into problems ("It takes me longer to enter a prescription into this computer than it does to write it on a prescription blank by hand"), and they correctly go back to the older manual systems that work better.

    Doctors aren't stupid. They use lots of new technology every day. When something works, they use it. When it doesn't work, they drop it.

    Medical practice is very complicated. The potential cost of error is very high. The cost of developing electronic medical records properly is very high. It's like developing a medical drug or device (which it is). You have to get back to your users (the doctors) continuously, find out what's going wrong, and fix it. Automating medical practice is a massive job, like automating the aerospace industry. At best, it will be slow and expensive. At worst, developers will take shortcuts, waste even more time and money, and have to scrap it all and start over.