No one chooses emergency medicine for continuity of care. It's undoubtedly rewarding for most doctors to help manage patients through growth or disease, but the emergency physician doesn't get the chance to appreciate a patient's development over time.
Until now.
Electronic health records at my institution now give us over 9 years of continuity. Even when I'm just meeting a patient for the first time, I can look back and see their first visit to the pediatric ED for asthma, the appendectomy from a few years ago, the gastritis visit during college break that may or may not have been related to alcohol.
For older patients, I can see the visits that described the out-of-control hypertension, the subsequent CABG and later, stents... the descent into heart and renal failure.
I get a little nostalgic to see almost-forgotten names of residents and attendings who cared for these patients, before they graduated or took other positions. It's like we were all a part of this patient's life.
Recently I re-open a patient's old chart, to show a resident an unusual EKG from a while back. I felt a pang, when my EHR alerted me the patient had died.
It's tempting to compare electronic health records to other means of keeping up with people, electronically. Research suggests social networks are changing the nature of relationships - rather than a few close friends with whom we share important news and feelings, we're instead updating a wider, but shallower, group of acquaintances with more mundane aspects of life.
While that may be true, EHR's ability to show the timeline of a patient's past care is having a profound effect on emergency medicine - where before there were discrete events, EHR is giving the semblance of an ongoing relationship. Physicians may debate whether electronic records are worth the investment, but to my mind there's no question the investment is more than financial.
Free to decide
Describe electronic clinical decision support to someone outside of healthcare.
They'll probably start conjuring images of Watson or Clippy - an automated guide who incorporates clinical data and provider habits, to offer suggestions to improve care.
And yet, despite advances in computing power and machine learning, and a massive push to adopt electronic records, clinical decision support remains stuck - so primitive that even 90's-era Clippy seems wise and helpful in comparison.
I think there are many culprits for this, but a big reason clinical decision support lags is that it's been hijacked by well-meaning hospital administrators. Instead of nudging doctors into learning new things, dosing meds appropriately for complex patients, ordering the proper tests, so far it's been mostly inane reminders written by bureaucrats. Instead of helping doctors make decisions, decision support has been used to boost compliance on various metrics of interest.
You see, when compliance with a federal measure at your institution is at 97%, steps need to be taken to get that up to 100%. And those steps often involve popups, and "hard stops" that disrupt your progress through a task, even if it doesn't quite apply to this particular patient or that particular situation.
Of course these metrics are important, and if you don't score well, CMS reimbursement could be withheld, and your institution will look bad on HospitalCompare.
But by hijacking an EHR's clinical decision support system to boost administrative goals, you've conditioned a generation of providers to ignore CDS popups and warnings. The alerts are often not really relevant to providing the best care. They're often not appropriate to the situation. And they get really annoying to busy physicians, when they delay appropriate care and add to their workload.
What's happened is that modern decision support has taken the most inquisitive, hardworking, and self-sacrificing group of people you're likely to meet, and turned them off to the idea that the EHR can be a teaching tool.
And I don't think there's a popup alert to reverse that clinical situation.
They'll probably start conjuring images of Watson or Clippy - an automated guide who incorporates clinical data and provider habits, to offer suggestions to improve care.
And yet, despite advances in computing power and machine learning, and a massive push to adopt electronic records, clinical decision support remains stuck - so primitive that even 90's-era Clippy seems wise and helpful in comparison.
I think there are many culprits for this, but a big reason clinical decision support lags is that it's been hijacked by well-meaning hospital administrators. Instead of nudging doctors into learning new things, dosing meds appropriately for complex patients, ordering the proper tests, so far it's been mostly inane reminders written by bureaucrats. Instead of helping doctors make decisions, decision support has been used to boost compliance on various metrics of interest.
You see, when compliance with a federal measure at your institution is at 97%, steps need to be taken to get that up to 100%. And those steps often involve popups, and "hard stops" that disrupt your progress through a task, even if it doesn't quite apply to this particular patient or that particular situation.
Of course these metrics are important, and if you don't score well, CMS reimbursement could be withheld, and your institution will look bad on HospitalCompare.
But by hijacking an EHR's clinical decision support system to boost administrative goals, you've conditioned a generation of providers to ignore CDS popups and warnings. The alerts are often not really relevant to providing the best care. They're often not appropriate to the situation. And they get really annoying to busy physicians, when they delay appropriate care and add to their workload.
What's happened is that modern decision support has taken the most inquisitive, hardworking, and self-sacrificing group of people you're likely to meet, and turned them off to the idea that the EHR can be a teaching tool.
And I don't think there's a popup alert to reverse that clinical situation.
Something's always wrong
I had an idea for an absurd indie comic strip about Corrective Action Plans - those in-depth multi-departmental administrative remedies that are developed in response to bad outcomes, dropped balls, missed communications. Over time we've taken to calling them CAPs.
The comic would follow the adventures of CAP'n America. He'd never arrive in time to save the day (in fact, he only worked business hours) but he would be sure to show up after a traffic accident or crime scene investigation and make sweeping pronouncements about how workflows will have to change.
Wouldn't that make for funny reading?
I still consider myself relatively new to medicine, but I think I've been around long enough to see the culture morph from closed-door, M&M-style reviews of cases... to CAPs. Maybe this is progress - certainly, it's change - but it's a far cry from the data-driven, common-sense laden discussions we don't seem to be having often enough.
CAPs seem like an outgrowth of Six Sigma mindset - getting to a defect rate of under 3 per million. Of course, when your ED sees half a million patients over five years, and a particular bad outcome happens once during that time, administrators don't like to be reminded that the failure rate is actually well in line with six sigma. They want a CAP, so that it never happens again.
Never is a powerful word in medicine. And "never" always comes with a cost, though sometimes it's not easy to measure that cost.
So what if the corrective action plan - to prevent that one really rare foulup - makes routine care a little bit worse? Length-of-stays and time-to-dispositions become a bit longer as doctors and nurses implement a new workflow. The delay is ok because it's all for a good cause, right?
But whenever we agree to a CAP - agree to revise our rules in response to a rare, awful event, rather than to address a routine awful events like extended boarding times or delays in prompt care, the process is stacked against us from the beginning. It's stacked against practitioners and against the vast majority of patients, too. And it becomes impossible to truly improve care.
How can we correct this process?
The comic would follow the adventures of CAP'n America. He'd never arrive in time to save the day (in fact, he only worked business hours) but he would be sure to show up after a traffic accident or crime scene investigation and make sweeping pronouncements about how workflows will have to change.
Wouldn't that make for funny reading?
I still consider myself relatively new to medicine, but I think I've been around long enough to see the culture morph from closed-door, M&M-style reviews of cases... to CAPs. Maybe this is progress - certainly, it's change - but it's a far cry from the data-driven, common-sense laden discussions we don't seem to be having often enough.
CAPs seem like an outgrowth of Six Sigma mindset - getting to a defect rate of under 3 per million. Of course, when your ED sees half a million patients over five years, and a particular bad outcome happens once during that time, administrators don't like to be reminded that the failure rate is actually well in line with six sigma. They want a CAP, so that it never happens again.
Never is a powerful word in medicine. And "never" always comes with a cost, though sometimes it's not easy to measure that cost.
So what if the corrective action plan - to prevent that one really rare foulup - makes routine care a little bit worse? Length-of-stays and time-to-dispositions become a bit longer as doctors and nurses implement a new workflow. The delay is ok because it's all for a good cause, right?
But whenever we agree to a CAP - agree to revise our rules in response to a rare, awful event, rather than to address a routine awful events like extended boarding times or delays in prompt care, the process is stacked against us from the beginning. It's stacked against practitioners and against the vast majority of patients, too. And it becomes impossible to truly improve care.
How can we correct this process?
In a manner of speaking
I never thought I'd embrace dictation.
For emergency physicians, dictation varies department to department. Some require it, some make it an option, and for others it's an exotic luxury. For my part, I'd seen too many errors in transcription and watched too many colleagues struggle with a phrase over and over to get enthusiastic about it. I never considered investing in dictation software for my own workflow, since I'm not a bad typist and it wasn't something my department was offering to us.
Along came iOS 5 for the new iPad, and suddenly I found myself dictating emails and short messages. Dictation seems like a natural fit for the iPad, since I can't type particularly quickly on it, and I'm not writing or editing long pieces on it.
Plus, I enjoy Apple's style of dictation, where my phrase or sentence remains invisible as it's transported to their servers for transcription. This process has been derided as a step backward, by folks who are used to desktop dictation software showing your transcription word-by-word, on screen. But mentally checking each word can be its own distraction, and disrupts flow. Plus, Apple's way of doing dictation seems more like a fun gamble to see if they got the whole sentence correct.
So, when OS X Mountain Lion came out this week, with its Siri-like dictation feature, I finally gave computerized dictation a chance to help with my charting.
First, a couple of concessions:
For emergency physicians, dictation varies department to department. Some require it, some make it an option, and for others it's an exotic luxury. For my part, I'd seen too many errors in transcription and watched too many colleagues struggle with a phrase over and over to get enthusiastic about it. I never considered investing in dictation software for my own workflow, since I'm not a bad typist and it wasn't something my department was offering to us.
Along came iOS 5 for the new iPad, and suddenly I found myself dictating emails and short messages. Dictation seems like a natural fit for the iPad, since I can't type particularly quickly on it, and I'm not writing or editing long pieces on it.
Plus, I enjoy Apple's style of dictation, where my phrase or sentence remains invisible as it's transported to their servers for transcription. This process has been derided as a step backward, by folks who are used to desktop dictation software showing your transcription word-by-word, on screen. But mentally checking each word can be its own distraction, and disrupts flow. Plus, Apple's way of doing dictation seems more like a fun gamble to see if they got the whole sentence correct.
So, when OS X Mountain Lion came out this week, with its Siri-like dictation feature, I finally gave computerized dictation a chance to help with my charting.
First, a couple of concessions:
Harmonize the wind
I always liked the idea of the original Ambient Orb -- a device that just sits there, but changes color based on metrics of interest to you. Bad weather approaching? It turns red. Stock market up? Green.
CIO John Halamka famously placed one on Paul Levy's desk, to help the CEO effortlessly monitor the ED waiting room situation.
But colors can only express so much - I think the Ambient Orb could just communicate a few things like "Good" or "Bad" or "Really Bad" on whatever you programmed it to care about.
What really held me back, though, was the idea of spending $150 on a ball that passively monitors some situation, when more "active" monitoring was never more than a few clicks away.
Then came Twitter.
I've given Twitter and its users a lot of grief over the years, even as I've come to spend more time with them than any other social network. But Twitter seems built around the concept of passive monitoring.
Skimming a Twitter feed is a nice way to check in with friends and colleagues, and pick up some news or useful links. I'm getting comfortable with the idea that Tweets are a workable proxy for thoughts, and also starting to accept that software can accurately categorize Tweet content and deduce sentiment.
So maybe a Twitter feed isn't the best way to survey the hive mind.
The Listening Machine (hat tip: the Verge) is a project to follow 500 UK Twitter accounts and figures out the positivity or negativity (or neutrality) of Tweet content, as well as categorize the Tweets into one of eight subjects. The Tweets are then converted to music.
It seems to me that music might be better than color, to reflect the complexity of the Twitter stream. I've been listening on and off for the past few hours, and can pick up without much difficulty when the overall sentiment turns negative, and when the rate of tweets pick up. I wonder if it's possible to tell if the stream is featuring ponderous topics or light chitchat - or if the current discussion is weighted toward politics, or the arts.
The idea of catching a snippet of music and knowing the mood, engagement, and to some extent, the content of conversations in an area, is very appealing, though I think DJs already make something like this possible, when reading a crowd, picking up a vibe. Twitter analytics will just make the crowd's thoughts and feelings more quantifiable.
CIO John Halamka famously placed one on Paul Levy's desk, to help the CEO effortlessly monitor the ED waiting room situation.
But colors can only express so much - I think the Ambient Orb could just communicate a few things like "Good" or "Bad" or "Really Bad" on whatever you programmed it to care about.
What really held me back, though, was the idea of spending $150 on a ball that passively monitors some situation, when more "active" monitoring was never more than a few clicks away.
Then came Twitter.
I've given Twitter and its users a lot of grief over the years, even as I've come to spend more time with them than any other social network. But Twitter seems built around the concept of passive monitoring.
Skimming a Twitter feed is a nice way to check in with friends and colleagues, and pick up some news or useful links. I'm getting comfortable with the idea that Tweets are a workable proxy for thoughts, and also starting to accept that software can accurately categorize Tweet content and deduce sentiment.
So maybe a Twitter feed isn't the best way to survey the hive mind.
The Listening Machine (hat tip: the Verge) is a project to follow 500 UK Twitter accounts and figures out the positivity or negativity (or neutrality) of Tweet content, as well as categorize the Tweets into one of eight subjects. The Tweets are then converted to music.
It seems to me that music might be better than color, to reflect the complexity of the Twitter stream. I've been listening on and off for the past few hours, and can pick up without much difficulty when the overall sentiment turns negative, and when the rate of tweets pick up. I wonder if it's possible to tell if the stream is featuring ponderous topics or light chitchat - or if the current discussion is weighted toward politics, or the arts.
The idea of catching a snippet of music and knowing the mood, engagement, and to some extent, the content of conversations in an area, is very appealing, though I think DJs already make something like this possible, when reading a crowd, picking up a vibe. Twitter analytics will just make the crowd's thoughts and feelings more quantifiable.
SAEM Didactic on Social Media and the Academic Physician
Members of the social media committee spoke today at SAEM about, well, online social networking and the emergency physician. Our presentations are available via Prezi: mine as well as Jason Nomura's and Rob Cooney's. We also had a conference call a few weeks ago where we covered some of these topics, moderated by Jim Miner and recorded by Scott Joing.
Below, I've added links to stories, papers and sources I referenced, in my presentation:
Matthew Strausburg's letter about Facebook's risk to his career
The malpractice trial outing of the anonymous pediatrician, Flea
Rhode Island EM physician Alexandra Thran's Facebook case and RISMB filing (PDF)
The nursing student, Doyle Byrnes, who posted a placenta photo on Facebook.
The Hayley Barbour clinic Tweet controversy
AMEDNews.com report on QuantiaMD survey of physicians use of social networks
Tips on HIPAA compliance while on social networks (and some notes on HIPAA compliance in social media policymaking)
Greysen et al, JAMA 2012 research letter on physician violations of online professionalism and state medical board disciplinary cases
Chretien et al, JAMA 2011 letter to the editor on physician usage of Twitter in 2010, broken down by specialty
Ed Bennett's list of hospital policies on social media
SAEM's social media guidelines
For further reading, Dr. Bryan Vartadedian's blog, 33 Charts has a lot of terrific writing on online professionalism: 1,2,3,4
Below, I've added links to stories, papers and sources I referenced, in my presentation:
Matthew Strausburg's letter about Facebook's risk to his career
The malpractice trial outing of the anonymous pediatrician, Flea
Rhode Island EM physician Alexandra Thran's Facebook case and RISMB filing (PDF)
The nursing student, Doyle Byrnes, who posted a placenta photo on Facebook.
The Hayley Barbour clinic Tweet controversy
AMEDNews.com report on QuantiaMD survey of physicians use of social networks
Tips on HIPAA compliance while on social networks (and some notes on HIPAA compliance in social media policymaking)
Greysen et al, JAMA 2012 research letter on physician violations of online professionalism and state medical board disciplinary cases
Chretien et al, JAMA 2011 letter to the editor on physician usage of Twitter in 2010, broken down by specialty
Ed Bennett's list of hospital policies on social media
SAEM's social media guidelines
For further reading, Dr. Bryan Vartadedian's blog, 33 Charts has a lot of terrific writing on online professionalism: 1,2,3,4
Eponymous
I love being at an academic institution for many reasons. One of which is that the residents keep me on my toes -- they've read the latest stuff, they want justification for proposed workups and treatment plans, and...
....they've got the sharpest grammar?
Yep. I can only imagine the patients watching us in the ED -- when we're not hunched over our keyboards, charting, reviewing and ordering, we can be seen gesticulating wildly about things like the placement of apostrophes in eponymous diseases.
For instance, can you pick what's proper, below?
Down Syndrome or Down's syndrome
Legionnaire disease or Legionnaires' diseaseThe rule I've always heard is: if the disease is named for a patient, it deserves an apostrophe. If the disease is named for its discoverer, the apostrophe is inappropriate. Lou Gehrig's disease was his, and the Legionnaires had theirs, but Down didn't own his syndrome. More here.
The most consistently misattributed apostrophe, I think, belongs to Crohn -- though that's slowly changing, too. We'll leave for another time the discussion of proper possessive apostrophe usage when a name ends in s (Legionnaires' or Legionnaires's), perhaps until agreement can be reached the Chicago Manual of Style and AP Style.
I just want to highlight the body of work of Tsung O. Cheng, who has been writing about eponymous diseases and the inappropriateness of apostrophes for fifteen years -- including how to handle the situation when the discoverer of the condition is also a patient.
While Dr. Cheng is Professor of Medicine and has been a prolific author, churning out 10 medline-indexed publications on this trivial topic means either medical grammar is an exciting and contentious field for potential scholarship -- or that our system of academic promotion and peer review is kind of messed up.
So, I humbly suggest: Deciding on whether to devote one's energies to adding lines on one's CV, or actually trying to contribute to the body of knowledge in medicine, shall henceforth be known as Genes' dilemma.
An inflammatory condition
I've noticed more patients complain of gout recently -- maybe it's a sign the economy is improving (gout has long been considered a disease of excess).
As I discovered while revising Rosen's new chapter on arthritis emergencies, the price of colchicine, an ancient gout treatment, has gone way up lately -- while suggested regimens for treating gout have recently changed. These are not unrelated, and the connection is actually quite interesting. You see, colchicine was grandfathered-in as an approved therapy by the FDA at its inception in 1938. Its safety and efficacy were never formally reviewed by the FDA...
...until 2009, when URL Pharma submitted the results of their new trial to the FDA, showing comparable efficacy and somewhat improved safety to high-dose colchicine regimens, using a simple 1.2 mg dose followed by a 0.6 mg chaser an hour later.
The FDA, bound by Hatch-Waxman exclusivity, considered this a new indication for colchicine, and responded by granting 3 years of market exclusivity to URL Pharma. The price of colchicine (now called Colcrys) subsequently shot up from 9 cents a pill to $4.85 -- though if you read the drug company's site, they make it sound like they're doing patients a favor, protecting them from unapproved forms of the drug that worked well enough, for decades.
I appreciate the way the authors of this NEJM opinion piece considered this state of affairs:
I'll just be interested to see what happens in August 2012, when the 3-year exclusivity period should end and colchicine prices return to, hopefully, to a less painful place.
As I discovered while revising Rosen's new chapter on arthritis emergencies, the price of colchicine, an ancient gout treatment, has gone way up lately -- while suggested regimens for treating gout have recently changed. These are not unrelated, and the connection is actually quite interesting. You see, colchicine was grandfathered-in as an approved therapy by the FDA at its inception in 1938. Its safety and efficacy were never formally reviewed by the FDA...
...until 2009, when URL Pharma submitted the results of their new trial to the FDA, showing comparable efficacy and somewhat improved safety to high-dose colchicine regimens, using a simple 1.2 mg dose followed by a 0.6 mg chaser an hour later.
The FDA, bound by Hatch-Waxman exclusivity, considered this a new indication for colchicine, and responded by granting 3 years of market exclusivity to URL Pharma. The price of colchicine (now called Colcrys) subsequently shot up from 9 cents a pill to $4.85 -- though if you read the drug company's site, they make it sound like they're doing patients a favor, protecting them from unapproved forms of the drug that worked well enough, for decades.
I appreciate the way the authors of this NEJM opinion piece considered this state of affairs:
It seems reasonable to expect that costly new drugs or increases in drug prices would be accompanied by a substantial benefit in disease management to be enjoyed by these patients. This standard is not met by Colcrys; in this instance, the public may bear considerable costs for a poorly executed administrative goal.Yep. A few US Senators sent URL Pharma a strongly-worded letter this past spring, but I don't think anything came of it (you can't say the company's done anything illegal, and our society is well past the point where we can expect corporations to act in any interest but their own).
I'll just be interested to see what happens in August 2012, when the 3-year exclusivity period should end and colchicine prices return to, hopefully, to a less painful place.
A few more iOS 5 considerations, for EM
I've been using my iPad in the ED, with my white coat's sewn-in iPad-sized pocket, for some time now -- mostly for patient and resident education, and to look up dosages or rashes. Hitting up my Evernote database or Dropbox documents is also useful. Occasionally I'll use my iPhone, for its LED light (when the otoscope can't reach to where I need to see) or rarely, its camera (in compliance with my hospital and department photo policy, naturally).
Our ED's EHR isn't quite accessible enough via iPad for me to quickly check results or place orders at the bedside -- right now it's just too cumbersome. But there's been progress -- enough so that I start to wonder about the flip side: instead of reviewing iOS medical apps and pining for an optimized EHR experience on the iPad, what if there are features of the iPad that could limit the utility of medical apps?
Well, there are some product design issues, like impact resistance and bacterial colonization, that have been discussed. But the operating system, iOS 5, has some quirks, too. Some have received a lot of attention. Some are maddening in their capriciousness.
But if you're an administrator, or an app developer, working on healthcare apps for folks with iPads, there's got to at least a few special considerations on your plate that you've never had to worry about, before (beyond the whole portrait/landscape issue). I've listed a few that seem novel to me, below:
Our ED's EHR isn't quite accessible enough via iPad for me to quickly check results or place orders at the bedside -- right now it's just too cumbersome. But there's been progress -- enough so that I start to wonder about the flip side: instead of reviewing iOS medical apps and pining for an optimized EHR experience on the iPad, what if there are features of the iPad that could limit the utility of medical apps?
Well, there are some product design issues, like impact resistance and bacterial colonization, that have been discussed. But the operating system, iOS 5, has some quirks, too. Some have received a lot of attention. Some are maddening in their capriciousness.
But if you're an administrator, or an app developer, working on healthcare apps for folks with iPads, there's got to at least a few special considerations on your plate that you've never had to worry about, before (beyond the whole portrait/landscape issue). I've listed a few that seem novel to me, below:
Subscribe to:
Posts (Atom)