Thursday, January 17, 2019

Seen in a chart

This isn't the first time I've put up stuff like this, and it won't be the last.

But you know what? Crap like this is no longer the exception. It's the rule. I'd say at least 50%-60% of charts I read from hospitals and practices that use computer charting systems (which is pretty much all of them) have errors of this kind.




And these are what the world is pushing us to use more and more of.

I'm not saying computer chart systems are bad things. They have a lot of advantages. But they also encourage the slacker inherent in all of us. It's easier and faster to check boxes, cut & paste, and use templates than it is to actually type out what's correct. Especially if you skip the critical step of proofreading what you've just done. Most do.

The majority of these errors are just amusing. This one is just stupid, but likely won't cause a serious patient outcome.

But if it can make an error about smoking, it can also make them about your allergies. Your current medications. What conditions you have. Your past surgeries.

And one "minor" error in any of those could lead to a disaster in the right setting.

21 comments:

  1. I am always having to correct my allergies. I had carpel tunnel release surgery last week and I spent hours inputting data only to find it was corrected somehow. I agree systems can create problems.

    ReplyDelete
  2. We don't even have the option for inputting data, it is all tick box. Oh, you are on medication that isn't on the list? Too bad, there is literally no way for me to list this as a medication. Becomes especially troublesome when the patient is part of a double blind trial, I can't put a maybe they are on it, maybe they are not.

    ReplyDelete
  3. I am a former medical transcriptionist turned editor and am one of thousands who have lost their jobs to voice recognition. You would not believe some of the things I've seen VR come up with. And you're right...most of it is funny (at least VR provides us with a comedy show in a job that has become very stressful). Most of us no longer make enough money to live on with just one medical transcription job. Most of us are now paid by productivity instead of hourly and are paid very low because supposedly we can do editing twice as fast as typing, but with all the editing we end up doing it doesn't quite work out that way.

    ReplyDelete
  4. I always go into any situation with this unfortunate fact in mind. And I remind my own physicians of my conditions at the outset of any exam. Though I know they should have just reviewed my chart, I know they don’t. A friend stopped a very reputable MD before he took her little boy in for g-tube surgery to remind said doc to be careful of the leads from the boys pacemaker. The surgeon said, and I quote, “Leads? What leads?” Even though warnings about the boys pacemaker were all over his chart.

    ReplyDelete
  5. And from the pt side of the computer screen -- I go or take Mom to the doctor. We don't see the doc. WE see the back of a computer and hear the doctor/nurse/whoeverisasking ask questions.

    Eye contact? Humpf.

    And yes, I know my allergies (genuine MD-certified penicillin from observation of my rxn.) and guess what was not on the little screen? Mom is allergic (not sure about this, but it is what she wrote down before dementia) to morphine, codeine, Dilaudid, etc. At 85 and her sister in the same state, I can't ask. So unless it is life or death, I don't want her taking that.

    So what happens if she gets to the hospital and the EHR has gone wacky and I am having one of the post-stroke, post chemo memory lapses?

    Please guys and gals, fix this!

    ReplyDelete
  6. "It" isn't making errors. HUMANS are making the errors. Of course there is clearly room for this system to be improved. But we can't just sit around blaming computers and software for this stuff. Users have to own it, and provide feedback to their employer & the EMR company, in order for changes to be made. Sitting around grousing is fun, and this is an easy target, but c'mon.

    And no, I don't work for Epic or Cerner or any of the others. I work on the provider side, MOF.

    ReplyDelete
  7. I agree strongly. The verbage is also often very wordy and hard to read. After generating (n.b. not writing) a long near-useless templated note, I often feel the urge to write a free text "real note" in there somewhere containing the relevant information. A community internist I talked to recently had basically come to the view that our notes as they are now are for the coders and the lawyers first and to himself or other physicians only second.

    ReplyDelete
  8. A while back, there was an error on my chart indicating I was a smoker. I was not, am not, and never have been a user of tobacco products. When automated charting and sharing among practices became more common, I found myself fighting a battle to correct all the records. Eventually, it made its way to my health insurance company. Now, all of my doctors and my health insurance provider are regularly trying to push smoking cessation programs my way. No amount of explaining on my part helps. Somebody put it in the chart at some point and it must be true. I wonder what would happen if I applied for life insurance?

    Then, there’s the patient portal side of these programs. They ask me to fill in my medical history, allergies, medications, etc. Someone at the office reviews it and it looks fine on my side. When I go back to the doctor, it is either missing altogether, partially missing, or the data is corrupted. I saw my primary doc yesterday. I have four drug allergies. On this visit, I had three plus an allergy to “????????”. I guess I’m allergic to question marks. The nurse ends up spending more time trying to fill in all the right boxes and correcting information than she does talking to me. Some of my actual medications are not listed, so my doc ends up hand writing the prescription and they do not show up properly in the software. They end up being in my medication list two or three times. The whole process is turning into a bit of a circus; but the penalties for not performing in the circus are too great.

    ReplyDelete
  9. Obviously, Anonymous@9:36 is allergic to the unknown!

    I think what we are going to have to do is do like we have with credit reports--a system that lets your review all such data and challenge it.

    ReplyDelete
  10. I work in a pharmacy and we are always getting electronic rxs over with 2 sets of directions. It's fine when they match (eg: take 1 tab twice a day twice a day), but often they don't (eg: take 1 tab twice a day three times a day"). And on maintenance meds, some times the error is locked into the patient profile somehow, so every time the doc sends over a new rx for that med for that patient, it comes over with the same wrong directions. It can be maddening.

    ReplyDelete
  11. I had my tubes removed last summer and the OB/GYN I went to for that had a patient portal where we were supposed to fill in a bunch of info before we go. I got a bit stumped when it asked about alcohol. First my options were never, rarely, sometimes, frequently (or something like that). Then it asked for a specific amount. My options were None, 1-2/week, 4-5/week, 1-2/day, 4+/day. I drink 3-4/year. I had already said I drink rarely, so never didn't seem right, but 1-2/week also seemed egregiously wrong. There was no way to give them the correct information.

    Also, I would consider all of those amounts to be frequent drinking, is my normal meter that far off?

    ReplyDelete
  12. LOL guys, if you think you've got it bad try being a rare disease patient who unfortunately had a 10 year long, extremely complicated and f-ed up path to the correct diagnosis. There is so much crap in my records that is just flat-out wrong. Either because medical providers clicked the wrong boxes, or entered info that seemed right at the time, but then was disproven as diseases that they *thought* I had eventually got ruled out. I'm told much of the incorrect stuff is in my records forever - apparently none of it can be edited out, which is complete and utter bullshit if you ask me (but when does a patient's opinion ever matter?).

    If my disease doesn't kill me outright, my medical records will.

    ReplyDelete
  13. An outpatient was being discharged on Friday, and transferred and scheduled their daily antibiotic infusion care to our small facility. Anticipating dosing issues since the therapy was starting on a weekend when most offices are closed, I hurriedly did some calling around to the previous hospital, pharmacy, laboratory, prescriber to find out some basic info such as diagnosis, underlying illness, allergies, other meds, weight, height, etc and had to use our lovely system to document patient parameters under a 'general' comments info template as the patient wasn't actually 'in the system'. So, I went through the whole retinue: smoking, drugs, alcohol, abuse, and laughed to myself with the new questions for the year about sex, sexual orientation, sexual preference, before I could actual document stated height and weight. But, our system has the choice of 'unable to determine' which I use freely, since I didn't have my crystal ball with me.

    ReplyDelete
  14. We receive quite a large percentage of electronic prescriptions at my pharmacy.
    Some Doctor's offices make mistake after mistake that we have to call on and correct. Some are funny - but some could be quite dangerous.
    Some Doctor's offices send clear, mistake free electronic prescriptions every time.

    No, its not the computer, it's the idiot entering the information.

    ReplyDelete
  15. From the researcher side it also drive me absolutely batty. If I looking at the effectiveness of say a smoking cessation implementation in an ante natal clinic, something like this could totally derail my data collection and make that mother/child pair invalid for analysis. Too many like that and I dont have the numbers to say if the intervention worked or not, which has long term effects with regards to standards of care and what is offered at the clinic.
    The number of times I have had to go back to the paper records and actually cross reference the midwives file with the delivery nurse file to work out who (if anyone) was right, is very annoying.

    ReplyDelete
  16. my husband had sinus surgery, and we both mentioned to the ent that he had had terrible reactions (stomach cramping, vomiting) to high dose naproxen. 'give him anything but naproxen' was my only request. The ent, who was busy charting during the pre-op, heard 'only naproxen.' Luckily (?) hubby had to have an artery cauterized so all nsaids were ruled out. It worked out, but what if his reaction was a true allergy, perhaps life threatening? hearing and listening are completely different.

    ReplyDelete
  17. Some of the mistakes in my electronic record:

    I was listed as 55 inches tall instead of 5 feet 5 inches i. That really skewed my calculated BMI.

    Another time the chart showed I had a fever "in excess of 105 degrees". But I had no fever at all.

    The chart also listed me as being retired from a job I never had. And I've never been retired.

    Family history showed wrong diseases for some relatives.

    ReplyDelete
  18. I'm a forensic pathologist, and these records are a mixed blessing for us. When I'm acting as a Medical Examiner, they are a horrible headache, since they are barely readable. They are bad enough when you are using the case management system that they are created on. They are almost useless when you get stacks of printouts in response to a subpoena. The worst are the ones where every sentence of information in the H&E is followed by a parenthetic sentence indicating the the name of the scribe, the name of the physician, and the date and time it was dictated and transcribed. It makes the chart *literally* unreadable. Since I have a limited amount of time for a chart review per case, I know I'm missing stuff.

    On the up side, in my private practice as a litigation consultant, it's been a gold mine. Unlike when I do my regular ME work, I charge billable hours when I review charts for a medical malpractice case or wrongful death case, and my client dictates how much they are willing to pay for me to review. I recently got a case that came with 3000 pages of medical records. I told counsel that they could either send me a synopsis or I could go over the full record, but if I did the latter, I would *really* go through the record, I wouldn't skip any of the pages. I don't dare show up for trial claiming to have reviewed a medical record but not knowing what is in it. And that means a lot of time. With an average of 2 minutes a page, that's about 100 hours. Counsel said "We don't care" -- because they pass the cost to the hospital or insurance company, and the insurance company decided not to settle with what they considered a predatory litigant. Similarly, on the plaintiffs side, I'm invovled in a case where they think they have a slam dunk case, and they believe that the settlement will pay for my work. So I spend hundred of hours going through these things with a fine-toothed comb. Most of those pages are content-free, but you don't know it until you read it. Ten years ago, back when we had "real" charts, it would have taken about 20 hours rather than 100.

    ReplyDelete
  19. I became adopted about 3 months after a new system was put into use by my PCP's office. Was discovered after and amazingly awkward conversation about possible familial risk factors. Such amazing tact from the tech when questioning why I thought my relative's blood pressure would be a cause for concern. The adoption was quickly cancelled- never did get an explanation.

    ReplyDelete
  20. I think we should all scream loud and long about the errors. And about lack of communication about what is in the patient's chart. Companies pay a fortune for electronic charts; there NEEDS to be a way for them to be more accurate, whether it offers an option for "other" that allows you to enter more information or warnings for people to read attached notes.

    I'm not sure of the best solution but I think we need to measure and improve accuracy rather than be satisfied with the status quo.

    ReplyDelete
  21. During my last hospitalization, the chart had erroneous information about my allergies. The doctor said HIPAA means those errors can only be fixed after I get out and requires a cash payment. (This doctor also withheld potassium as a punishment for having neuro symptoms even though I was hypokalemic and insists that calling my emergency contact and asking if I probably know who the President is is a "comprehensive neurological exam.") Now that I am out I was able to edit -- but it's multiple choice. No room to type out an explanation. So most of the info is as close as it can get, but wrong. Moreover, drugs "migrate" from the allergies section to the current meds section.

    I don't get it. This seems like a recipe for someone dying.

    ReplyDelete

So wadda you think?