25 December 2012

Merry Christmas!

From one godless atheist to your family. Hoping it's a great holiday for all of you.

16 December 2012

Tis the Season

Grew up in a very big Irish Family in Chicago, and this pretty well encapsulates my experience of the holidays growing up:

My cousin Tom and I were the nephews designated Mischief and Mayhem...

12 December 2012

(Only) human

Chatting with some med students, a good question was raised: how do we, as doctors, deal with the emotional baggage we encounter in our profession? It's high stress, we see disturbing things, and sometimes we make mistakes that can result in harm to patients. The pressure and responsibility can be very hard to handle.

These stresses, if unmanaged or poorly managed, can carry severe consequences for physicians. Burnout is rampant among docs (and trainees, too). Doctors have high rates of divorce, substance abuse and have the highest suicide rate of any profession.

A normal day at my job is hard: I'm running nonstop for 8-12 hours, I'm constantly interrupted, I have patients making demands of my attention and empathy, I'm saturated with information and need to make rapid decision without adequate information, and I know that if I make an error or miss some important piece of information, the human, professional and financial consequences can be disastrous. It's a pressure cooker.

And that's a day where things go well. A bad day can be very bad indeed. Sometimes it's just the emotional strain of dealing with particularly difficult patients. Maybe you go through a run of giving out terrible diagnoses. Maybe you deal with the death of a child. Or a patient who pulls at your heartstrings in some unique and personal way. Maybe someone dies on you unexpectedly. Worse, maybe someone dies on you and you're not sure if it was your fault or not. Perhaps you know you made an error, and that you're going to have to face accountability for it.

These are the days that drive physicians over the edge. I've had them, and I remember them so vividly even years later. There was the one lady with a gallbladder attack on Thanksgiving, many years ago. She had classic signs and I saw gallstones on my bedside ultrasound. She crashed and died right in front of me from a ruptured thoraco-abdominal aortic aneurysm. Her abdominal aorta had looked normal on my scan; the aneurysm was in the chest and ruptured into the thorax, which is very unusual. That didn't make it any easier to go home and sleep that night.

So I guess my take on the question is not how do we deal with the psychological stress but how should we? I am not an expert, but here are my thoughts.

The first step, which most practicing professionals have already accomplished, is to learn what is called "professional detachment." This is an unnatural skill in which you must suppress your innate sympathy for the suffering experienced by a fellow human being, pain which you may be personally inflicting. The first time you stick someone with a needle, it's probably as traumatic for you as for the patient. More advanced applications involve you ignoring someone's pain or personal tragedy while trying to figure out the hidden life threat. This is a necessary skill if you are to function in the medical environment.

Another way to think of the same skill is to maintain a sense of distance. Remember, an older teaching physician once told me, the patient is the one with the disease. This helps you remember that the patient's condition is not your doing (usually) and their outcome, if negative, is the result of their disease and not necessarily a reflection on your care.

While this detachment is useful and necessary, it can be maladaptive if taken to extremes. First of all, as a physician you do need to express empathy and compassion. It's part of the job. But the emotional demands will be overwhelming if not governed in some fashion; we have limited capacity for caring. My solution is to dole out my compassion and empathy in measured doses, as appropriate to the case and my own mental state. This is not a license to be callous and uncaring in other cases, but rather to be polite, professional and reserved, emotionally.

Furthermore, you need to understand that the professional reserve does not equate to repression of emotion. You suppress it, in the moment, set it aside to get the job done, but that doesn't mean it never happened. For minor stuff it probably is okay to suppress it & forget it.  But the bad things — they won't go away on their own, but will fester and bubble up at the most inopportune moments. You need to take some time, when appropriate, to unpack the experience and re-live the emotions to deal with them. Maybe it will be just turning the case over in your head the next day. Maybe it needs to be more immediate. We've sent docs home after bad pediatric arrests when it was clear they were so upset they needed some time. It's essential, in any case, to explore the disturbing feelings so you can come to a resolution and move on.

Many institutions will have formal critical incident debriefings for the entire team, for particularly awful events. While this doesn't need to be performed formally for routine events, it's a good idea to informally debrief with a trusted partner, superior or mentor. Talk through the case, review the medicine and the science, review your actions and outcomes, and your emotional response to the situation. It is helpful to do this with someone you respect, so he or she can give you valuable feedback. This can be over coffee or a beer or three; possibly better that way.

There can be a lot of shame involved when there was a bad case, even when well-handled, but especially so when you know that you made an error or may have. A lot of docs like to bury these as deep as possible. But these in particular are helpful to talk about, and the more publicly the better. This is not easy, but can be invaluable. We instinctively shy away from openly talking about our mistakes, but when you do you will probably receive a lot of support from your colleagues, many of whom have done the same or understand that "there but for the grace of god go I." An additional benefit is that your mistakes may have been due to a system error or a cognitive bias and by reframing the discussion in an educational light, by seeking out the root causes, you can improve the quality of your own care and that of your partners.

Keep a sense of perspective, and try to stay positive. When the job is really getting you down, take a break, go out to the ambulance bay, take a few deep breaths and try to remember the big picture. We have a great job. It's a privilege and an honor to be allowed to care for patients. We can sometimes make a huge difference in people's lives. We have respect and status in society, and are quite well paid for it. Many people would give their right arm to be where you are. Yes, seeing the 10th drug seeker of your shift is a drag, but damn, it's still better than sitting at a desk and moving numbers from column A to column B.

Sublimation is a defense technique that is particularly valuable in the ER. It is a form of displacement where the negative feelings are transformed into something positive, or at least more-or-less acceptable. The most common form it takes is "gallows humor." Tragedy and comedy are deeply linked, and a morbid witticism can provide a lot of relief of the emotional tension that builds up in a clinical setting. Others may channel these feelings into art or literature. To each their own. If this is not your thing, find an outlet. I practice karate, and there's nothing like pounding the hell out of the heavy bag — or a white belt —after a bad day.

Finally, and possibly most importantly, when you know you screwed up, when you know there was an error that harmed or may have harmed a patient: forgive yourself. You are human, as are we all, and we make mistakes. Take the time to understand it, do your best to learn from it, and forgive yourself. Let go of it, file it away, and move on. If you don't or can't, self-doubt and self-hate will paralyze you and in the end it will sink you.

One last thing: if you are really having trouble, get professional help. If you're self-medicating, or if you are bringing work home to the point it's affecting your family, be humble and realize that doctors can benefit as much as (or more than) any other patient from psychological counseling and support. Many hospitals have a confidential Physician Assistance Program, staffed by professional counselors trained to deal with the issues doctors struggle with. I've seen doctors torpedo their careers with behavior and substance issues, and I've seen programs like these successfully rehabilitate physicians who were in a downward spiral. Check with your medical staff office and use the resources that they offer.

30 November 2012

What malpractice looks like

I review a lot of cases in my professional life. Some of them are just ones that our QA group comes across in our practice. Some are cases related to our liability policy. Some are cases I'm sent for review, or educational cases I present. We see a lot of cases which could have been done better, or in which the documentation is imperfect (or even downright bad). But, fortunately, most of the cases that pass across my desk are within the standard of care.

We get into a lot of arguments over when care provided (or documented) falls below the "standard of care." This term is widely misunderstood, especially in academic circles, and this causes a lot of controversy. Many docs interpret the "standard of care" to mean "best practice." So any care that deviates from best practice, they contend, is prima facie a failure to meet the standard of care (and hence, malpractice).  Unfortunately, this is the interpretation that plaintiff's experts also prefer to embrace! However, it's important to understand that "standard of care" is a legal term with a clear definition that is much more expansive: the level at which an ordinary, prudent professional having the same training would practice under the same or similar circumstances. So the standard of care is not only not perfect care, it is not even average care, because by definition that would imply that 50% of care is below the standard.

This is a pretty low bar, actually. As I explain to our docs and trainees, you are allowed to be wrong. You are allowed to make errors. You are not allowed to be negligent. There is a difference. This is all, of course, limited to the abstract world of theory and pre-trial evaluation. Actual juries have notoriously variable determinations as to the standard of care. But when reviewing cases in advance, deciding which to defend, or what you would testify in favor of, it's a good guideline.

The cases I review tend (obviously) to involve bad outcomes, and generally present with varying degrees of imperfection, but it's pretty rare for me to see a case and stone cold identify it as malpractice. Part of this is because most docs are not, in fact, negligent, and part may be because I have a bias towards the defendant physicians. Most of the deficiencies I see generally involved a diagnostic error, or a minor lapse that probably did not impact the outcome of the case, or simply poor supportive documentation of the thought processes that drove the decision-making the way it went.

Sometimes, though, there is a case that you review and immediately reach for your checkbook.

This is an example of one such case.

A 19-year old male presented to the ER with a fever and headache. He was generally well-appearing, though febrile and tachycardic and as ill-appearing as a young person with the flu typically appears. He had no focal symptoms to suggest a source for the fever (i.e. no cough or sore throat, etc), just generalized fatigue and bodyaches. He was alert with a totally normal neurologic exam. He had no meningismus; his neck was described as supple on two separate exams. He was given 2 liters of IV fluids and tylenol after which his vital signs normalized and he felt much better. He was re-examined twice and demonstrated improvement on both exams, which were well documented and timed. Nursing notes agreed that the patient was much improved. The doc, a conscientious and compulsive sort, did a fairly thorough work-up. Chest x-ray was normal, as was bloodwork, with the exception of a WBC 11,000, just at the upper limit of normal. Influenza swab was negative. Blood cultures were sent, but antibiotics were not given. Because of the severity of the headache, he also did a spinal tap, which was normal. The patient was discharged home in the care of his parents with instructions to follow up with his doctor the next day for a recheck if he wasn't feeling better, and a voicemail was left with the PCP to ensure access to follow-up care. The discharge diagnosis was "Fever, uncertain source; possible viral syndrome."

So... before reading on, do you see any inadequacies in this case? I don't. If anything, the case was more aggressively worked up than was indicated, and for sure more workup was done than I would have, generally.

Except for one thing. The doctor documented a "normal" spinal tap when in fact the lab reported 110 WBCs, mostly neutrophils. This indicates that the patient had meningitis, quite probably bacterial.

More baffling, the doctor knew about this. The lab called the charge RN, and the charge RN notified the doctor, who added on CSF PCR studies for viral pathogens.

And yet he discharged the patient. Didn't call the diagnosis meningitis. Didn't tell him there was a possibility of serious illness. I have no clue why. It's baffling.

Now it's really easy to bash him as incompetent and dangerous, but I know this guy well. He's an MD/PhD who is double boarded in EM and critical care. He's smart as hell, and generally a great and conscientious physician. We don't know what happened here. Of course this case went on to the predictable bad outcome. The doc does not remember the case, so he can't really explain or defend it either. One can only presume that it was busy and he got confused or distracted, maybe had the discharge teed up and ready to go, expecting the negative LP results, and failed to change course on getting the results. It is, in any event, as clear-cut a case of a medical error as I can ever recall seeing. Most of us will never see such a case, unless you're doing expert review.

Now ask yourself, if he had not done the LP, the outcome would have been the same, and the allegation of negligence would still have been there: Fever and headache — how can you justify not doing the LP? If you've been in the trenches, though, you know that everyone with the flu also has a headache. It's part of the febrile syndrome. But the decision whether or not to LP is a judgement call. you can make a wrong judgement without being negligent. I would not have done the LP, based on the case as presented. I'd have been wrong, but in such a case that decision would have been well within the standard of care.

This is also a trend that I see when reviewing series of closed cases where the doctor lost in court or settled. Sure, there are cases where the care was fine but it settled because of a sympathetic plaintiff, or where a jury miscarried justice. But remember that the odds that a physician will prevail in a malpractice case is about five to one. We almost always win. When we lose, more often than not, there was a "WTF?" moment when you review the doctor's actions. It makes it really hard to present these cases for educational purposes: the docs reviewing the case can't put themselves in the position of making such an egregious error. The only possible conclusion is that the doctor who screwed up was an idiot or lazy or a "bad doctor." It's not true, though. There are bad doctors out there, but there are many more good ones. Of the good ones, we are all human and we all are subject to cognitive biases and errors, no matter how smart we are. And ER docs all bear the burden of a distracting environment with systems prone to error (hand-offs, triage cuing, overcrowding), working night shifts, seeing patients who may not be able to tell us what's going on. A set-up for errors.

In the last decade I have cared for about 15,000 patients, and I am sure that I have made an error just like this. I must have been lucky, since mine didn't blow up in my face. Maybe I caught it, or a nurse did, or it was for a less lethal condition. If you're honest with yourself, know that you will make errors like this, too.

So bear this in mind, when you think about "malpractice" and the "standard of care." Negligence, when you see it, is usually not debatable; it's obvious and flagrant. If there's a reasonable case to be made that the care provided was within the standard, it probably was an ordinary error or a mistake of judgement. This is not to say you will win in court! But perhaps you can think of it like pornography, in the words of Justice Potter Stewart, "I know it when I see it."

27 November 2012


I love, it, I really love it, when one of my strongly-held prejudices is borne out by actual, you know, facts and science.

For years, I have been arguing against the practice of performing a routine lumbar puncture (aka LP or spinal tap) in patients with the "worst headache of their life." This is done after a CT scan of the brain, typically, to look for a subarachnoid hemorrhage (SAH). The SAH is feared because in some cases they represent a leaking aneurysm which is at risk of bursting, often with devastating or lethal consequences.

The need to do the LP is one of the sacred cows of Emergency Medicine, written in stone, and has been for longer than I have been practicing. The reason is that SAH is dangerous, the CT scan is imperfectly sensitive for SAH whereas LP is highly sensitive (in fact, the "gold standard") and relatively easy and safe to do. This was perhaps more true long ago when the resolution of a CT scan was lower than it is with modern machines, but the dogma remains. There is, however, a huge variation in actual practice out there. Many docs seem to do very few LPs for headaches, and some seem to LP everybody. I performed a unscientific survey of ER docs on twitter and found that about half "always" still do the LP or are strongly inclined to do it routinely. Some were, in fact, required by their employer to do the LP!

Now my experience over the years was that the LP seemed to be a horrific waste of time. It was traumatic for the patient, consumed a lot of ER resources, and never ever showed anything. Twice -- twice! -- in a decade I spotted the unicorn and had a genuine negative CT followed by a positive LP. In both cases, the patient went on to have negative angiograms, so either the LP was a false positive or they were non-aneurysmal bleeds (which, as it happens, do not require treatment).

So I dug into our data. Pulling a year's worth of cases, I found that we had about 2,800 headaches present annually, slightly under 3% of all of our visits. 18 of those were subsequently diagnosed as SAH, for a prevalence of about 0.6% within all-comers of headaches. But that's not entirely fair, since over half of the headaches were either migraine type headaches or other chronic/recurrent headaches, and these folks are not those for whom we are highly suspicious of SAH. Of the headache patients, about 900 had CT scans ordered. While I might argue that not all of those truly needed a CT, and certainly not all would have gotten one in other countries, for this discussion it's reasonable to use that as an index of how many headache patients we had for whom our doctors were worried about SAH. So we have about a 2% prevalence of disease in our "acute" headache population (18/900). The traditional data was that CT was about 90% sensitive for SAH, so the negative predictive value of a CT is very good -- somewhere well north of 99% likelihood that the patient does not have SAH. Now you can play with the numbers and tighten it up a bit by more rigorously screening out headaches that are not "worst ever" and not sudden onset, but even if you get to a pretest prevalence of 10%, which would be quite high, the NPV is still very good, certainly better than we can rule out other serious diseases like PE or unstable angina.

But this was very rough math from a single practice with small numbers. So it is not exactly something I was able to endorse as a standard of care. Just contextual information I could offer a patient guiding them whether or not to accept the LP I was offering. Most declined, but some preferred the assurance that the gold standard test offers.

I've been quite pleased, though, to see more and more new and more rigorous data emerge on the topic. It seems, ever so slowly, the tide of opinion is turning against the routine LP.  First David Newman over at SMART EM did a great deep dive on the topic, showing that for LP, the Number Needed to Treat is somewhere around 500, which means that you'll do a lot of LPs to find a single SAH in a patient for whom it will make a difference. (Updated podcast on SAH here - worth listening to!) Then there was the Perry article in the BMJ last year which showed that the sensitivity of early CT is very very good, perhaps as high as 100% for SAH. Then there was this August 2012 article in the highly influential journal, Stroke, authored by none other than Dr. Jonathan Edlow:

Diagnosis of Subarachnoid Hemorrhage : Time to Change the Guidelines?[...] Given this analysis, we believe that practice should change. Neurologically intact patients who present with thunderclap headache and undergo CT scan within 6 hours of symptom onset no longer need an LP to exclude SAH if the CT scan is negative.
This is the same Dr Edlow who was lead author on the ACEP clinical policy, only 4 years ago, which did recommend routine LPs! (Link: PDF) The times, they are a-changin'!

So I feel comfortable claiming victory here. I was right all along and shame on you for ever doubting me.

(insert nuanced discussion here about shared decision-making with patients and the need to assess each patient as an individual.)

26 November 2012

This is what it takes to get published nowadays

For Pete's sake. This popped up in my newsfeed today, with multiple lay media citations:

Pediatric Inflatable Bouncer–Related Injuries in the United States, 1990–2010 
METHODS: Records were analyzed from the National Electronic Injury Surveillance System for patients ≤17 years old treated in US emergency departments (EDs) for inflatable bouncer–related injuries from 1990 to 2010. 
RESULTS: An estimated 64 657 (95% confidence interval [CI]: 32 420–96 893) children ≤17 years of age with inflatable bouncer–related injuries were treated in US EDs from 1990 to 2010. From 1995 to 2010, there was a statistically significant 15-fold increase in the number and rate of these injuries, with an average annual rate of 5.28 injuries per 100 000 US children [...] Most injuries were fractures (27.5%) and strains or sprains (27.3%), and most injuries occurred to the lower (32.9%) or upper (29.7%) extremities.  
CONCLUSIONS: The number and rate of pediatric inflatable bouncer–related injuries have increased rapidly in recent years. This increase, along with similarities to trampoline-related injuries, underscores the need for guidelines for safer bouncer usage and improvements in bouncer design to prevent these injuries among children.
Sweet Jesus on a pogo stick. So you mine a database for some trivial but catchy mechanism of injury and slap a ramshackle statisical analysis on it (somewhere between 30K-100K injuries? that confidence interval is as wide as a barn door) and presto blammo you're in Pediatrics and USA Today and on CNN solemnly intoning on the dangers of letting your kids go to Jump Planet.

Is this where we are as a society? Have we run out of actual public health concerns that we find this sort of minutia worth researching? Or are car crashes and gun accidents and drug overdoses gotten too boring to publish and report on? Or, I suspect, is the culture of academia so degenerate that the mandate of "publish or perish" overwhelms common-sense judgement in deciding whether a topic is publication-worthy? Yup, that's it. Bring on the trivia!

Next thing you know they will be warning you of the dangers of tripping over your pets. Oh, wait. They already did that study.

That meteor can't get here soon enough.

EDIT: Great minds etc etc etc

22 November 2012

Be careful out there

Cute little PSA -- Dumb ways to die

21 November 2012

On objectivity

Now, I'm not a radiologist, oncologist or an epidemiologist. So I am not claiming any expert opinion of the science, but I was not surprised to see yet another major article released regarding the value of early detection of breast cancer via screening mammography -- it tends to detect a lot more early cancers, but doesn't seem to reduce the number of advanced cancers.

From today's NEJM:

The introduction of screening mammography in the United States has been associated with a doubling in the number of cases of early-stage breast cancer that are detected each year, from 112 to 234 cases per 100,000 women — an absolute increase of 122 cases per 100,000 women. Concomitantly, the rate at which women present with late-stage cancer has decreased by 8%, from 102 to 94 cases per 100,000 women — an absolute decrease of 8 cases per 100,000 women. With the assumption of a constant underlying disease burden, only 8 of the 122 additional early-stage cancers diagnosed were expected to progress to advanced disease. ... breast cancer was overdiagnosed (i.e., tumors were detected on screening that would never have led to clinical symptoms) in 1.3 million U.S. women in the past 30 years. We estimated that in 2008, breast cancer was overdiagnosed in more than 70,000 women; this accounted for 31% of all breast cancers diagnosed.
This is not a new finding at all. Numerous previous studies have been published questioning the value of rountine mammography screening in younger women, and a major controversy was ignited a few years ago when the experts at the USPSTF finally recommended that women under the age of 50 not be routinely screened by mammograms.

As I disclaim above, I do not have a vested interest or a strong opinion on this, though I do have a bias towards accepting the conclusion as the body of science accumulates. But one thing that I noted in the media coverage of this newest study was this statement which was almost universally cited:
The American College of Radiology issued a statement saying the report was "deeply flawed and misleading"
While I understand that journalists should try to present both sides of an issue, especially one which is so controversial and emotionally charged, maybe an organization which has such a strong, vested, economic interest in the value of mammography might not be the most credible source to turn to for an expert opinion? As Upton Sinclair famously said, "It is difficult to get a man to understand something, when his salary depends upon his not understanding it!"

Discharge a PE? that's crazy talk!

So I recently sent home a patient with a Pulmonary Embolism (PE) for the first time. Or perhaps I should say that that it was the first time I've knowingly sent home a patient with a PE, but that's neither here nor there.

This was an unusual case, to be sure. The patient was young and healthy, a triathlete in exceptional condition. He had had arthroscopic surgery on his left knee about a month ago, and a few days after that developed this sharp pleuritic left chest pain. The pain was quite severe, but he ignored it for about three weeks until finally, since it wasn't going away he presented to his doctor, who diagnosed the PE on CT and sent him to me for treatment.

The PE was small but not tiny, segmental as I recall. He otherwise looked great, with no tachycardia or shortness of breath. Functionally, he was doing great. He wasn't back to running yet, but he was cycling and swimming and performing at about his usual level. So I guess that made him functionally "well-preserved." Given that he had symptoms for over three weeks, I guess that qualified him as stable, so we started him on low molecular weight heparin (LMWH) and sent him home.

And I suspect that this is where we are going in the future - outpatient management of stable PE patients.

I threw out the question on twitter at 2AM, and woke to find a vigorous conversation ongoing on the topic among ER physicians on three continents, including one principle investigator of a major trial on the topic. Twitter is awesome. You can read the conversation, in part, here on Storify. Michelle Lin over at Academic Life in Emergency Medicine put together a PV Card on the topic and received some more feedback.The consensus was that most non-US ER docs have already or are beginning to embrace the concept of risk stratifying and discharging some PE patients, while the US practice has not moved much and is deeply skeptical of the idea.

Can we safely send home some PE Patients? 

There are many patients with PEs who are clearly ill. They're easy to spot if you've a smidgen of clinical judgement - they're dyspneic, tachycardia, hypoxic, hypotensive, etc. There is a nicely validated scoring system to sort out those who are more likely to have a bad outcome, and presumably, these folks are the ones who would benefit from hospitalization. But, of the well-appearing PEs with lower risk, the risk is still not zero. There are some people who present with small clots who will proceed to have recurrent embolic events and die. We've all seen it. Is it possible to quantify how commonly that happens? More importantly, is it possible to predict which of the well-seeming patients are more likely to have these bad outcomes?

There is some research out there to support a selective approach to outpatient management of PEs. There was this study which supported the safety of early discharge. More recently there is the Hestia trial which was a prospective study supporting the safety of outpatient treatment, and one unblinded randomized controlled trial of outpatient treatment which also supported outpatient management. If you haven't, I would strongly encourage you to listen to Rob Orman's ERCast podcast on this topic.

I would also add that the value of inpatient treatment as currently practiced seems limited. The well-appearing PEs in the US tend to get a very brief inpatient stay, less than 24 hours, which I suppose might screen for stability but I'm not sure there's any evidence to support the utility of the brief admission. Talking with some european docs, not only is outpatient management common over there (in some countries), it can take 3 days to get a CT-PA, so in many cases they are discharging suspected PEs on LMWH until they get their study, and if it's positive then they get admitted. (Which makes no sense at all, but there you have it.)

The signs seem pretty clear: low-risk patients, as judged by an objective risk stratification score like PESI plus some good old-fashioned clinical judgement (size and location of clot, total clot burden, risk indicators maybe not built into PESI) probably will allow us to safely discharge patients with PE. But can we get there? I'm not sure. The culture of the ER, especially with a perceived high-mortality diagnosis like PE, is highly risk-averse. Merely mentioning the notion elicits gasps of horror from my colleagues, and mutters of "over my cold, dead body." A further, and larger, obstacle to changing practice is our zero-risk-tolerance, highly litigious medical environment. Who wants to be the first ER doc sued for sending home a PE? Plaintiffs' experts will be lining up around the block to testify against you.

And this is a problem. We know that some people with PE will suffer recurrent embolic events despite anticoagulation, though it's a small number. Being hospitalized will not prevent the recurrent embolization, though it may provide earlier detection and therapy. Since we do not know in advance among the low-risk group who will suffer recurrent emboli, it's a catch-22. You can admit them all, a very large number of patients, to detect a very rare complication, or send them home with the risk that when a complication does happen, you are ar risk for being "blamed" for the decision to discharge.

I think we are not ready for prime time here, but it's coming. US docs will demand better data before warming to the notion. Strong institutional support will be needed from hospitals, meaning defined care protocols supporting the practice, in order to convince skittish doctors that they have the backing of the facility in the event of a bad outcome.

You've been warned.

15 November 2012

The Catch-22 of documentation fraud

I just wanted to expand on something I wrote yesterday, which relates to my other sort-of-recent post on upcoding. I wrote, about scribes and compliance:

Knowing that the scribe cannot document a complete ROS unless I actually did that ROS, I am more compulsive about making sure I hit all ten systems. (Even when it's not clinically relevant. Such is the Kafkaesque world we live in.) And I make sure to do a full exam where before I may have elided over a few systems. This is, of course, only for cases where the complexity of the case will justify a service level requiring the complete H&P. 
This hits at the heart of the upcoding debate. Remember this front-page article in the New York Times from six weeks ago, in which the increased billing levels of ER doctors is asserted as prima facie evidence of fraud and abuse, and the follow-up in which the powers that be asserted their intent to reclaim these hundreds of millions of dollars in "inappropriate payments." We are not looking at a hypothetical threat here, and the financial risk to care providers is enormous.

The rules, for those not familiar with them (and who the hell would be reading a blog post about medical coding if you weren't?) are that to bill at a level 5, which is the highest ordinary level of service in the ER, the physician must document the following:

  • An extended history 
  • A complete review of systems
  • A comprehensive exam 
  • High complexity medical decision-making

In order to quality for a level 5, all of these must be met, but the sine qua non is the medical decision-making (MDM). This is, in fact, the ultimate driver of the visit level. MDM consists of three components: the number of diagnostic options (i.e. your differential), the amount of data you must review (i.e. tests, re-examinations) and the risk inherent in the presenting problem. If the MDM isn't met, no matter how nicely documented the rest of the chart is, a high service level may not be justified. To put it another way, an ankle sprain, no matter how thoroughly documented, is still just an ankle sprain.

Previously, it was common to have cases "downcoded" when a doctor had a high-complexity MDM but slipped up on the other items, most commonly on the ROS. Over the years, physicians have gotten better educated about the system and more sophisticated at making sure the ROS and other requirements have been met so that the billing level can, appropriately, be determined by the MDM.

This rankles. Always has. When I see a patient with chest pain and a heart attack, in order to get paid appropriately I have to ask a bunch of completely irrelevant questions about unrelated systems: do you have burning when you urinate? Do you have any rashes? Nobody would argue that the complexity and risk don't justify the level 5, but I have to document a bunch of medically unnecessary trivia to compliantly bill at the level the MDM deserves.

And this is where the bureaucratic hassle now becomes a catch-22. "Medical Necessity." Medicare considers it fraud to bill for things which are medically unnecessary. If I see an ankle sprain and order blood tests and a CT scan to try and get the bill up to a high level, that legitimately is fraud because the tests ordered are not medically necessary. But what is happening now is that Medicare (in the form of the private contractors who administer it regionally, along with some private payers) are reviewing charts and claiming that the physicians are fraudulently upcoding because we are documenting complete Reviews of Systems when they were not ... wait for it ... medically necessary.

To be clear: Medicare set the rules, and made them arbitrary and disconnected from reality, and now is coming back and punishing physicians for attempting to follow the rules to the letter of the law.

And the format this takes is scary. You get a letter from the Medicare carrier (or a RAC or a Medicare Advantage administrator) telling you that you've been reviewed, found guilty of upcoding, and this finding, based on a handful of charts, is extrapolated back several years. The result is a large demand for reparations, usually in the mid-to-high six figures. The physician group can either write a check or lawyer up and argue it chart by chart in front of an administrative law judge.

What I hate about this is the underlying dishonesty. This is about saving money. I get that, and that is in fact a reasonable goal. Healthcare is astoundingly expensive, and as a society we need to ratchet back the expense. If there's an argument to be made that physicians are paid too much, then let's have that debate on its merits. But the attempt to save money by harassing physicians and exploiting the contradictions within the rules that the government itself wrote is beyond maddening.

14 November 2012

Write this down for me

I have a lovely pen. It's a Mont Blanc Meisterst├╝ck fountain pen. My group bought it for me on my tenth anniversary as a partner in our Emergency Medicine practice.

It's a luxury I would never have paid for myself, though I have loved and used fountain pens since I was in college. Ironically, about the time I got it, the window of opportunity to use it in my professional life closed. For a decade, we had a hybrid paper-and-dictation documentation system, but around the time I hit my milestone, we went to an Electronic Medical Record (EMR). And with that, I never again had to touch pen to paper, except to sign the odd prescription. Such is life.

I am a computer guy, tech-savvy and fearless, and I was one of the few docs who saw the move to an EMR as a good thing.  My documentation improved, and now that we are with Epic I would say it's even better. As I am a quick typist, the workload of documentation was only modestly increased by the transition to full physician documentation in the EMR. The other docs in my group varied in how well they adapted, from a few whose productivity improved, to the mass who accepted it with grumbles and minor complaints, to a few outliers who simply refused to use it at all.

Recently, though, we started a pilot program using medical scribes.

Honestly, I resisted the scribe initiative for years, though there were a few docs who really wanted them. I wasn't opposed, but I was too busy to do it, and it wasn't high enough on my priority list to make it happen. It finally happened when I challenged one of our younger, energetic docs to "make it happen," and she went out and did just that. Very impressive initiative. She formed a committee, put together a business plan, had presentations from scribe vendors, took competitive bids, and soon enough there were young enthusiastic faces greeting us in the ER. I watched, bemused, from the sidelines for a couple of months and finally took the plunge and signed up for a scribe myself for a few shifts.

These are my thoughts and observations so far, after about a dozen shifts with my own personal scribe.

First, the general structure of the program, for our group. We pay a flat hourly rate to a scribe vendor. The vendor recruits the scribes from a local university, mostly pre-med students, and manages all the HR functions associated with such a program. Docs who are interested in having scribes sign up and choose which shifts they want a scribe for. The cost of the scribe is deducted (pretax) from the doc's individual paycheck. The program is entirely voluntary and about a third of our docs have signed up so far, usually just for the busier shifts.

The social aspect of having a scribe is more than a little weird, though I got used to it quick enough. I added another line to my standard introduction: "I'm Dr Shadowfax, and this is Jenny, who is working with me today." Almost never has the presence of the scribe occasioned any further comment or discussion. The scribes step out of the room for pelvics or other uncomfortably intimate exams and are generally invisible during the H&P (hidden by the large monitor of the computer on wheels they bring with them). During the physical exam, I verbalize what I'm seeing/doing, as if I am talking to the patient. "Your lungs are clear and your heart is regular without murmurs." This allows the scribe to document my exam in real time, and, from what I can tell, patients seem to like it, since they are getting a sense of what I am looking for and seeing. If there are "issues" such as psych, substance abuse or simply an unpleasant patient, I'll wait till we're out of the room to tell the scribe what I want documented.

I've never had a secretary or personal assistant before and have always prided myself on self-sufficiency, so it feels odd to have someone whose whole job is to do the little scut work (like putting a chart in the rack or pulling reports off the fax machine) for me. I can do that perfectly well myself. I can also document perfectly well myself. Better, in fact, than most. Getting over the idea of someone else doing "my" work for me has been and remains probably the biggest barrier for me in fully accepting the scribe. But these small efficiencies are of course the whole purpose of having a scribe in the first place, so I am getting over that.

The workflow is quite different now. It's actually very pleasant. I have the freedom to simply sit down and talk to the patient. I can take a bit longer and have more of a free-flowing conversation. I'm facing the patient, not facing a computer screen, I'm not making notes on a clipboard, and I'm not frantically trying to remember the necessary data points for the chart. I just chat. I feel like I have more mental energy to spend on the patient and I can simply forget about the chart, confident that the scribe is capturing the important data points. Simply put, I can focus on the patient, and I feel like that allows me to be a better doctor. I suspect, though I have no proof, that it also helps with patient satisfaction, which matters a lot these days.

The quality of the documentation is a little more variable. It's hard to let go of control of the chart. There are some odd little verbal tics some of the scribes have that I would never use. To me, reading these charts are like fingernails on a chalkboard, though they're perfectly accurate and acceptable. Sometimes a really important historical point gets left out of the chart because the scribe didn't realize its significance. It is very important to proofread the charts and make sure they say what you need them to say. I'm learning to "let go" and not spend so much time editing each chart that it negates the point of having a scribe in the first place. And I think the scribes, as they learn, are getting better and better at picking out the important bits of the conversations they are documenting. When there is an important point I want emphasized I can simply repeat it back to the patient as a cue that I want this verbatim in the chart, and if I note an omission I review that afterwards with the scribe as a "teaching point" for them, as I would with a med student. Since they are all pre-med, they really seem to appreciate it. One of the best points (and a pleasant surprise) was when I reviewed my charts and found entries like:

1645 - patient re-evaluated. Abdomen still nontender. Taking po well.
or, 1015 - neurosurgery paged. 1025 - Dr Shadowfax speaking with Dr Jones, who requests MRI

Stuff that I had never before had the discipline to document and time, now 100% of the time in the chart. This is a huge benefit, especially when it comes to med mal defense.

Another thing that this has forced me to do is be more rigorous with my H&P. Once you have been working in an ER for a while, there are quite a few diagnoses you can literally make from the doorway. Say, a kidney stone. I don't need to do a Review of Systems or even a physical exam for a kidney stone patient, and over the years I may have become a little lax on this point from time to time. But we have trained the scribes that "if it didn't happen, you cannot document it." So now, knowing that the scribe cannot document a complete ROS unless I actually did that ROS, I am more compulsive about making sure I hit all ten systems. (Even when it's not clinically relevant. Such is the Kafkaesque world we live in.) And I make sure to do a full exam where before I may have elided over a few systems. This is, of course, only for cases where the complexity of the case will justify a service level requiring the complete H&P. So the scribe effectively helps keep me honest and improves my compliance.

The productivity side is also a net positive. Once I learned to let go and trust the scribe to get all the charting with minimal oversight, this freed up my time enormously. I can go from room to room to room seeing new patients, with only a brief interlude to enter orders (which the scribes are not allowed to do in our hospital). I've always been able to see 2+ patients per hour with no problem, and with the scribes 3+ has been easy, when volumes permit. I think I could go even higher but I haven't had a really busy shift since the program began.

At this point I am, I think, not making money on the scribes. I think, in fact, that I am losing money. I have been told by experts that in the startup phase of a scribe program you should expect to lose money for the first year. This seems reasonable with our experience. We have 8 docs on duty in our ER at peak times, and only a fixed number of patients. To the degree that I can see more patients, that's taking money from my partners' wallets, which puts an upper bound on my appetite, out of courtesy. Worse, if I have a scribe on a slow shift, it grates on me that I am paying for them do essentially do nothing. If I have a scribe, I feel pressure to be more productive than I otherwise would. Over time, I hope, we can contract physician staffing to the point that we will all realize increased productivity and revenue. This requires more than a 1/3 physician buy-in, which we have yet to achieve. We will see. For the moment, I can at least hope to break even on the program, though some of it may come at my partners' expense. Maybe that will induce them to get their own scribes as a defensive measure.

The final, and perhaps most important, point for me is this: quality of life. If I have a scribe shift, it's a good shift. I save so much mental energy not having to chart. When I have a five-minute conversation with a patient, ordinarily, I am carefully committing about 30 key points to my short-term memory. I then have to dash out of the room, while it's still fresh in my mind, and enter that into the computer. I never realized how much that was wearing me down till I didn't have to do that any more. My "external memory" is passively (from my point of view) capturing all these data points and I can focus on my clinical impression from the get-go. I can forget the details and focus on the big picture. The saved "brain strain" takes a busy shift and makes it seem nearly effortless. When I have five free minutes, which is rare enough, I can check twitter or my email or text my wife rather than frantically trying to catch up on my charting. And when my shift is over, I am generally done with my charts and can walk out the door as soon as the last patient is dispo'd. Granted, I was generally one to leave at the end of my shift even without a scribe, but that took work. Now it's easy. I like my job better. I've never felt like I was one of those docs susceptible to burnout, but it is endemic within emergency medicine, but for someone who is riding that razor's edge, a scribe could be the difference in job satisfaction between having to leave the field and keeping their career going another decade.

I'll update this when I've more experience, but so far I am continuing my scribe utilization and would describe myself as very happy with the experiment. Now I just need to figure out how to get them to blog for me.

01 October 2012

Are low acuity patients congesting the ER?

I had an interesting twitter chat the other day with one of my colleagues, a young ER grad studying healthcare policy by the name of Seth Trueger. You should check out his blog over at MDaware.org. The twitter conversation wound up involving about a dozen ER docs and nurses. You can review an edited summary over at storify. Yes, of course Gruntdoc was involved. That guy never shuts up.

The point in contention is an interesting one: we know the nation's ERs are overwhelmed and overcrowded. That's old news. We also know a big driver of this is boarding of admitted patients in the ER due to limited inpatient beds. If you're a 20-bed ER and you're boarding 5 patients, you've lost 25% of your throughput capacity. Common sense that this is a big issue. But, the argument hinged on, what is the contribution of the proportion of ER patients who "don't need to be there," the patients whose care could have better been delivered elsewhere?

There was, I hasten to add, no disagreement as to why the "worried well" and the "walking wounded" come to the ER. PCPs are too busy to see them, both because their clinics are booked up and also because they often don't have the resources to provide much in the way of acute care services — IVs, nursing staff, etc. In part this is because it may not be economical to provide this care in the office. Furthermore, most medical offices are only open during working hours and acute care centers are only slightly more accessible. There are also many patient-side barriers, including the hassle involved in making an appointment, the need for co-pays and insurance status. So, given the many obstacles involved in getting care in the more appropriate, most cost-effective settings, the ER becomes the default for many of these patients.

Now Seth argued at some length, that these low-acuity, ambulatory care patients were "a drop in the bucket" of ER overcrowding and cited the example of the ubiquitous URI patient who can be seen and streeted in less than 20 minutes. These folks, he and others argued, are not the problem that our nation's ERs struggle with. This is, I might add, in line with ACEP's argument that only 7% of ER patients are non-emergent.

I have pointed out in the past that my BS-meter starts pinging when people start claiming that the ER is only caring for emergent patients and that non-emergency cases are rare. So this set me off, of course. My perception — and that of many of us in the trenches — is that we are absolutely beset by non-emergencies and that the ER is viewed by many as the "convenience clinic," if not the "vicodin clinic." But is this true? How can we quantify this?

ACEP has, for their PR campaign, relied on the National Health Statistics Report to establish whether a patient in the ER was actually an "emergency." The problem with that method is that it takes the 5-point triage scale perhaps a bit more literally than the typical triage nurse does. To the point, it considers a level 4 "green" patient to still be an emergency since the definition of that level is that a patient needs to be seen in 1-2 hours. Which is not at all the way it is applied in the real world.

In this case, I would advocate using the coded level of service by the ER physician to stratify patients. Low acuity probably correlates nicely with the E/M code applied. The lower-level ER codes, level 1, 2 and 3 tend to be associated with not being admitted to the hospital, with not receiving advanced imaging studies like CT scans and with not receiving complex work-ups with blood tests, CT scans and X-rays. Put simply, an ER patient level 3 or lower may receive one or two ancillary tests (like a simple x-ray or a single lab test) but not much else before they get bumped into the level 4 range. So a level 3 patient is one that generally is simple and hopefully quick and one who, in theory, could have been cared for elsewhere were an appropriate care environment available.

So I pulled our numbers. Our ER is a pretty typical, moderately high-acuity community ER. We seen nearly 300 patients every day, and of those about 20% get admitted to the hospital. Our numbers indicate that we see about 100 patients every day who fit this definition, about 1/3 of the total volume and about 45% of the non-admitted volume.

Is this "a drop in the bucket"? Are these patients who we should be seeing and streeting in 20 minutes? Put simply, the answer is no.

When I look at the time stamps on the charts of these patients, it's clear that they are not in and out of the ER all that fast. It's all relative, of course, and your mileage may vary. We have a very efficient ER. We don't board patients and our average door-to-bed time is about 15 minutes. Pretty good. But for the majority of patients who are not admitted, the typical time in the ER is still in the 1-3 hour range:

The lower-acuity patients are there less time, it is true. About 1-2 hours on average.  Why so long?

There are two factors at play here. First of all is the fallacy that just because a patient is easy and quick from the physician's point of view, they are also quick and easy for the facility. But unfortunately, the many steps which a patient must go through in the ER are fixed and take about the same amount of time for each and every patient. Let's use our shop's numbers for perspective. The patient must present at the greet desk and be entered into the system, must be placed in a bed, must be triaged, must be registered, and there are often obligate waiting periods between each step. In a highly efficient ER, where patients are bedded rapidly, much of this takes place in the treatment area. Then the doctor swoops in, does his or her black magic and is gone. If there are orders or treatments to be applied, that takes time from when the orders are entered to when they are executed. Eventually, the physician enters a discharge order, and after some time, the patient is actually discharged. Each step in this process takes time. So from door to bed: 15 minutes. The triage process itself takes a good ten minutes (bear in mind all the irrelevant data points ERs are required to capture, like domestic abuse screening, etc). The patient must be registered, which takes another 5-10 minutes. Then there is a waiting period until the doctor gets in to see the patient. That's another 5-10 minutes. So we are talking 30-45 minutes even before the doctor assesses the patient, on average. Assuming the doctor has no orders, the time from the decision to discharge to the actual discharge may be another 15 minutes, depending on nursing workload. So in a typical case where the doctor's face time is very minimal and there are zero orders entered, the process phase of the ER visit takes an entire hour! This is, I might add, a "best case" for an ER visit. (Bear in mind that we are talking average times here. So for each person who comes in at 5AM and is seen immediately there's one who comes in at 7PM and has to wait twice as long.)

The second fallacy at play is the idea that a low-acuity patient is in fact a low workload patient. Seth cited the URI patient. Nothing faster from the MD's point of view. In our department, that represents about 4% of patients. That's maybe 15% of the total low acuity patient load. What are the other typical level 1-3 patients here for? Well, based on our ICD-9 coding, in rough order of frequency, things like:

  • Back pain
  • Headache
  • UTI
  • Neck Strain
  • Minor head injury
  • URI/Bronchitis/Pharyngitis/Sinusitis
  • Extremity Cellulitis
  • Laceration
  • Dental pain
  • Extremity sprains/strains/contusions
  • Pediatric fever (non-infant)
  • Abscesses
  • Corneal Abrasions
  • Rashes
  • Allergic reactions

A lot of these are, in fact, not at all easy from a time point of view. Abscesses/Lacerations/Corneal abrasions all take physician time doing procedures (and associated set-up time, etc). The musculoskeletal injuries often require imaging and splinting. The back pain/headache/cellulitis often require medication administration. Concussions and toddlers with fevers may not require tests but do consume a lot of physician time face-to-face. Some of these cases require multiple physician assessments. Some require labs. Each additional step adds time, sometimes quite a lot of time, to that one-hour best case baseline I described.

So what's the total time burden?

Our experience is that for the ambulatory population, i.e. excluding admitted patients, we have about 510 patient-hours per day in our ER. Of this, almost exactly a third, 160 patient-hours is attributable to the lowest-acuity patients, the E/M level 1-3s. That correlates also to nearly a third of physician staffing and RN staffing. Bear in mind that our institution just built a whole new hospital at a cost of $500 million, in part because the ER needed a much larger physical plant. The costs involved in this care are not insignificant, and the burden placed on the nation's ERs from these less acute cases is major, not at all a drop in the bucket.

I want to take a moment here that I am not commenting on whether these patients should be in the ER. In the current healthcare environment, they have to be here because there is nowhere else to go as often as not. And many of these cases will always be with us: if you have neck pain after a car accident at 2AM, the ER is the right place to be treated, even if it winds up being just a sprain. We embrace our role as the care provider of last resort, the ones who are always open and always available, no matter what. It is also true, however, that we are an expensive place to receive care. The fixed costs of operating an ER are horrendous, compared to a clinic. Our health care system would be far better served if there were accessible sites of care that could care for these less intense patients in a more cost effective manner.

24 September 2012

Upcoding is in the eye of the beholder

There was an interesting and important article in the NY Times the other day about the gradual increase in the average E/M coding levels used by doctors over the last few years. For the non-docs, med students and ER trainees out there, here is a brief summary of the way physician billing works in the ER:

During and after the patient encounter, the physician creates a medical record. This is a pretty standardized document including the history the doctor got from the patient, the exam, any tests and the medical decision-making. And, of course, the diagnosis. This gets reviewed, typically, by a professional coder who then applies Medicare's rules to determine what level of complexity the service was, and the associated cost of that service. (In some cases, the doctors code their own charts, and some are automatically coded by a computer, but both of these are exceptions to the rule.) ER visits are coded on a 5-level scale, with a Level 1 (99281) being the simplest and a Level 5 being the most intense. These are called E/M codes because they refer to the "Evaluation and Management" of the case, as opposed to, say, a surgical procedure.

The amount Medicare pays the doctor for an E/M code in the ER ranges from about $20 to $175. (If you've ever received a multi-thousand-dollar ER bill, chances are good that was the hospital bill. The physician's fees are much more modest in most cases.

Medicare, of course, tracks utilization and recently published a report (PDF Link) about the recent trends in the doctors' use of the codes. What they found is that over the last decade, we have shifted from using the lower codes and towards the higher ones:

This is where it gets interesting. The feds and some consumer watchdogs view this trend as clear evidence of fraud and abuse, that the physicians are "upcoding" the visit levels to increase their income. On the other hand, I've spent the last decade trying to educate physicians on how to document the patient encounter so that you can accurately capture the legitimate value of the service provided. So I look at that trend and think to myself "Job well done."

From the perspective of a practicing physician, the rules that govern the documentation required to capture a service level are deliberately onerous and designed to produce downcodes. They require the doc to collect far more data than is actually necessary based on the actual condition of the patient. You forget to check one box, you leave out one required element, and despite the complexity, gravity and risk of a patient's condition, you will not be paid for the service. For years, we have been losing money to these archaic rules (they date from 1995), and we have been struggling to stem the leakage of revenue from our practices.

The industry responded to these rules by developing tools to comply with them. The first thing was to have professional coders. Prior to 2000, a substantial majority of ER charts were hand-coded by the physician; now that is quite rare. Then we got templated paper records which prompted the docs to get all the required data points. Now we have EMRs which do the same thing more efficiently. It's no surprise that as an industry we have gotten better at meeting the guidelines.

And then there is the fact that the ER is a different place than it was in 2000. Our patients are older and sicker. We do more in the ER than was true in the past. Patients are rarely directly admitted any more, but rather get the majority of their admitting workup done in the ER. I don't know how much of the skew in the above graph is due to these factors, but they shouldn't be disregarded.

But I don't like where this is going. The government is desperate, understandably, to save money on healthcare expenditures. They seem to have assumed their conclusion that the increased coding levels is fraudulent and unjustified, and there seem to be few voices disagreeing with them. Furthermore, there is some inappropriate upcoding, and it's very easy for a patient with an egregious bill or a certain physician (or group) who pushed the envelope too far to be held up as anecdotal proof that doctors are all a bunch of thieves.

The article quotes some insurers as saying that nearly half of the charges being submitted are "upcodes," without noting that the insurers have a vested interest in not paying and may not be presenting an honest picture of the claims. We had a run-in with a certain national insurer who analyzed a few hundred claims, found that 80% of them were upcoded and demanded a couple of million dollars in restitution. We fought back, of course, defended each chart on the merits of the care, and established that in fact 97% of the time the correct, higher, code was appropriately used. That the insurers are not necessarily honest players here passes unremarked.

I think changes are coming, and it worries me. We already have one insurer saying that they simply won't pay for cloned records — but how will they tell if a chart has been cloned? Do charts have DNA? I think what we are going to see is the carriers becoming very aggressive in arbitrarily denying payment or demanding restitution. I would love to see revised and streamlined coding rules to replace the old ones, but I suspect that if that were to happen, the new rules will be designed to be more restrictive and more burdensome.

Just go to the ER, redux

Sweet Breakdancing Jeebus, we are back to this:

Downplaying the need for the government to ensure that every person has health insurance, Mitt Romney on Sunday suggested that emergency room care suffices as a substitute for the uninsured. 
"Well, we do provide care for people who don't have insurance," he said in an interview with Scott Pelley of CBS's "60 Minutes" that aired Sunday night. "If someone has a heart attack, they don't sit in their apartment and die. We pick them up in an ambulance, and take them to the hospital, and give them care."

If this quote sounds familiar, it was previously uttered, near word for word, by the intellectual giant who was the standard bearer of the Republican party for most of the last decade, George W Bush (remember him?). He said, in 2007:
"The immediate goal is to make sure there are more people on private insurance plans. I mean, people have access to health care in America," he said. "After all, you just go to an emergency room."
I was going to expound on my long-held point that "Just go to the ER" is not a substitute for national health policy, and that the ER is unable to provide comprehensive and preventative care yadda yadda yadda but fuck it. You all know that already. I can't stand it. I'm gonna go have a drink instead.


29 July 2012

The Looming Physician Shortage

Hardly news, but worth repeating:

The Association of American Medical Colleges estimates that in 2015 the country will have 62,900 fewer doctors than needed. And that number will more than double by 2025, as the expansion of insurance coverage and the aging of baby boomers drive up demand for care. Even without the health care law, the shortfall of doctors in 2025 would still exceed 100,000.
In fairness, I've always felt these numbers are a bit inflated. Still, the shortage is quite real, and as the article makes clear, much worse in the less desirable areas of the country to live and work. Being in a trendy cosmopolitan area, I have no trouble recruiting. In fact, I tend to be swamped with applicants. However, I know of many rural ERs where they have to pay recruiters exorbitant sums to induce a board certified ER doc to come work there. There are also many more ERs where no certified ER docs will work (or too few, in any case) and they wind up making do with family practitioners or docs with more dubious backgrounds.

On an unrelated note, the job market for midlevel health care providers such as nurse practitioners and physician assistants is booming. 

11 July 2012

Medicaid clients in the ER: emergent or not?

It's an article of faith in the Emergency Medicine community that patients on Medicaid over-use the ER; many would go further to say that they abuse the ER. I am (rarely) inclined not to be so judgmental on this point because the cause is in part rooted in inadequate access to ambulatory care. But the fact of Medicaid overuse of the ER is probably one of the very few things that most ER providers, doctors, nurses and techs would be in unanimous agreement on.
 So I was surprised to see Sarah Kliff over at WonkBlog link to the following article:

Study: Medicaid patients aren’t using the emergency department for routine care

Huh. That's counter-intuitive. But I am not sure it passes the sniff test. I'm about to go on shift, so I haven't had time to dissect the study yet. The gist is that Medicaid patients do visit the ED more, but not for less acute conditions, compared to privately funded patients. Here's a link to the study proper. A couple of quick thoughts:

1. This contradicts a reasonably robust body of research (for example) and the rule that applies to polls also applies to studies: if a given result is markedly different from the pre-existing data, it's more likely to be spurious.

2. I can't tell on first blush whether this was a scientific paper, peer reviewed, or from an advocacy organization. The study is published on a website for a policy think tank, Health System Change, which automatically makes me wonder about possible bias in the process. I am not implying shenanigans, but I don't automatically trust think tank papers, especially when they support a certain agenda.

3. The data source is the National Hospital Ambulatory Medical Care Survey of Emergency Departments, which is a great and reliable data source. However, I've previously seen its data abused to support the absurd claim that only 7% of ER patients are non-urgent. So I'm similarly skeptical of this result, especially when they seem to have dropped one of the triage categories (from a five-point to a four-point scale). F

Furthermore, triage level has at best, a poor correlation with the true urgency of a patient's condition. By which I mean that it provides a reasonable sort-order to guide which patients get seen first. But it is a poor guide to whether a patient needed to be seen in the ER or whether he or she could have been cared for in another, less resource-intensive setting. For example, a big, ugly hand lac may well be triaged as a green, since as long as the bleeding is controlled, it can wait, but it's not appropriate for an internist's office. conversely, an 9 month-old with a fever may be triaged as yellow or even orange, but would be perfectly appropriate for a pediatric clinic.

I'll review this in more detail when I have time and energy, but I would be very very cautious in accepting this as strong evidence against the conventional wisdom that Medicaid patients do over-use the ER.

28 June 2012

Defending the Mandate and its Tax/Penalty

From a contributor at the Daily Dish:

I do not understand why Democrats don't embrace the newly defined "tax", saying: you bet we raised taxes, but not on the hard-working, responsible middle class. This is a tax on those deadbeats who don't pay for their own insurance but still expect care when they show up at emergency rooms. It's a tax, all right, and I think we should agree to raise it even higher so they have more of an incentive to buy their own damned insurance and leave the rest of us alone. Let the Republicans protect the rights of deadbeats; Democrats are fighting for people who play by the rules.
This. This is exactly the right approach.

Matt, in the comments to the last post, asked a good question:
What is to stop people from paying a marginal penalty until they get a catastrophic disease, and then applying for insurance? This seems like rational behavior that will undercut the whole premise of "insurance". If I could buy car insurance with "pre-existing conditions" I'd never buy it until after I was in an accident.
Fair question. As I understand it there are two factors that make people willing to sign up for insurance when they are still healthy. The first is simply risk aversion. If you get hit by the proverbial bus or have any unexpected medical emergency, you can't sign up for insurance that day, or retroactively. So the possibility of a financially crippling medical emergency looms if you are healthy and uninsured. In fact, I believe you can't sign up until the next open enrollment period, which I think is every six months. You could be on the hook for quite a bit of medical expense (or denial of care) as a result.  This should be a powerful motivation. 

The second motivation is that in most cases, people want insurance. Sure, there are some objectors, but by and large, people perceive healthcare insurance to be a desirable thing to have. So the penalty is an incentive to buy something you already want to have. Would you rather pay $1000 and get nothing, or pay some higher number and have the thing that you actually wanted? Of course this doesn't work absent the subsidies that make the insurance affordable for working families and the medicaid expansion. 

The Massachusetts experience is that about 2% of people choose to pay the penalty, either because they are gaming the system, conscientious objectors, or in a particular financial hardship. If that number is replicated nationally, it's probably not enough to create a serious adverse selection death spiral.


So this is great news and a great day. My quick reactions:

First of all, CNN & FOX. Once again proving that being first is a higher priority than being right. How embarrassing. At least their soon-to-be-fired producers will be able to get health care coverage.

Second, Roberts: I have to give him a little credit. I fully anticipated him to be a partisan hack and invalidate the law. Perhaps the burden of history weighed heavily on him, perhaps the delegitimization of the court influenced him, or perhaps the sheer radicalism of Scalia's dissent drove him to uphold the law. Regardless, he got it right. Make no mistake, though, he did what he could to advance his long-term agenda by slowly restricting the reach of the commerce clause.

While most of the focus centered on the the Heritage-Foundation-developed mandate (aka the greatest threat to liberty ever), it's important to note that the 4 conservatives wanted to invalidate the entire law, and there is far, far more than the mandate in Obamacare. 12 million Americans will get rebates from their insurers this year based on the ACA's insurance regulations. Rescissions of policies is now prohibited. In a couple of years, pre-existing conditions will be covered under the guaranteed issue provisions, and the moribund individual market will be resuscitated by the insurance exchanges. All of these huge reforms survived and will transform healthcare in a good way.

The mandate itself may work, or people may prefer to pay the "tax" penalty and go without insurance. We will see. If enough people opt out and insurers are experiencing serious adverse selection in a few years, perhaps the partisan rhetoric will have died down enough that Congress can tweak the incentives at bit. One can hope, anyway.

The Medicaid ruling was disappointing but not fatal. The gist is this: the ACA expanded eligibility for Medicaid all the way up to 133% of the poverty line. This is significant because in many states, the eligibility thresholds are very stingy, and in some if you're male and without dependents or a disability you are never Medicaid eligible no matter how little you make. So this is a very large expansion of coverage. It is still allowed under this ruling, but no longer can the Feds make it compulsory for the states.

So what does that mean? Not clear. Most states probably will implement the Medicaid expansion, which is for the moment fully funded by the feds. There are concerns that the feds may shift the costs back to the states in a few years, but that's not clear yet. I can see some GOP governors refusing to implement the expansion on these grounds (really just to be recalcitrant dicks; I'm looking at you Scott Walker), but there will be huge pressure on them from their medical community to accept what is essentially free money. If they do refuse, then it's going to leave a large portion of their lower classes without access to health insurance.

For the ER, I predict little will change. ER utilization has been going up nationwide for two decades, and that trend will not change, regardless of the fate of the ACA. When, in five years, ERs are busier and more overcrowded, I predict it will be held that this is a consequence of the "failure of Obamacare" and I'm going to call BS on that in advance unless it is shown that the rate of ER volume growth accelerated after 2014.  Which is possible, but I don't think is super likely. Many of the soon-to-be insured are already coming to the ERs as no-pay patients, and the only difference is that we will be reimbursed for those services that we are already providing.  Some of them may be diverted to PCPs' offices, though the limited capacity of the primary care network sadly ensures that will be a small number.  To the extent that ER volumes do increase, it's a failure on the part of the system to create enough primary care capacity, not the expansion of coverage.

It's worth noting that the ACA does contain a 10% boost to primary care providers' reimbursements. So that may help improve access to primary care, and there are also significant expansions of Community Health Centers.

Cost control is where ACA is weakest, and more will need to be done to bend the curve of health care inflation. Sure, the IPAB may have some effect, if it is ever implemented. Medicare is already putting into place other reforms such as value based purchasing and others. But this is certainly the point where the ACA is only a start. Wether Congress can get it together well enough to add onto it in a productive fashion is to be seen. 

The saddest element of this whole kerfuffle is that liberals and policy wonks are celebrating the survival, by the thinnest of margins, of reforms which in the best-case scenarios will leave the US with the worst access to health care and health insurance in the OECD, with the highest cost per capita in the developed world, and with the worst outcomes in the industrialized countries. The passage and survival of the ACA are big wins, but they still leave the US with the worst health care in the world, and one party is hell-bent on dragging us backwards. So I will celebrate the win and spike the football and all that fun stuff, but tomorrow morning we've got to get up and keep working to reform our system further. Because what we have is not good enough.

19 June 2012

Let us give thanks

I began my shift with a small prayer:

Let us give thanks for Alcohol, Anxiety, and Gravity, without which we would be out of business...

Adds my partner, who happened to overhear me:

"Don't forget methamphetamines and impulsive decision-making."

Obamacare and EMTALA

From today's LA Times:

ACA Strikedown would complicate who pays for ER care for the uninsured:

"Whether people know it or not, whether people appreciate it or not, access to emergency care became a right in this country in 1986," said Dr. Wesley Fields, an emergency room physician in Orange County. "But the law that did that never addressed the big question of whose responsibility it was to deal with the cost." 
That unresolved question — who pays? — helped shape President Obama's 2010 healthcare law and its requirement that Americans get health insurance. For years, it even convinced many Republicans, including former Massachusetts Gov. Mitt Romney, to champion an insurance mandate. But today, the insurance mandate is the central target of GOP opposition to the law. 
Within days, the Supreme Court will rule on whether the new law is constitutional. If the law is upheld, millions of newly insured patients will have many of their hospital bills covered by insurance. But if the law, or just the insurance mandate, is struck down, those bills will be passed on to taxpayers, hospitals and privately insured patients, as they have been for the last quarter century.
As they have for the last quarter century.

The whole thing is well worth the read, and nothing that I haven't been saying for years. The coda, however, is striking for its understated demonstration of the cognitive dissonance displayed by the opponents of the mandate:

In the past, the cost shifting was cited by many conservatives as a reason why the federal government should require Americans to have health insurance. 
"If a man is struck down by a heart attack in the street, Americans will care for him whether or not he has insurance," the Heritage Foundation's Stuart Butler said in 1989. "We will not deny him services — even if that means more prudent citizens end up paying the tab." 
Butler, like many Republicans, has since renounced the insurance mandate.
They don't care about healthcare. They just care about scoring political points.  And now we wait to see if that neutral arbiter, the Roberts Court, the umpire whose only job is to call balls and strikes, will rewrite 75 years of constitutional law to deny his political opponents a victory, thereby shifting the cost of care for the uninsured back onto healthcare providers.

12 June 2012

Go Live

Whew. Mind = blown.

Our entire hospital booted up a new Electronic Medical Record (EMR), from top to bottom, we are now an Epic facility. Today was my first shift in the ER after go-live, which was Saturday. Holy smokes, what a project it was to get it up and running. This system now runs everything in the hospital, from the ER to the OR to the wards, to the business and billing function, stocking, housekeeping, nursing, RT/PT/OT, social work — EVERYTHING. And we went live with a "big bang," all at once.


The good news? It went, if not perfectly, very well, and certainly better than expected. There were no major issues, which was a huge relief since the programming team was frantically building critical elements until the day before go-live. It's really disconcerting to sit with tutors three days before the event, ask how to order labs, and be told, "well, this is how you'll do it, but you can't do it now since that module is still in development." But to their credit, they got it done and it works.

The bad news? Not too much, other than the fact that the system is massive and really, really complex. This makes the learning curve super steep, and the impact on operations during the first week has been substantial. We have lots of support, with tutors and specialists standing by our elbows guiding us through each workflow, but every simple little thing takes forever as you're learning it.

How complex is this system? Just my interface, and I am but one provider of many classes, has by my count at least 15 different screens I interact with, and each screen has dozens of widgets and elements I need to operate. Worse, the behavior of each widget isn't always consistent from one context to the next. There are multiple ways to get some common tasks done, which is nice, but it's so easy to get lost in all the menus, windows and panes. It's pretty overwhelming, and mistakes can be frustrating to undo.

And I'm an eager adopter, a computer savvy guy. For me to be completely on overload, I pity some of the less nerdy folks working in our hospital.

I don't want you to think I'm down on this system. I loved our old EMR, Picis, because it was super elegant and simple and easy to get stuff done. This is much more intricate, which is a big challenge to learn but — I think — will be more powerful once I get it mastered. I can see myself being much more efficient than I was before within a couple of months. I hope.

For those who are interested, we have the ASAP module with the Notewriter function, but my off-the-cuff reaction is that Notewriter is utter crap and I don't think I'll ever use it. I've constructed a H&P skeleton with a lot of datapoints auto-populated from the chart and I am using Dragon dictation. The current release of Dragon seems much more accurate than older ones I have used, and there were Dragon experts there giving us lots of tips & tricks to really take advantage of the shortcuts available.

For example, I can order meds & labs verbally, and even common lab panels, using the mic. I can also drop in a standard age/gender/complaint specific physical exam with three words. (IMPORTANT: proofread/edit the output to make sure it's appropriate for the actual patient!) Also, I've made a slew of medical decision making notes with links that pull in personal/clinical data from the chart for common situations.

So it's a powerful tool, and I may wonder how I ever got by without it in a year or two. But for now, my head is spinning and I've gotta go lie down.

[EDIT: The de-identified screenshot was provided by the nice folks at Epic.]