A physician pulls up a patient’s medication list mid-visit and stops. Something’s off. The patient stopped that drug months ago, and the list never caught up. Nothing in the system flagged it; the physician just happened to notice. Ask who’s responsible for making sure that gets caught every time, and nobody has an answer.
Someone built that list years ago, correctly, for a patient who’s since moved on to a different regimen twice over. Nobody’s updated it since, because it never became anyone’s specific job to check.
Fifteen years of the same note, getting longer every year
A decade-long study at a major academic medical center put real numbers on it. Nearly 3 million outpatient notes, 46 specialties, a straight line running from 2009 to 2018: median note length up 60 percent, from 401 words to 642.
Real physician voices, twice a week
Free, and one click to unsubscribe.
Length alone would be a minor complaint if the extra words were new information. Median redundancy, how much of a note matched the patient’s last one, climbed from 48 percent to 59 percent across those same years. By 2018, barely a third of the average note came from someone typing in the moment. The rest arrived pre-assembled, pulled from a template or carried forward from whatever came before.
Notes got longer for a specific reason: Less of what’s in them came from anyone actually in the room.
More templates isn’t the fix, and neither is fewer
A second study followed 203,728 physicians on Epic and measured what that redundancy costs. Physicians in the top decile for note length spent 39 percent more time in the electronic health record (EHR) after hours and closed 5.6 percentage points fewer visits same day. Heavy copy-paste users closed 6.8 points fewer.
Templates surprised the researchers. Heavy use of them predicted worse outcomes. So did almost never using them. The physicians doing best sat in the middle, somewhere between relying on a template and living inside one.
The authors stayed careful about what that middle zone actually means, saying only that tools like copy-paste and templated text “may have limited efficacy,” a narrower claim than the headline finding suggests. Still, the worst outcomes sitting at both extremes raises an obvious question about the practices in the middle. Did they land there by luck, or because somebody was actually tending those templates instead of just clicking through them?
What a stale template actually risks
The everyday cost starts small. A smart phrase still pulls language from a guideline that’s two updates out of date. Somewhere in the problem list, entries that stopped mattering years ago are still sitting there, and so is a login for someone who left the building last spring. None of it costs more than a few extra clicks, which is exactly why none of it feels worth fixing today.
The gap widens on its own. Ninety seconds of manual correction last year turns into closer to three minutes this year, the distance between what the template shows and what’s actually true stretching a little further every month nobody looks at it.
Medication lists are where small stops covering the whole story. A list still naming a drug the patient stopped in the spring isn’t clutter anymore; it’s something a physician might glance at and trust while deciding what to prescribe next, whether it interacts, whether a dose needs adjusting. The list looks authoritative on the screen. Whether it’s actually current is a separate question, and in most practices nothing checks. Every physician who’s caught one of these has felt the same two things at once: relief that it didn’t go further, and a flicker of doubt about how many times it hasn’t been caught at all.
A missed referral eventually generates a phone call. A stale medication list generates silence, right up until the day it doesn’t. None of this is fair to expect the front desk to catch in a spare minute, or a physician to catch mid-visit, every time, forever. The gap isn’t anyone’s failure. It’s just never had anyone’s name on it.
Somebody has to own the system, not just use it
The data already answered the obvious fixes, and none of them held up. Typing less doesn’t help, since physicians who barely use templates struggle as much as physicians who lean on them heavily. Trimming down to fewer templates hits the same wall; the worst outcomes sit at both extremes, not one. Switching EHR vendors misses the point entirely, since the software was never really the issue; what happens to it once it’s installed was. A different vendor’s templates go stale on exactly the same schedule, the moment nobody’s assigned to keep them current.
Physicians in practices they own themselves report meaningfully higher EHR satisfaction, a stronger sense that their documentation time is reasonable, and more people around them helping with the work, than physicians in practices someone else owns. The study stops short of saying documentation support alone explains that gap. It does show the two showing up together often enough to take seriously.
Every complex system a practice depends on has a name attached to it somewhere. Someone runs the schedule and gets asked when it breaks. Someone owns the billing platform the same way. The templates and smart phrases and problem lists underneath every note a physician writes get built once, at go-live, and after that belong to whoever happens to notice something’s wrong. In practice, that means they belong to no one.
Real ownership here isn’t complicated: Someone reviews templates on an actual schedule, prunes the lists before they go stale, and closes access the day someone leaves instead of six months after. That’s not something anyone does well in the gaps between other work. It has to actually be someone’s job, not the fourth item on a list for someone already too busy to get to the first three. It means the physician who caught that medication list once this time doesn’t have to be the one catching it every time from now on.
Karan Kanwar is the CEO of Central AI and Wing Assistant. Central AI provides an AI operating system for appointment-driven medical practices, handling calls, texts, web chat, and messages around the clock, qualifying patients and booking appointments directly into a practice’s existing systems. It is HIPAA compliant and designed to complement the tools practices already use.
Wing Assistant pairs that technology with skilled virtual assistants who take on the administrative work that pulls teams away from patient care. Kanwar writes on where automation and human support meet in health care operations, and how practice leaders can adopt both effectively. He shares updates on LinkedIn.
