41 replies »

1. You are obviously correct, Richard. there is no turning back and we shouldn’t look for a way back either. We just need a good and responsible way to move forward.
   The more folks get on the software, the faster it will improve because there is nothing more effective than customer input and customer complaints.
   However, since other than inconveniencing early adopters, these software products can adversely affect patient care, we also need a well regulated way of moving forward.

2. Margalit,
   You made some very intereting points in your blog. What most people don’t realize is that EMR software has been around for more than a decade, but that most healthcare professionals have not used it as of yet. Now that EMR software is becoming mainstream, healthcare professionals are beginning to analize,and critize it every chance that they get. Like always no one likes a change, especially healthcare professionals, and they most realize that this change is inevitable. They should learn to use the software that their facility has to its fullest, and realize that better software (updates, etc) will be on the way. EMRs are basically still in their infancy, and they will “grow up” rapidly.
   CCHIT on the other hand was placed in a position to rate and accredit these EMRs by the government. This position has only been for only about a year now, and again, it will eventually progress into its place. All new fields have found there ways in the past, as will EMRs.
   As for the usability of EMRs, it to will take time for healthcare professionals to be able to learn to use a EMR in a matter of 5 mins. EMRs are not as simple as are internet browsers, or word processors. They are complex systems that deal with not only very confidential information, but also many other systems like x-ray images, lab systems and others that interface into the EMR. EMRs are not plug and play type of software. As for the healthcare professionals complaints, in the past they would have training on the paperwork used in the facilty for at least 4-8 hrs as well.
   We need to continue to go forward into the future of EMRs, and not look back to the days of patient’s charts overfilled with handwritten notes, labs, and unorganized paperwork.

3. I’m working my way through the supplement from the bottom up, but one sentence caught my eye:
   “Future PCMH recognition and certification processes should focus more on patient-centered attributes and the proven, valuable key features of primary care than on the features of disease management and information technology.”
   There have been studies before that showed indication of what they called “high touch” having higher correlation with outcomes than high tech had.
   I never really understood why the PCMH criteria were so heavily tilted to EHRs. I do understand the basic need for having a comprehensive record stored at the MH, as it was always intended, but I think they went overboard with specific requirements.
   Anyway, maybe we can learn something going forward.

4. Margalit:
   Yes, there is value in data collection and population management, but those are very expensive activities and cannot/should not be allowed to interfere with one-on-one patient care.
   Did you see that the AAFP just released the independent evaluation of the medical home National Demonstration Project? In a word: disaster.

5. pcp,
   Your description of what passes as CDS is indeed useless and unfortunately all too prevalent.
   If we assume, just for the sake of this discussion, that reimbursement is available for both just talking to the patient and data collection, would you see any value in population management? Would registries help in managing a large panel of folks with multiple chronic conditions?
   What do you think about Dr. Wagner’s chronic care model?

6. The devices are impediments to safe cost effective care.
   They are clinical disrupters with inexpicable idiosynchracies that impede comparison of multiple data points at the speed with which the clinicians’ mind operates. They are disruptive to the creativity that is required to evaluate and manage multi-organ disease in patients on multiple medications. They are good for sandbox care, eg a simple one problem surgical situation, but otherwise, rigidly require workflow adjustments that are counter intuitive to effective care. They facilitate errors.

7. Margalit:
   “However, I am looking at all the requirements for “meaningful use” and most things in there seem to go to patient care, such as registries, CDS, sharing information with patients and other providers of care.”
   When I say patient care, I mean me sitting in the exam room, face to face with the patient. I see nothing in the meaningful use standards that would improve that interaction. All I see is a tremendous administrative load that means I need to see more patients to cover the overhead!
   To be more specific, my day typically begins with seeing five or six Type II diabetics, usually poorly controlled. I want to use my time with them talking about diet and exercise. The last thing I want to do is, 30 times every week, reading a prompt from the EMR about the latest data on diabetes, clicking to indicate that I’ve read the prompt, and then clicking to acknowledge that it is or is not relevant to the patient I am trying to talk to. What a waste of time and energy, and what a disservice to the patient.

8. I don’t think any physician who gets his patients’ lab results and radiology reports off the hospital computer (which is most physicians these days) can claim EHR’s are mostly made for billing. But I worked mostly in the lab, so what is it you clinicians really dislike – is it CPOE? Having to document your notes in the computer? What? Just come out and say it.
   Now physician office systems are another issue and one I know nothing about. But again, I would ask the doc commenters to be specific about what they don’t like. EHR’s have multiple different components. What is it you do like and what you don’t?

9. pcp, you are making a point frequently made by many docs (EHRs are designed for optimizing billings, not patient care).
   However, I am looking at all the requirements for “meaningful use” and most things in there seem to go to patient care, such as registries, CDS, sharing information with patients and other providers of care. Even recording of orders and receipt of electronic results can be viewed as an attempt to improve patient care.
   So what would you, or anyone else that cares to comment, consider a worthwhile feature in an EHR, a feature that will improve patient care?

10. Tsuris and pcp are spot on. These devices should be recalled until there is proof of efficacy and safety. The data at present prove no better outcomes and no cost savings.
    Equipment is being sold illegally. Best not to purchase.

11. The problem with EHRs is that they’re not designed to improve patient care: they’re designed primarily to deal with coding, billing, documentation, insurance issues, data collection, and so on.
    Good products sell themselves. Bad products are subsidized and mandated.

12. In response to your questions, Margalit,
    “So, tsuris, let’s take this a bit further. How do you propose the FDA test these software tools?”
    HIT caused cases of failed care and lost specimen, lost results, hidden results, death and other, as reported on MAUDE and in Florida are frequent and give the equipment a rightfully deserved “black eye”. The user is not in error, but is, instead, bewildered and ostracized by the unusable equipment.
    There needs to be candid in situ evaluations of CPOE and EMR equipment spanning the training, implementation, complications, defects, flaws, cognitive disturbance, usability, and chronic efficacy with efficient after market surveillance (if any of these devices would meet the safety and efficacy requirement of the FDA). Since the systems controlled by these HIT devices alter the entirety of communication and care of the hospitalized patient, their impact on each and every component of the complex interdigitated system of care requires detailed evaluation.
    Most importantly, the patients and doctors must provide consent to have their care and workflow altered and affected by these systems.
    The gag clauses between vendors and hospitals must go away, and all complaints, adverse events, confusion, disrupted communications be vetted.
    When compared to the systems of care these non approved devices are replacing, they may produce worse outcomes and more complications. No one knows but it is likely according to the front line users, doctors and nurses.
    They may be better, also, but why spend $ billions on devices that are of unporven safety and efficacy? HIM$$ and C$HIT propaganda is insufficient to support unfettered use.

13. Dr. Rosen,
    What I find most disturbing about the Florida incident is the following quote from Dr. Rosenberg:
    “This year we have seen as many if not more medical records violations from electronic medical records as we saw from hand-written records violations.”
    As to good EMRs, there are several small EMRs, some written by their users, some commercial, that consistently manage to satisfy customers. I just reviewed one on my blog http://bit.ly/d5D1NG and I was fairly surprised by the scarcity of complaints. Maybe because these small software packages have no pretenses of saving the world or reducing the US deficit.
    smart patient, I cannot speak for the HIT industry and the many appearances of impropriety it displays.
    In my opinion, the industry avoided and is trying to avoid even now any FDA involvement because the high road is more expensive than the low one. The most well intended argument against FDA oversight is about “stifling innovation”. I don’t buy that either. Innovation for the sake of innovation is silly, and harmful innovation is not a desirable thing to have.

14. If this is true–“I also think that the FDA has a role in ensuring patient safety.”
    Margalit, why have the HIT industry’s devices avoided any and all scrutiny?
    Were the authorites duped by the C$HIT certification or were there greenbacks being exchanged. The government is supposed to protect the patients, but the government has been commandeered by the self interested “intellectuals” from HIMSS and C$HIT amd others with a revolving door between ONCHIT, HHS and elsewhere. Is not N. deParle tagged to Cerner?

15. Margalit:
    You said: “I don’t think all EMRs out there are bad, some are actually pretty good, but this indiscriminate forced shopping spree will inevitably create large numbers of dissatisfied customers.”
    What proof do you have for this statement?
    Was the EMR involved in the missed uterine cancer case in Florida a “pretty good” EMR? See HIStalk report on this.

16. Merle,
    I wanted to think a little bit before I answered and I think I agree with some of what you wrote. I believe spending money to encourage HIT adoption is a good thing, but I think the money is incorrectly allocated. The Regional Extension Centers (REC) for example are a good idea, but their charter is too narrow. Maybe they should have gotten more money and be asked to also work with patients, particularly the underserved. I think the SHARP grants and the BEACON communities are a wonderful way to develop and test concepts and as you said, maybe more research money would have also been a good thing.
    I am not certain that throwing a few thousand dollars out there for each physician was the best idea. It created a flood of advertising and lots of pressure on doctors to buy products against their better judgment with absolutely no evidence that the docs are wrong and the HIT vendors are right.
    When you put so much money on the table, and during recession times too, even the honest and well intentioned are temporarily blinded by temptation. I think this is what is happening today.
    I don’t think all EMRs out there are bad, some are actually pretty good, but this indiscriminate forced shopping spree will inevitably create large numbers of dissatisfied customers. I don’t see how this helps with our long term goals of universal HIT adoption.
    I also think that the FDA has a role in ensuring patient safety. How and what exactly should be done may need a bit more thinking (thanks bev, I’m looking at the lab systems). I hope we take the necessary time to think things through before we act and then go back to rationalize why we did what we did.

17. Margalit;
    You post an interesting question about how the FDA should require validation of EMR software. As I said, this process already exists for blood bank software. Such companies as Sunquest and Soft Computer produce such products; if you have any contacts with them they will know how it’s done. I do not know the answer, from the buyer’s perspective.

18. It seems there is a delay with the adoption of every EMR system among all medical staff in my experience.
    W

19. So, tsuris, let’s take this a bit further. How do you propose the FDA test these software tools?
    Should they have the equivalent of clinical trials, and deploy the software and any subsequent modifications, in a controlled environment for, say, six months?
    Should they devise scripts for testing as many possible pathways (hundreds and thousands) and have each software execute the scripts in a laboratory environment before it is released for sale?
    Should we also have a “consumers reports” type of organization where owners and users can fill out surveys so the rest of us can benefit from their experience?
    Should vendors be held legally liable for any damages to patients caused directly by the software (it may add an extra incentive to do well)?
    Dr. Seker, if the EHR is diverting attention from the patient to degree you are describing, may I suggest that it is a pretty bad EHR. EHRs were not meant to interfere with the patient-doctor relationship, or with listening and taking thoughtful histories. We are no where near where we need to be, but we can, and we must, improve very quickly.
    Remember the first Motorola cell phones? It didn’t take that long to produce an iPhone, but it did take massive user participation.

20. Gee, I wonder why there are complaints about computerization of medical care. Doctors are supposed to be so smart that no matter what obstacles are put in their way, they should be able to cure. Big deal, they have to now type and click while ignoring the patient. The patient is just one big bag of biochemical results anyway that can be controlled from mission control.

21. Hey Tsuris__Why do you think these are devices and why do you think they are being sold illegally? Those are rather strong accusations.

22. Ms. Margalit,
    There have been many innocent people prematurely killed because of user unfriendly and flawed HIT devices. The word is out that these systems can not be trusted. They are impediments to effective care of the complex patient with multi system disease.
    Even the consulting docs at the Cleveland Clinic agree that tracking the course of clinical events and interventions to determine the diagnosis and best management of a patient in hospital for a week or more are disrupted by the pages of useless legible verbiage and cognitive blunting detail.
    The doctors work and devote intellectual energy to protect their patients from the new dangers posed by this illegally sold equipment.
    If there is any chance that doctors will embrace and promote this technology (rather than having it rammed into their brains), they need assurance that it has had a modicum of vetting for safety and efficacy, including usability (and not by deceptive HIT vendors and HIMSS subsidiary C$HIT), performed according to the requirements of the Federal Food Drug and Cosmetic Act.

23. Margalit, I’d like to offer a simple suggestion. Not only should we not add further government involvement — FDA or otherwise — but we should get government out of dictating the adoption of and approving EMR systems.
    The legacy EMR systems available today have failed to meet the needs of docs and hospitals for years, which is why so few docs and hospitals have adopted them. They have spoken with their proverbial feet. Now, in what seems to me to be an irrational act, the government has decided to force docs and hospitals to adopt these very systems and will subsidize the cost of the EMR system — ignoring the fact that the all-in cost for a doc can run six or seven times the cost of the EMR software!
    In short, government involvement is trying to force an unacceptable solution on physicians and hospitals. From what I read, I suspect many docs and perhaps hospitals will rebel and refuse to oblige.
    My suggestion is for the government to nurture the development of new approaches to EMR systems by making grants to entrepreneurs who appear to offer a promising, financially sound EMR system.
    The criteria for making such grants should be the projected economic viability of the EMR system. If the system can demonstrate a high probability of acceptance by docs, it most likely will be financially viable, and a grant should be approved. If it can’t demonstrate that docs will adopt it, it should not be funded.
    Such funding will result in the reasonably quick development of products that actually help and meet the needs of docs and will not require government subsidies.

24. tsuris, it’s funny that both tsuris and zealotry have their origin in the same part of the world….
    I guess it depends how you define an HIT zealot. If it means somebody who believes that proper technology when properly used can provide measurable benefits to health care, than I plead guilty.
    I do agree with you that usability, both in the context of patient safety and in the context making a doctor’s work easier, were grossly neglected in the past and I also agree that past CCHIT certification did not and could not measure either one. Furthermore, the new and improved minimalistic Meaningful Use certification does not measure these factors either. As pcp said above, these sort of certifications cannot really measure usability, let alone patient safety.
    As implausible as it may sound to you, I would like to know what you think we should do moving forward. I tend to agree with bev MD on the need for FDA involvement and I also think we need to slow down a bit, but I am really and truly curious whether you have any suggestions other than trash the whole thing and stick with paper.

25. Margalit said “So what do we do? We could choose to stop the effort to implement HIT. We could choose to have HIT go through FDA approval processes. We could just go ahead and hope for the best. We could try to think creatively and constructively and come up with a workable solution. I was hoping this post will trigger such conversation here…..”
    I doubt you were. You are an HIT zealot, but never have to use it.
    There are patients dead because of the crap that is being sold and forced upon the doctors and nurses. Best advice for all: do not buy until they are approved as being safe and efficacious.
    The vendors have ignored issues of usability to the detriment of patients. They knew and deliberately chose the path of deception, using CCHIT Certification as the vehicle of fakery.
    This is criminal.

26. I guess the take-home message for the one system that was rated “3” is that they better get serious and do some major improvements on their vendor demo.

27. Margalit,
    Thanks for your balanced perspectives.
    It will be obvious to Joe the Doctor that CCHIT is destroying what little is left of their credibility with their new usability rating system and process.
    The fact that 18/19 of vendors can get a 4 or 5 usability score doesn’t pass the straight face test….
    If there is a doctor on the planet that finds this information useful in making an EHR purchasing decision, would you please identify yourself and help me understand what I’m missing.

28. I am increasingly of the opinion that at least parts of an EMR do constitute a medical device and should be so regulated by the FDA. As I have mentioned before, there is precedent for this in their regulation of blood bank/transfusion service software. I recognize the delays and bureaucracy this will add to this process, but it works quite well with the blood bank software. Besides, the onus should be on the vendors to prove that, as Margalit says,
    “some clinically pertinent information is, for example, not displayed when it should be or there is a software bug manifesting itself in rare circumstances and corrupting actual data.”
    We have actually seen this happen while implementing new laboratory information systems, where critical pieces of information in, say, a microbiology report are omitted. This is particularly problematic in interfaces between the LIS and HIS. Why should the hospitals have to find these bugs for the vendors?

29. Thanks.
    As if Consumer Reports rated cars by watching advertisements, but never actually driving them?
    It’s measuring something, but doesn’t sound like it’s something worth measuring.

30. pcp,
    From CCHIT’s website http://www.cchit.org/participate/jurors
    “For its CCHIT Certified comprehensive program inspection, the Commission empanels a team of three clinical jurors, one of whom must be a practicing physician, and an IT security evaluator to assess a technology’s conformance to the CCHIT certification criteria. The inspection occurs by observing the performance of the applicant’s technology in executing a series of test scripts and reviewing required materials supplied by the applicant.”
    However, as I said before, this is much like watching a scripted, and rehearsed ad nauseam, vendor demo.

31. Who are the jurors?
    If they’re not docs actually in the office and on the floor seeing patients, this is just rating sales pitches.

32. Gary,
    The rules are in the Usability Guide link above. Here it is explicitly http://www.cchit.org/sites/all/files/CCHIT%20Usability%20Testing%20Guide%20Ambulatory%20EHRs%202011%20v6.pdf
    If I remember correctly, a typical CCHIT inspections takes about 8-10 hours and is done over Webex with the vendor driving and the jurors watching. The usability ratings are performed during and immediately after the inspection.

33. propensity, let me address what I believe you are asking, not what you actually typed in.
    We can measure basic design features that are prerequisite to usability, like whether fonts are readable, buttons are consistently labeled and consistently placed on screens, checboxes are aligned, response time, etc. However, a software product having satisfactory grades on all these basic items, is not necessarily usable, or not judged equally usable by different users. For example, I prefer clicking to scrolling and I know that other folks prefer the opposite, and there are many more such examples as Michelle also indicated. This is problem #1.
    Problem #2, which is much more explosive (as the HIStalk thread bev mentioned certainly was), is the contention that lack of usability kills people. There is absolutely no way to test that during a certification inspection. It may be that the screens flow nicely and all the buttons are as they should be, but some clinically pertinent information is, for example, not displayed when it should be or there is a software bug manifesting itself in rare circumstances and corrupting actual data. These sort of usability measures can only be assessed by long term use in a real live environment by clinician users.
    For ambulatory products, the AAFP for example is running surveys and publishes results regarding user satisfaction. This may be a better usability assessment, but still imperfect.
    All in all, I don’t believe either CCHIT or the ONC/NIST can really and truly certify usability. They can certify the basic prerequisites, and maybe they should, but just like a car that passes inspection can turn out to be a horrific lemon, an EMR certified this way may turn out to have fatal flaws.
    So what do we do? We could choose to stop the effort to implement HIT. We could choose to have HIT go through FDA approval processes. We could just go ahead and hope for the best. We could try to think creatively and constructively and come up with a workable solution. I was hoping this post will trigger such conversation here…..

34. Where can one find the usability indices? I agree that true usability cannot be judged until six months after installation. Is that how they were rated? A one or two hour webex or even onsite evaluation is meaningless.

35. Propensity is correct: CCHIT is a scam. CCHIT Jurors are sworn to secrecy. Their tests are meaningfully useless for usability. It is an organization of HIMSS to create a facade of legitimacy for the defective care products they are promoting.

36. Propensity,
    I don’t know if you have a serious question here or if you are just trolling. I will assume (perhaps incorrectly) that you have a deficit in reading comprehension and ask you to re-read the article and note: the article explained CCHIT usability testing in detail as well as giving references (those “linky” things) to the usability testing documents (which contain further links to additional documents). If you can’t follow these, then I suggest Google which I have found to be very useful. A simple search for “CCHIT Usability” returns a wealth of information.
    If your comment was intended to make some other point then I can’t help you.

37. Congrants on appearing on The Health Care Blog, Margalit. I had read this one over at your blog, but it’s still good food for thought. Usability is a tricky thing: for example, I’m happy to just pull up an empty Notepad file, type some HTML, and then post it online. Of course, that’s not how most people would want to keep a blog, but I don’t like using programs that “simplify” to the point that typing a simple sentence requires clicking five different buttons. That type of system is “unfriendly” to this particular user.
    Plus, what’s most important for one practice might not be for another. What parts of an EHR should be given highest priority (i.e. less clicks to get to) and what should not? How do you make a structure general enough for anyone to use but also able to be customized for the different types of practices out there? What about color, contrast, font size, screen resolution, popups, and code language requirements? Touchscreens or mouse functions? Keyboard shortcuts? And finally, how do you make data easy to access and use, but still consider privacy and security?
    @propensity: I believe Maraglit summarized the CCHIT testing process starting in paragraph 4, with a link to CCHIT’s own testing guidelines for further reading.

38. Does the author know anything about the CCHIT process of testing usability? Please explain how CCHIT does that to us naive users.

39. Thank you for this posting. It’s the CENTRALl issue in rapid adoption or use of physician EHR’s. Unfortunately, I agree w/ John. If those system are five stars, then we need another fifteen stars in the rating system.

40. Margalit;
    Thanks for a fair post from a trustworthy expert. I’m sure you have been following the Epic brouhaha as exposed in HIStalk 6/2/10 and on the Health Care Renewal blog. There seems to be great disagreement on how subjective or objective usability measurement is, but from my computer-dumb user perspective, it’s pretty darned important to us. And you are absolutely right about vendor demos vs. real life experience. Our institutions got caught on that multiple times.
    I don’t really know what the answer is for the average buyer, a hospital or dr’s office who has little comprehension of technicalities and much vulnerability to sales tactics.

41. It is absolutely astonishing that Epic’s EMR received 5 stars for its usability. That rating alone should trigger an investigation of CCHIT’s business practices.