Those who advocate for electronic medical records cite a decreased incidence of medical errors.
The VA’s universal EMR, VistA, has been hailed as a model to aspire to. That confidence was recently shaken by an AP report, which disclosed a “software glitch” which exposed patients to wrong doses of medications.
One example included heparin, a blood thinner that requires close monitoring. Other problems included vital signs, active medications, and laboratory data being shown under the wrong patient’s name.
Although no one was harmed, these are serious, and potentially deadly, mistakes.
This clearly isn’t a shining moment of EMR proponents, and goes to show that just because a system is digital doesn’t automatically make them better. It’s another reason why we should be vigilant in ensuring the quality of the current health IT infrastructure before pouring millions of federal dollars into it.
Related posts:
- Do electronic medical records really reduce malpractice risk?
- Working harder won’t reduce medical errors
- Medical errors: Impact on physicians
- Poll: Will electronic medical records really save money?
- How to reduce the risk of medical errors from patient hand-offs
- Electronic medical records and the iPhone
- Do electronic medical records increase physician communication of critical test results to patients?
{ 7 comments }
While EHRs and e-prescribing will certainly help reduce some types of errors the VA example shows technology can introduce new types of errors.
I’ve been e-prescribing for over a year now and while my prescriptions are legibile, I’ve seen how easy it is to check the wrong box, highlight the wrong choice in a drop down menu or prescribe for the wrong patient.
I wrote an essay on unintended consequences and HIT in Family Practice Management a couple of years ago at: http://www.aafp.org/fpm/20060900/90whyi.html.
Proponent of EHRs, like proponents of IT anywhere – who isn’t these days? – never said there won’t be errors in a computer system. They are made by and managed by people. The lesson to be learned is not “beware of the EHR hype” it is that we cannot sit on our laurels and stop thinking critically just because we are using EHRs.
To error is human, to really screw it up, use a computer, to really, really screw it up use the government and a computer!
All information systems have bugs – this is why an intelligent human should double check.
However, humans make errors to. My mother is on coumadin. There have been numerous times when doctors prescribed her drugs that increased the action of coumadin yet forgot to tell my mother that so she could check her INR sooner. Pharmacists failed to notice it either. Good thing, there is internet so I can double check these things. One time, after I cross-referenced my mother’s new prescription, she went to check INR two days after starting the new drug instead of after 2 weeks as she was supposed to according to her schedule. Guess what – it was 7(!!!) when they try to keep it 2-3, actually closer to 2 because my mother also has macular degeneration and higher INR increase risk of immediate blindness. If I hadn’t checked the internet, my mother would’ve continued on the same dose of coumadin for another two weeks, running the risk of bleeding. Care to guess what her INR would’ve been then? Hint: it took my mother two reductions in dose to get her INR back to desired range. Is it really my – layperson’s – job to check prescribed drugs for interactions?
Oh, and when you call the doctor and mention it, the receptionist often say “did doctor tell you that?” “If not, don’t worry about it”.
Back on topic. Yes, computer systems make mistakes. But so do humans. So maybe one should use both?
So
Computer entry error now greatly exceeds handwriting illegibility as a source of inpatient med errors.
At least when the handwriting is illegible, the reader knows the information is suspect. When it is all spelled out plainly in 12 point type, the tired, the lazy, the stupid, the poorly trained individual will often simply not question what they see even if it doesn’t fit.
I have serious objections to EMR’s and none of them is the Luddite curmudgeon dislike of change.
1) If nobody reads the EMR, how are you better off than paper records? I’ve gotten calls for consults when I’ve seen the patient recently, uploaded my note and no one bothered to do anything except say, ‘Oh, this is on the problem list; let’s call a consult’.
2) If doctors actually read the EMR, they often do cut and pastes that are ludicrous – rehashing old stuff, not checking if other consultants have changed meds, contradicting themselves by putting new stuff at the end and not indicating that this is a change from the old info.
3) Given the hard work of thinking, it is near impossible to get MD’s to repeat a workup appropriately if they have the old record. Actual example in point: patient w/glaucoma nonresponsive to 4 meds. Patient went to new ophtho, who went ballistic that she refused to bring old records and he had to repeat entire w/u. In fact, patient did not have primary open angle glaucoma, for which she’d been treated, but exfoliative glaucoma, which needs laser tx to let meds work!
4) Notes in the chart are ridiculously long b/o cut and paste – ever try to figure out what was the patient’s last recorded clinical status in an emergency?
5) It is really easy to steal huge am’ts of chart info w/EMR’s. Chart info is always vulnerable to disclosure, but if charts are not aggregated, it is more work.
So – do you really want YOUR psychiatry, STD hx and genetic disease profile available in an EMR?
“5) It is really easy to steal huge am’ts of chart info w/EMR’s. Chart info is always vulnerable to disclosure, but if charts are not aggregated, it is more work.”
I think there are three separate concerns that tend to be lumped together when people start debating the privacy of electronic health records, and we need to address all three to convince the doubtful that EHRs can keep patient data secure. The first concern is that of hackers independent of a practice or medical institution somehow accessing patient data care for malicious purposes. The chances of a hacker accessing Internet-based EHRs and client server-based EHRs that don’t transmit their patient data across the public forum of the Web are miniscule (note that the security differences between Internet-based and Web-based applications are not known by many people, who tend to lump the two technologies together; see http://www.nuesoft.com/evolve to understand the difference). If Obama’s administration sets some minimum security standards relating to interoperability and the way that EHRs transmit data, patient information will quite literally be as secure as data held by the Federal Reserve or Defense Department.
The second privacy concern is that people within the practice or medical institution, whether a care provider or support staff, may access patient records needlessly out of curiosity (most likely with celebrities). However, electronic health records make this type of security breach less likely, not more likely; different permissions levels for different users and the “audit trail” in place on robust EHRs make records more secure than paper records.
Finally, there’s a worry that health information will be shared with companies against our wishes. Although EHRs make data easier to transmit, this is not really a concern about implementing EHRs; it’s a concern about federal regulations – and the ethical standards of the health care industry. Choosing to not store patient data electronically won’t prevent changes in privacy regulations, although it might stunt progress in health care.
Comments on this entry are closed.