HIT Apologia
Health Affairs recently announced its top 15 articles for 2013, and has made them available to nonsubscribers.
The top article was by a pair of RAND researchers updating what is known about the health information technology (HIT) roll out from the 2009 HITECH law, appropriating $20 billion to upgrade information technology throughout the health care system.
It doesn’t take long ― like just the abstract ― to figure out that people haven’t learned a blessed thing from flushing $20 billion down the toilet. Here’s the complete abstract with my comments −
A team of RAND Corporation researchers projected in 2005 that rapid adoption of health information technology (IT) could save the United States more than $81 billion annually.
This original “study” was horrendously flawed. They deliberately chose (and said so) to ignore any contra information, basing their projections on a best possible scenario that couldn’t possibly come true in real life. In the latest report:
Seven years later the empirical data on the technology’s impact on health care efficiency and safety are mixed, and annual health care expenditures in the United States have grown by $800 billion.
I’m sorry, so sorry
Please accept my apology
They are still minimizing the evidence. There is plenty of evidence that the HIT adoption process has increased errors and slowed productivity. Rather than “mixed” the evidence is overwhelmingly negative. Again, from the latest report:
In our view, the disappointing performance of health IT to date can be largely attributed to several factors: sluggish adoption of health IT systems, coupled with the choice of systems that are neither interoperable nor easy to use; and the failure of health care providers and institutions to reengineer care processes to reap the full benefits of health IT.
They have it exactly backwards. Rather than reengineering clinical practice to suit the demands of IT, IT must be developed to enhance what is happening clinically. This is fundamental to any acceptance of IT. There is a reason programmers used terminology like “desktop”, “inbox” and “trash” in creating office systems. You have to start with what the users are familiar with doing and adopt your technology to their comfort zone. The problem with this whole top-down HIT enterprise is that they put IT engineers in charge instead of letting clinicians develop what would improve their operations:
We believe that the original promise of health IT can be met if the systems are redesigned to address these flaws by creating more-standardized systems that are easier to use, are truly interoperable, and afford patients more access to and control over their health data. Providers must do their part by reengineering care processes to take full advantage of efficiencies offered by health IT, in the context of redesigned payment models that favor value over volume.
NO! Start with making it work in the doctor’s office and THEN build interoperability. DO NOT standardize it from the start, but allow a variety of systems in real-life settings and see what works best and THEN move to standardization.
Now, it might surprise you to learn that abstracts rarely reflect what is actually in the paper. In fact, the abstract is typically written by the editors of the journal, not the authors of the article, and these editors often have an agenda of their own. People who confine their reading to abstracts are doomed to be misinformed.
So, let’s look a little more closely at the article itself.
After a bit of tap dancing around the issue, the authors of this article, Arthur Kellerman and Spencer Jones, tell us −
…[I]t needs to be noted that the (original RAND) researchers attached the following caveats to their analysis early in their article: “Here we summarize the methodologies we used to estimate the current adoption of [electronic medical record] systems and the potential savings, costs, and health and safety benefits. We use the word potential to mean ‘assuming that interconnected and interoperable [electronic medical record] systems are adopted widely and used effectively.’ Thus, our estimates of potential savings are not predictions of what will happen but of what could happen with HIT and appropriate changes in health care.”
So, the original estimates on how well HIT will work are based on the assumption that HIT will work well (be widely adopted and used effectively). It’s not their fault that the health care system did not find all this attractive — even though virtually all the evidence in existence at the time predicted exactly that.
And, sure enough, the new article ticks off the failures one by one:
- Are modern health IT systems inter-connected and interoperable? The answer to this question, quite clearly, is no.
- Are modern health IT systems widely adopted? The answer here is no as well.
- Are modern health IT systems used effectively? Again, the answer is no.
- Has appropriate change in health care been made? Sadly, the answer here is no.
The authors explain a bit of this by writing:
Considering the theoretical benefits of health IT, it is remarkable how few fans it has among health care professionals. The lack of enthusiasm might be attributed, in part, to the sobering results of studies showing that in many cases health IT has failed to deliver promised gains in productivity and patient safety. An even more plausible cause for providers’ lack of enthusiasm is that few health IT vendors make products that are easy to use. As a result, many doctors and nurses complain that health IT systems slow them down.
Nonetheless, these writers continue to insist it is not the fault of the IT industry or the government, but of those damned providers who fail to “reengineer existing processes of care to take full advantage of the efficiencies offered by health IT.” So, doctors should practice medicine the way the IT people tell them to instead of having the IT people work to help the people who actually provide the care.
There was nothing wrong with RAND’s initial analysis, they write. The analysis would have worked out fine if the vendors and providers had fewer “shortcomings” −
The optimistic predictions of Hillestad and colleagues in their 2005 analysis of the potential benefits of health IT have not yet come to pass. This is not because of shortcomings in their analysis but rather because of shortcomings in the design, implementation, and use of health IT in the United States. When the preconditions these authors posited are finally realized, the benefits they predicted will be realized as well.
And we would all live in Paradise if only we had fewer shortcomings.
“They deliberately chose (and said so) to ignore any contra information, basing their projections on a best possible scenario that couldn’t possibly come true in real life.”
Of course they want their projections to meet the best possible scenario in their study to pass such a thing. However, ignoring information that would negatively affect their projections is a flawed way to go about it.
“This is not because of shortcomings in their analysis but rather because of shortcomings in the design, implementation, and use of health IT in the United States”
Don’t dare blame their analysis, but rather the shortcomings of implementation. Yes because their analysis was surely not flawed.
Surely they would’ve included any sort of shortcomings in their analysis. Or maybe not…
“In our view, the disappointing performance of health IT to date can be largely attributed to several factors.”
In their view, all of the factors are based on blaming the providers for not correctly adopting the program as opposed to enhancing what the providers already have.
Less than a decade ago there was an article in Health Affairs by RAND researchers that estimated $77 billion in savings from HIT if a 90% participation rate was achieved. See Richard Hillestad et al., “Can Electronic Medical Record Systems Transform Health Care? Potential Health Benefits, Savings and Costs,” Health Affairs, Vol. 24, No. 5, September/October 2005, pages 1,103-17
“NO! Start with making it work in the doctor’s office and THEN build interoperability”
“You would think this is how they would start versions of these systems. Make it work first, then make it standardized instead of having those adopt an unfamiliar system.”
“Considering the theoretical benefits of health IT, it is remarkable how few fans it has among health care professionals”
Remarkable indeed. How could they not see the THEORETICAL benefits?
Obviously, everything works theoretically. In writing everything seems beautiful, flawless. We normally picture a utopic scenario. We don’t want to think in what could go wrong, we tend to highlight the positives and forget about the negatives. In theory even Communism works. But reality is different. Good policy makers are those who seek attainable goals, not those who legislate on the premise that everything will go perfectly according to plan. It is a pity when we realize that theoretical benefits are not achievable.
The final conclusion is amusing. Everything would be better if there were fewer problems. What a wonderful realization. It is true, if everything goes as planned and there were no issues we would live in a perfect world. The problem is that we spend too much time pointing out errors and not actually trying to solve them.
There is a spreading aphorism in the medical community, at least among those using electronic medical record (EMR), that one is “treating the computer” and not the patient in the exam room or wherever the device is located.
The combination of more and more limited time with a patient and need to meets the needs of the EMR – it needs to be fed data, mostly, to satisfy the billing people and the insurance companies -means increasing difficulty in trying to do a good job.
I’ve taken to using my slightly better than average (at least average physician) typing skills to “free text” clinical information in the chart because some of these systems are useless in collecting clinically useful and understandable information about a patient.
Except for allowing organizations to bill at a higher level the EMR in the main, as currently used, is a huge waste of money and time.
“NO! Start with making it work in the doctor’s office and THEN build interoperability. DO NOT standardize it from the start, but allow a variety of systems in real-life settings and see what works best and THEN move to standardization”
Bu-bu-bu-bu-but that would be MANAGING.
This administration clearly does not know how to manage anything but a campaign.
Which is probably why the Obamacare advertising campaign eagerly promised how great it’s gonna be – – but when the campaign is finished, so are the promises.
It is amazing that we ever thought that we would get anything useful out of a Federally sponsored EMR inititative.
The Federal govt could have solved this problem years ago by just requiring that any interaction/procedure/process that was paid for by Federal dollars…Medicare…be sharable with patients electronically. Then the vendors would have to open up the back end of their systems…or…place data in retrievable files. If retrievable by patient…it is by definition…retrievable by other physicians if they have a key/password.
At some point this becomes morally reprehensible.
In the modern world IT has significantly increased its importance. Everything is electronic nowadays; there are systems that must be maintained and new programs to be written daily. Every industry is gradually more reliant of IT, which has brought many problems. People who work in IT are very skilled, knowledgeable about systems, and they are capable to adapt their knowledge to support a variety of industries. The problem is when IT believes it has the knowledge to change the industry. Using as an example healthcare. As the article states, they want the industry to adapt to what IT can do, not the other way around. Those in the health care industry studied to become experts in the field, they know what works best. They should be the ones dictating what is included in HIT and what the best way for them to use it is. Not the other way around. IT doesn’t know how doctors treat their patients, they don’t know what should be written down in a patient’s medical history or how. They why are we asking doctors to adjust to a program written by those who don’t know the field. Let doctors be doctors, let IT worry about IT. Realize that IT works for the doctors; it is the doctor’s job which it is supposed to support. It is not the job of the doctors to make IT’s job easier.
It seems as if the United States was determined to waste money. Faulty after faulty legislation have been passed in the recent years that hasn’t achieved anything. Taxpayer dollars are being burned every time there is a new law. I haven’t heard about a law that has actually solved something in the recent years. This administration will go down in history as a big spender, one that wasted its money when the country’s debt reached critical levels.
On the other hand, every penny of that $20 billion HITECH money (and all of the other wasteful programs you can think of) ends up in somebody’s pocket. So if your goal is to enrich your friends at the expense of taxpayers, this isn’t “wasted” at all. The law becomes just an excuse to give (certain) people money. Those people in turn give a portion back in the form of campaign contributions. Good deal all around.
Which is why….we have the best govt money can buy!
HI Greg,
Ever since I started writing and speaking about electronic medical records (which was 1993) I have always said that “the state of the art isn’t”. We still don’t have a sensible system that works the way it should. Greed over who “owns” the record has prevented standardization and the kind of development we need.
The push for the electronic record, however, is allowing massive data collection and mining. It is permitting payors to require providers to turn over a patient’s complete chart just to get paid for an office visit for the sniffles. (And people worry about the NSA . . . .)
The noble French put it all on a smart card a long time ago. They had no fight over who owned the data and got the project done. We are a long way from that uniformity.
There are, however, some very cool innovations in the pipeline. One that works is CareScreen which is an adjunct of the only national ACO. It is a handy tool linked to deep informatics. It allows real time case management, and “keeps score” on how the provider is doing on gainsharing. It is a totally useful, relevant, and streamlined little “app” that has proven HIGHLY effective in promoting cost-effective care. Indeed, its use has been proven to save millions. It isn’t a complete record, but does all the heavy lifting needed to succeed under the new payment models while boosting the quality of care the patient receives.
Another outfit in San Francisco is trying to move the “electronic record” to an offshore scribe system using Google Glass. While Glass solves the problem of the doctor never looking at the patient, it is remarkable that the written record has become so important (legally and to the payment processing) that a typewritten text has to be composed, sent overseas and back–when a complete videographic record of everything that transpires could be created and stored simply by turning on Glass’s video camera. Hmm. In the future, the potential telemetric uses of Glass are obvious.
When I started practice in 1974, the medical record was comprised of the doctor’s illegible scribbles simply sufficient to remind him or her of what happened during the patient encounter. It has since evolved into a defensive medicine document with enormous potential liability attached to it. In many states like California, the failure to maintain an “adequate” medical record is unprofessional conduct that could lose a doctor his or her license. It goes without saying that the record is no longer written for the doctor, but for the payor and for 12 unknown potential jurors who one day might judge the care and treatment received during one of the 600 million medical encounters that occur every day in this country. No amount of automation will reverse the absurd lengths to which we have taken the simple “note.” It has been elevated beyond all reason. So, let’s face it, the electronic medical record now seems to be coming back to BYTE us in the . . . .
Have a great day,
Charlie Bond
I have thought for a while that with cameras getting cheaper and smaller, and data storage so cheap, that we are headed to 100% AV recording of all patient encounters.
Look where the police are going. We are next.
I spend a lot my time on patient safety issues.
IT has proved to improve patient safety in a number of ways. It adds some time and costs to diagnosis, treatment, and rehab but the ROI in better, safer, less costly outcomes — including lives saved — is worth it.
The article and comments suggest that physicians and health care professionals were not consulted when more IT was implemented into healthcare. Maybe they weren’t the right ones to consult, but that the whole process was controlled by tekkies, insurers, and b’crats seems hard to believe. I welcome comments to correct my perception.
Jan,
I don’t know were you get your impression. Even these authors (HIT apologists) report — “in many cases health IT has failed to deliver promised gains in productivity and patient safety.” Please provide some evidence for your view.
Excellent article, Greg.
When I first wrote about the value of electronic records to document outcomes in chronic disease in 1975(Archives of Internal Medicine), those of us gathering the data at Duke knew that we had to devise entry systems that would not increase the world load of doctors and that would give them reports that would save them time in dictation. All of that succeeded but the insurance industry didn’t want to pay for outcomes data and the Robert Wood Johnson Foundation decided they didn’t want to put any more money into the use of computers in medicine. As a clinician, I have used several EHR systems and all make life miserable for physicians, encourage upcoding, don’t insist on common definitions ( What do you mean when you say “Myocardial Infarction”?) and are not searchable in terms outcomes. But the insurance companies want them because they are easier to search for pre-existing conditions and great for defending medical malpractice actions because even though a doctor used the term “fibromyalgia” incorrectly or not in reference to any recognized definition, that term can be used to discredit any plaintiff complaining of pain as a result of alleged negligent acts. We have a mess. We have gone backwards since the 1970s in terms of electronic records. At least the size and price of the computer have come down dramatically, but using the machine to help doctors and patients evaluate prognosis has barely inched forward. For those who don’t remember, the use of the computerized cardiovascular databank at Duke led directly in the 1970s to the precipitous drop in hospital time for acute MI, because it was possible to define subgroups of uncomplicated acute MI patients who could be safely discharged early as opposed to being kept in the hospital for 3 weeks or longer. EHRs have been used as glorified word processors that have made a lot of money for vendors and consultants and easy fodder journalists intent on bashing doctors as being incompetent greedy Luddites.
IN the early 90’s the hospital lab I was medical director of adopted an upgrade of a popular lab system. When I tried to point out the problems the reaction was that I was the problem, I needed to shut up and do as the system vendor instructed. There are many comparisons of health care to aviation, mostly to call attention to the admirable safety record of (commercial) aviation. A fundamental difference between aviation and health care is that aviation was governed from the start by an agency which dictated the vocabulary used by pilots and controllers. A great many errors in health care result from ambiguity brought about by terminology with vague meaning and even a lack of clear thinking enabled by vague terminology. It’s damn near impossible to see any way to clean up the mess unfortunately
Seems no one in the administration learned anything from the UK experience:
The costly programme to make patient records digitally available to every hospital, originally scheduled for completion in 2006, will not be ready until at least 2015. Even if the NHS meets that deadline, the system is certain to be obsolete the moment it goes online. University of London Professor Trisha Greenhalgh explained it best:
“Depressingly, outside the world of the carefully-controlled trial, between 50% and 80% of electronic health record projects fail – and the larger the project, the more likely it is to fail. […] Our results provide no simple solutions to the problem of failed electronic patient records projects, nor do they support an anti-technology policy of returning to paper. Rather, they suggest it is time for researchers and policymakers to move beyond simplistic, technology-push models and consider how to capture the messiness and unpredictability of the real world.”
http://www.theguardian.com/commentisfree/cifamerica/2011/apr/23/nhs-health
Excellent essay as usual, Greg.
“Bruce Landes, MD, is a confirmed skeptic when it comes to electronic medical records. But he insists he is no technophobe. “I am not a Luddite,” Landes says, working up some steam on a topic he knows well. “Physicians are people who have been through the development of CT, MRI and robotic surgery. We are not afraid of technology, but we are afraid of bad technology.”
That was a quote of mine in an interview from an article titled “EMR Wars” on the subject in 2007. http://www.healthleadersmedia.com/print.cfm?content_id=92119&parent=107
I think that the HITEC act (part of ARRA 2009) made things worse by throwing more money into health IT. That extra money helped keep bad companies and programs alive. Without the stimulus of HITEC, the EMR market would have narrowed to a few IT companies with user-friendly, interoperable products. When you are headed in the wrong direction, the last thing you need is more speed.
Remember how quickly the dial-up modem industry narrowed in the 1990’s to two proprietary, incompatible standards 56Kflex (Rockwell and Lucent) and X2 (US Robotics) that then, by mutual consent became the V.90 interoperable standard in 1998? No government money was injected and the market fixed the problem.
Further from the same article:
“As president and CEO of Southwest Physician Associates, Landes has done a fair share of analysis of EMR technology, describing it as “a good example of bad software.”
“You have to realize that physicians have been trained four years in med school, then three to seven years in post-graduate training,” Landes says. “The funny thing is that they want to take care of patients. They don’t want to become specialists in creating medical records. They look at the medical record as an incidental cost of doing business. Many EMR programs act as if the medical record is the whole point of the patient encounter. It is just not.”
Good post, Greg.
Nice post, Greg.
Question I always have is what do we need additional electronic records for? Do you really need to know whether I had my eyes checked 10 years ago?
The US health system was already one of the most wired in the world before Obamacare passed. (http://healthblog.ncpathinktank.org/surprise-finding-the-u-s-has-more-health-it-than-other-countries/) Leave people alone and they’ll use electronic records where they make sense. And offer electronic records for people with complex diseases who need them as Lynx Care did.
For those who think that it doesn’t matter if records reduce medical productivity, consider that a loss in medical productivity means that health care costs more. If health care costs more and incomes aren’t rising, either some people don’t get treated or people have to go without some other necessity.
We certainly cannot say that decreased productivity doesn’t matter without knowing what people have to give up to put up with it. Especially as the US medical school cartel refuses to graduate more physicians despite a vastly expanded population, and has instead made a conscious decision to enforce quotas resulting in the admission of more people who will work fewer hours over their lifetimes.
Great article comment about the myopia permeating the Rand Article. Although the landscape is littered with dead medical record companies, from the ashes arise a never ending stream of the same, thereby fulfilling the definition of insanity commonly attributed to Einstein. However, while engineers, scientists, architects and others have managed to maintain control of their own profession and restrict the coders and developers to producing computer programs which assist at their daily tasks, in medicine the software developers has decided that they know best how the doc should practice medicine and the docs need to restyle their practice patterns to get with the program. The end result is decreased productivity, alienation from the patient and frustration on the part of the caregiver. Yet there are countless ways in which computers can help docs, and would be welcomed. They just cannot replace the doctor or the office and everyone looses when they try.
The original RAND study was funded by GE and Cerner! What a surprise that the conclusion was so flawed, in the favor of the vendors.
Also striking is the absence of any oversight for safety, efficacy, and usability.
They not only waste money, but they kill patients who otherwise would not have died under the system of care that was displaced by these toxic devices.
Not long ago I destroyed over 200,000 individual notes that were not in cyberspace and nothing happened. What a shock.
I am sure 10 year old eye reports were destroyed as well. In fact after the second or third year of storage the desire to obtain those original notes became non existent.
I met you at the AAHKS meeting two or three years back. I am nearing the end of my career as an orthopedic surgeon but in a former life I was a systems analyst for UniRoyal tire company and wrote programs to facilitate workflow in laboratories. In my not so confident youth I went to the actual workers and asked how they did their work and developed a program that took hours out of the lab’s workday and as a bonus produced nice reports for management.
Athena is the provider of the IT software for our ten surgeon group and we believed many of their hollow promises. I have offered my services free to help them improve what is a very awkward and inefficient user interface but they were not interested.
Your article hit the proverbial nail on the head. Thank you for seeing the obvious.
Great article, Greg.
The idea that doctors should “reengineer” their practices to match some IT design is ridiculous. The person who wrote that should consider whether their academic enterprise had to be reengineered to suit electronic grading or whether the IT processes are trying to match the academic environment. HINT: it’s the latter. Similarly for grocery stores, auto repair shops, and manufacturing floors. The IT system design should always match the business problem, not the other way around. Enterprises that stick to systems because the IT guys say it’s going to be that way are businesses whose lunch gets eaten by smarter, more nimble firms.
It’s the same problem we face in the wider health policy and, indeed, political economic discussion: do you let the best ideas compete, or do you enforce the ideas experts think are the best ones?
These comments are spot on. Past technology has failed to recognise the need for Continuous Process Improvement and thus frustrated the clinical teams by not providing systems that match their workflow and being shackled in getting modifications.
Our approach we call DREWP (Design-Redesign-Expand-without-Programming). The aim is to get the clinical team to Design as much of their own system as far as it feels the team can cope with – Redesign it with experimentation so that it reaches an optimum and then install the current development. Subsequently, after using their design for some time, they can Expand it with the next set of features that will produce the greatest productivity gain for the least effort. In this way the technology adoption begins small and grows over time. We further support the clinical team by providing a tool (and process analysts to aid them) that is used to design the system without the need for programming as the design is turned into an operational system immediately, by our under-the-covers-code. Hence the team make changes in near real-time.
Our testing shows that a system for ED built in this way was 40% more efficient than a rigid proprietary system and saved 23.9 hours staff work (equivalent to 3 staff) per day on an average case load of 165 patients per day.
Jan Peter Ozga says: “I welcome comments to correct my perception.”
By way of background – MD here, background in computing dating to 1970, postdoc in Medical Informatics at Yale 1992-4.
How about dead bodies to alter your perceptions?
1. Sweet case – http://hcrenewal.blogspot.com/2011/09/sweet-death-that-wasnt-very-sweet-how_24.html – settled for $1,375,000.
2. Baby deaths due to interference in care by bad health IT “Babies’ deaths spotlight safety risks linked to computerized systems”- http://hcrenewal.blogspot.com/2011/06/babys-death-spotlights-safety-risks.html – $1,500,000 settlement
3. Another baby death due to distraction “Thanks to the wonders of EHR, this premature baby went to the grave, prematurely” – http://hcrenewal.blogspot.com/2011/03/thanks-to-wonders-of-ehr-this-premature.html – $1,000,000 settlement
I’d give details of my mother’s case of EHR-related injury and death, but the case is in litigation. She died in 2011.
In the meantime, do some study at my academic site “Common Examples of Healthcare IT Failure” (including the linked ECRI Deep Dive study) at http://www.cci.drexel.edu/faculty/ssilverstein/cases
Also see my populist writings at the multi-author Healthcare Renewal blog. Google it.
I especially want reaction to the ECRI Institute PSO study as linked at my Drexel site – 171 “IT incidents” serious enough to cause harm, voluntarily reported from 36 hospitals over just 9 weeks, with 8 reported injuries and 3 possible deaths as a result.
And perhaps unionized nurses, protected by their union, letting the sun shine – http://hcrenewal.blogspot.com/2013/11/another-survey-on-ehrs-affinity-medical.html
Any statements that “paper causes error too” must be accompanied by the ethical guidelines that permit experimentation in medicine without knowledge of a new intervention’s risk profile (per the IOM study), and avoidance of patient consent.
When my husband and I imported a medical record system for the Mac in 1984, we found a cultural problem; Doctors thought that computers were for clerical employees; they would feel diminished if they had to learn to type. So much for efficiency.
People seldom appreciate the value vacuum; the nearly total lack of benefit to management decisions, compared with the cost of accumulating data points. Support for clinical decisions is the supposed objective, but data for budgets, staffing, reporting, equipment planning, and so on are not summarized with ease in most IT systems. Simulation of future operations is not an IT objective in most cases. Moreover, from actual experience, use of computers in patient rooms means that the nurse has her back to the patient.
Excellent post, Greg. And thanks, Charlie, for your information on Care Screen and Google–and I agree about Google Glass. Coming soon to the operating room near you.
Wanda J. Jones, President
New Century Healthcare Institute
San Francisco
Correct me if I am wrong: the original Rand study was commissioned by the Institute of Medicine and largely funded by General Electric and Cerner. I also believe that it was IBM who originally perfected the bait and switch tactics of systems development. What can I say? Silicon valley has us all by the tail!
The President went to UnitedHealthcare for help for the exchange. I suggest a better source for healthcare systems development: Walmart. I call Walgreens about an Rx written for a patient. After 3-4 minutes of small talk, I get a “sort of” answer. After asking the same question,the same call to a Walmart Pharmacy takes 4-5 seconds to receive a precise answer. The Walmart folks are driven by efficiency for the end-users. The Rand folks are not.
This post and the replies became a giant gab-fest about a topic that physicians have always secretly wanted to go away. The elephant in the room is the fact that it is very difficult to hide errors from a proper system of EMR, and that creates tremendous liability for the practitioner. Given the present state of affairs with no tort reform in the horizon, I would be commenting in exactly the same vein were I an MD.
To those that commented that the process should have worked from the bottom up using whatever disparate systems exist now, and later attempted to standardize, those folks have obviously never attempted to herd a thousand cats.
A lot of people who commented failed to see the idea of what properly designed IT systems could bring to bear in terms of clinical results. 40 years ago as an incipient computer engineer, I tried to convince an old country doctor (my dad) about the benefits of computer assisted diagnosis. He shot me down, but in fact, computer assisted diagnosis with a database fed by thousands of inputs from wise physicians, can out diagnose any human being.
This summer a client of mine who also uses a VA clinic had a series of tests performed by a well known and excellent primary care physician who pronounced him healthy. A week later he went to his VA clinic for a routine checkup performed by a crass rookie VA nurse practitioner with a world class EMR system behind him. The rookie called out a potential cancer which when double-checked by his Primary, was found to be a correct diagnosis . It was not brilliance by the rookie nor lack of diligence by the expert. It was simply an EMR system noticing a trend line in blood test results over multiple years that proved impossible for the expert human to detect.
“40 years ago as an incipient computer engineer, I tried to convince an old country doctor (my dad) about the benefits of computer assisted diagnosis. He shot me down, but in fact, computer assisted diagnosis with a database fed by thousands of inputs from wise physicians, can out diagnose any human being.”
Interesting but I wonder if you had told your Dad he had to “re-engineer” his practice to accommodate your software how that would have gone over. Plus how do you know those thousands of physicians are any wiser than your Dad was.
This is a particularly myopic view of the purpose of the Clinical Information System (aka EMR). The modern clinician wants workflow system that assists in the activities of caring for patients. That means many functionalities are needed quite apart from the predictive modelling cited here.
The gabfest as you decry it is bringing attention to the fact that current systems fail miserably at supporting the practicalities of work but rather make work more inefficient.
You notion of a “properly designed IT system” is also entirely naive. Each clinical team has to work in the context of many variables that cannot be entirely predicted and are certainly not amenable rigid control unless of course you want to turn healthcare into police activity.
Gitmoray, you writ:
“It was simply an EMR system noticing a trend line in blood test results over multiple years that proved impossible for the expert human to detect.”
That is highly unlikely. Please specify the cancer and the “blood test results over multiple years” that this cybernetic miracle detected, plus evidence a human could not have detected it.
also, a “potential cancer” is not a diagnosis. It is a theory.
Current electronic record systems are garbage bloatware. I speak as a pediatrician who hasn’t adopted them, but gets plenty of notes from people who have.
I see a kid with a diaper rash and it generates a 1/2 page note. I get a note from the urgent care about my patient with the diaper rash who was taken there and get 5 pages of notes, all of which are pointless except one paragraph at the beginning and one paragraph at the end. How is this in any way helpful to patient care?