An ongoing dialogue on HIV/AIDS, infectious diseases,
September 19th, 2024
How Electronic Health Records Tyrannize Doctors — ID Doctors in Particular
A paper just appeared in the Journal of General Internal Medicine entitled “National Comparison of Ambulatory Physician Electronic Health Record Use Across Specialties.” The goal of the study was to track clinician workload by specialty, divided into various functions — documentation, chart review, orders, inbox.
Importantly, there was no gaming the system. By using Epic’s built-in function, they tracked “active” EHR time (any mouse activity or keystrokes) using a 5-second inactivity timeout. They additionally measured time spent on the EHR outside of scheduled hours on days with scheduled appointments, and time on unscheduled days.
Remember, some of this is time working on notes, follow-ups, and inbox wasn’t possible in the days of paper charts. Easy access to patient records for clinicians is mostly a good thing, but it has brought with it several untoward consequences, with longer hours of EHR use associated with physician burnout.
The results? Here’s the figure, reproduced with the kind permission of the lead author:
Gosh, does this ring true. Hey, I’m logged into Epic right now as I write this, and it’s 5:42 a.m. on a Wednesday, reviewing patient and clinician messages, test results, and — most importantly — prepping for the clinic session I have this afternoon and peeking ahead to tomorrow morning’s appointments, reading through charts to be ready for the visits.
Now none of this is unique to ID docs. I’m married to a primary care pediatrician, and her inbox activity easily exceeds mine. But here are several reasons why ID doctors finished #1 in this review, at least based on my highly anecdotal and admittedly biased perspective.
- Chart review. Before, during, and after a visit. It’s so critical. It would be impossible to do this work without meticulous attention to the history and results. You know that Media tab in Epic, the one you’d like to ignore? That place where “information goes to die”? We ID docs dive right in, painful as opening those scanned documents and inscrutable PDFs might be. If we see someone is scheduled to see us and we don’t have the records to review ahead of time, this elicits deep anxiety and an all-out effort to remedy the situation ASAP. Code Chart.
- Notes. The other day, we hosted some medical students for dinner at our house, and a (quite brilliant) future surgeon recounted something she learned on a recent Transplant Surgery rotation: “Just read the ID notes,” she said. “My resident said you can get rid of the rest of the chart documentation and find a complete and accurate summary of even the most complicated cases.” Indeed. These works of art take time.
- Complexity. People don’t refer to or consult ID for routine issues in Infectious Diseases — they manage them on their own. That community-acquired pneumonia responding to empiric antibiotics? That outpatient with a UTI getting better on nitrofurantoin? That drained abscess, now healing on cephalexin and local care? It’s the opposite of those cases that make up the daily ID doctor’s work — the diagnostic dilemma, the failure to respond, the highly resistant or confusing microbiology. Tough stuff all of them, and I’ve been at this a while.
- OCD. To varying degrees, all us ID doctors suffer from obsessive-compulsive disorder. I’ll confess — the struggle is fierce. If you have never written a note that starts out, “Briefly, …” and is then followed by a scree of prose longer than any other note in the chart, the Infectious Diseases Society of America deserves the right to wonder about your ID credentials.
- Breadth. If there’s a medical or surgical service out there that hasn’t had a patient with an ID complication or issue, I haven’t heard of it. From the broadest primary care clinicians to the super-specialized surgeons who only manage one component of a given body part, we’ve seen patients from them all. This creates quite the pressure to review records and do some pre-visit research about the latest obscure medical treatments or surgical techniques.
All of this requires a lot of EHR chart use, and time spent outside of clinical hours finishing up the work. Look at the distribution of activities in the figure — it shows it’s not just one thing we’re doing more than others. It’s the entire bundle, the results of an extremely diversified portfolio of clicks, keystrokes, and scrolling.
Some might argue that ID doctors should just write shorter notes, and I agree. Notes really should focus on our interpretation of what has happened, why we think it’s going on, and what we recommend — not just a re-statement of material that’s already available elsewhere, if others took the time to look at it.
But importantly, writing shorter notes is easier said than done. Many of my colleagues tell me that if they don’t write out the details of the history, or re-type all the results, they don’t really learn the full story, analogous to taking notes during an important lecture. Others cite the positive feedback they receive from others (see #2 in the above list), saying they don’t want to disappoint their non-ID consulters.
But here’s another motivator to stop providing this chronicling service. We non-procedural specialists consistently find ourselves at the low end of the payment scale for MDs, a situation that will never change with Relative Value Units (RVUs) providing the metric for determining salaries. Sadly, the recent trend wasn’t encouraging — our latest reported salaries were lower than the previous year.
And, as noted in this compelling post, I doubt anyone is getting paid for time spent on the EHR outside of work hours.
Or getting paid by the word.
14 Responses to “How Electronic Health Records Tyrannize Doctors — ID Doctors in Particular”
Leave a Reply
Paul E. Sax, MD
Contributing Editor
NEJM Journal Watch
Infectious Diseases
Biography | Disclosures | Summaries
Learn more about HIV and ID Observations.
NEJM Journal Watch — Recent Infectious Disease Articles- Sulopenem Etzadroxil plus Probenecid (Orlynvah) for Uncomplicated Urinary Tract Infections
- Radiolabeled p-Aminobenzoic Acid as a Marker to Detect Bacterial Endocarditis
- Clostridioides difficile Vaccines: Promising Contenders
- Observations from ID and Beyond: Marking a Social Media Mass Migration — Until the Next One
- Incidence of Autoimmune Diseases Involving the Skin, After Acute COVID-19
I literally need to open four different EMR systems to see one new consult. Cerner inpatient, Epic outpatient and care everywhere, CCHIe for OSH that doesn’t use epic, and PowerShare for OSH image.
Unfortunately it is also common practice for some specialties to consult ID prior to discharge to have one complete document to use or copy for the discharge summary.
Amen, amen!
We are story tellers–we weave a narrative of events and we adduce cause and consequence–we are witnesses. WE are the ones who described the sad evolution from heroic hope to hopeless impairment–it is our note that often shows the ultimate folly of our optimistic colleagues who get to sign off. Later today I will see (again) a patient who fractured his spine in May after a syncopal collapse at age 78–ortho spine signed off. Operation success. Trach, PEG, deep decubitus ulcers but he is awake, aware, anguished and suffering–he evokes Edvard Munch’s The Scream. Surviving sepsis, I ask, then what?
EPIC has a function where you can see how much of a note was pasted (regurgitated) and how much is new. Our notes might win a prize there too. My notes are prose. An oncologist (who writes superb notes) likened my notes to reading the New Yorker magazine–truly a meaningful compliment I believe, not a comment of length).
I am no longer complaining about being underpaid. Soon, payors may realize how little relationship there is between cost and value;.
Lastly, what about time spent trying to learn? About our colleagues’ specialties? About everything? When I am not on EPIC, I am often reading–it all comes back to learning for the job. But our job is still enviable.
blah, blah, blah!
One stand out observation- you, Paul Sax, do at least 2 clinics a week. I do 4 with my various roles and the 4 bleed into every space of my 10 hour days. Yes, the scree of prose I write, but also the dozens of labs I review for my patients, now exclusively those w HIV, NTM, complex fungal (cocci, mostly, one difficult lomentaspora formerly scedo), rare recurrent UTIs and hospital follow ups. I respond to a minimum of 8 patient messages a day, all requiring decisions, none of which I have the time to pause and figure out how to drop a charge. Oh and that cybersecurity training that is overdue? Just got a message that I may take a salary penalty for not completing. It is Sept 21, and I have 9 days to finish the 30 LKA questions. It will take me 45 min max, but do I have this time given all the above?) Oh, yes, and ps…. 11 yo texts me the dog barfed three times. Thank you always for speaking the truth and validating us all.
My notes are shrinking as I get older . Most don’t care about the history including Medicare
I’m a big believer in the inverse relationship between note length and probability of it being read. I know my ID notes will never be the source of the encyclopedic recitation of all medical events (A standard pmh for my notes- “past history reviewed; most notable for 2 prosthetic joints and no immune suppression”). Certainly if I illicit new information that ISN’T in the past history I include it, but otherwise I hate cataloging information that has it’s own searchable tab. The biggest issue is when people say past history reviewed (or complete review of systems), it actually needs to have happened.
Love that PMHX!
-Paul
The problem isn’t EHRs. The ID physicians I work with at an HIV center, as well as the primary care providers (like myself) at the center, are very conscientious. This is a good thing. The problem is summed up in Dr. Sax’s final paragraph. The corrupt RVU system does not value ID and primary care services.
OCD is considered a badge of honor but truth be told it is an illness.. For example atypical chest pain is not better served by a virtue signaling extravagant note and an expensive stress test, echo, labs, PET scan , calcium scoring and a CT angio under the guise of thoroughness, its better served by a short course of ibuprofen and observation for a week. So maybe get of the high horse of I am “so through” and treat the problem and not yourself.
There is something demonstrative/performative about those notes! And I sometimes picture a bow-tie-wearing ID doc preening. But those notes ARE a resource to others when needed in a copy-paste world. It takes all types!
This was one of the reasons that I left ID practice (due to what turned out to be a fortunate turn of events) during the latter part of my career. Also one of the reasons that I retired at age 66 instead of continuing clinical work. So glad I’m no longer a slave to Epic. Retirement has opened up a new and wonderful chapter in my life.
Is it just me, or is does physician pay inversely correlate with time spent on charting?
It’s a problem of our society which is an overabundance and wrongful application of IT systems to everything in our lives. Electronics are not always better than what we did in the analog world. There are MANY examples in all industries where IT/computer systems ARE actually worse. In medicine we keep adding wrong applications to our everyday clinic lives – and make it worse. Paper CHARTS have a lot of advantages over EHR. E-prescription has A LOT MORE disadvantages over paper.
Sadly, it keeps getting worse in medicine, every year there are more IT-requirements added that just make our lives (and that of patients!) more difficult.
The importance of data analysis is essential to any physician’s practice. In Epic, however, I have encountered significant difficulties in accessing and analyzing data. Having utilized this system across three different institutions since the early 2000s, it appears that its design has not evolved significantly in over 20 years. For instance, lab reports from various institutions are often misaligned, complicating even straightforward tasks such as comparing CBC and CMP results. Additionally, the need to open PDF files to locate essential data can be quite time-consuming, and aligning a patient’s treatment history with these results requires manual intervention and navigation across multiple tabs to retrieve relevant medication histories.
In my role as the Director of OPAT at my institution, I have found that the most effective way to track lab results involves creating an Excel flowsheet from faxed lab reports, which are then uploaded as PDF files in Epic. This reliance on a separate system for recording data seems inefficient.
A few years ago, I developed a best practice alert to remind providers of the need for PJP prophylaxis for patients receiving more than 4 weeks of prednisone at 20 mg per day. Unfortunately, Epic is unable to perform this straightforward calculation, despite efforts to address the issue with the company.
I believe that modernizing and enhancing Epic’s capabilities for data collection and analysis could significantly improve efficiency for not only infectious disease physicians but for all users. Such improvements could potentially save hours of time navigating the EMR.