Monday, June 1, 2015

The Doctor's Dilemma

At the start of every shift I go through my e-mails. I find it a nice way to ease into my workday, and it's sort of fun to count the minutes you spend reading e-mails, multiply it by the amount you're getting paid per minute (and make no mistake, every ER doc knows what that number is), and realize you've just made money by clicking on announcements of new hires you'll never encounter, seminars you'll never attend, and the administrative glad-handing that passes for the aura of employee relations.  I could tell you that I'm looking intently for updates on clinical policies and care pathways that will lead to improved patient outcomes, or opportunities to fulfill  my educational needs, but the truth is I'm looking for Nastygrams.  Those are the e-mails that start out with some variation of "Remember that patient?" Any written communication that starts this way is by definition bad.  In the ER world, good news is handed out in person, while bad news is always in writing (to cover everyone involved except the proposed offender).
(You can also use the “I know how much I’m making per minute” theory when you’re doing personal tasks like going to the bathroom.  I like to think about it when I’m fetching the patient some water or juice.  It takes me two minutes to walk across the ER, open the refrigerator, get the juice, fill a cup from the ice machine, find a straw and a lid, and take it back to the patient, drag a tray over to the bedside, open and pour…yep.  Just bought lunch.)
The Nastygrams come in several varieties.  The primary one is the Patient Complaint.  The patient disliked something about what you did or didn’t do, and the Patient Advocate (now there’s a neutral job title) and other Administrative Poobahs would like you to review the chart before they confirm your anticipated guilt.  If that sounds alarmist, it needs to be understood that one of the current paradigms in health care is that the customer (patient) is always right, no matter how that plays out against the backdrop of clinical care.  It’s all about the satisfaction scores, about seeing the patient quickly, about making sure they’re happy during their stay and leaving with a positive impression so the patient, family, and friends will continue to seek care, and bring their dollars, to this hospital.  Which is all well and good, and something to be pursued, until you recognize that a lot of medicine doesn’t involve making people happy.  It’s about doing the right thing and the right time within the limits of your abilities and resources, and as often as not it means doing what might not be considered optimal customer service.  And if the scores come in low, no matter what the reason, it’s the physician who’s on the hook because otherwise it’s a problem with the system, and that might implicate the folks in suits.
Once you review the record, the complaint is usually not a surprise; and if you've done your documentation correctly (including direct quotes from the patient of phrases that rhyme with "brother ducker," "whole mitt," and "brass mole"), the Cubicle Dwellers have no choice but to reluctantly send a letter to the patient apologizing for the hard feelings, but regretfully concluding the care was appropriate. (The line, "It was documented that you acted like an orifice, and our staff was exactly right to send you on your happy way," is apparently not part of the template.)
(Speaking of which, my colleagues and I recently got an e-mail from one of our Medical Directors complimenting our group on the fact that there had been no patient complaints in the  prior two weeks.  I sent back a note and explained that I was on vacation.)
The “customer service” mantra leads to a strange and costly practice of medicine, where tests are ordered simply to be doing something or to meet patient  expectations, drugs are dispensed to make patients happy or simply because the doctor isn’t up for the fight (I think every ER doc has experienced their own version of “Give’em What They Want Wednesday,”), and the time needed to fully understand what’s going on with a patient, to set realistic expectations for the ER visit, and to provide good explanations of diagnoses or discharge instructions is truncated by the need to deal with those complaining more loudly and threatening to leave.  (I’ve never quite understood why, when someone wants to leave the ER because they’re not being seen fast enough or they don’t like the care they’ve received, we need to fall all over ourselves to get them to stay.  They are self-triaging themselves back to the community despite our legally-obligated offer of care.  Isn’t that what we want people to do…to take control of their own health care?  Yes, unless their departure means less revenue.)
If clinical correlations mean anything…as they rarely do in the world of customer service…there are studies that suggest that more “satisfied” patients actually get worse health care and experience worse outcomes.  And what’s more interesting is that those very things you can do to enhance customer satisfaction…like taking more time with patients to provide teaching and explanations, making sure family and friends are involved and informed, and going through discharge instructions carefully and completely…work against the goal of enhanced patient turnaround times. (More on that later.)
Which leads us to the Nastygram by the Numbers.  I truly do think it's a welcome development that an increasing amount of medical care is judged by objective criteria.  For example, if we know that patients should get certain medications after a heart attack, and that old people should get influenza and pneumonia vaccines, it makes sense to keep count of how many actually get the recommended care.  These are objective, specific, and measurable criteria proven to show a benefit to patient care.  The problem is that many other criteria we're judged on, especially in the ER, are totally abstract numerical goals without any clinical basis or demonstrated outcomes.  Many of them are promulgated by what I think of as the "Center" industry, which are those organizations set up to certify institutions as a Stroke Center, a Chest Pain Center, and others of that ilk for a healthy fee.  Which ultimately is not a clinical designation, but a marketing one.  (Ever notice how no hospital markets itself as a "Methamphetamine Center?"  It's because the bean counters don't want those folks...perhaps better termed as "non-revenue clients"...darkening the door of their ER.  Strokes and heart attacks mostly involve old people with Medicare, which means payment. Meth intoxication?  Not so much.)
Despite the fact that these criteria have no basis in reality, when one of them is not met the Nastygram follows.  For example, many have heard of "clotbuster" treatment for strokes.  The theory goes that if strokes are caused by small blood clots that limit blood flow to the brain, these medications get into the vessels and blast these clots away.  The catch is that you have to give these medications under four hours or so from the onset of the symptoms of stroke; beyond that time, the risks of life-threatening bleeding from use of the drug outweigh any benefit you might see  The criteria is that once the patient arrives in the ER, you have an hour from the moment the patient walks through the door to give the medication.  If it's given after that first hour, or there's no reason on the chart why you didn't give the medication (not everyone qualifies and it's not universally indicated), you "fall out" and your e-mail inbox fills.
Why one hour?  Nobody knows.  It's a totally abstract number.  Nobody has ever been able to quantify how much brain tissue is lost between 59 and 61 minutes.  And while I would surely agree that quicker treatment times are generally better, sometimes you need to take the time to discuss with the patient and the family the risks and benefits of treatment, and the speed of their decision process often does not match yours. They may want to talk to additional family members, or you may simply have problems finding anyone to talk to.  Maybe you need some lab work to make sure the patient actually qualifies for the drugs, or simply time to get the story straight so you know what you're doing.  And it's worth noting that in a recent study of Certified Stroke Centers, the majority were unable to meet their own select criteria for administration of the agent in less than an hour.  Rather than citing unproven numbers with the force of law, wouldn't it make more sense to track your own facility's times, come up with a goal, and look for roadblocks in the process rather than to come down on individuals for falling out on criteria that clinically means nothing and can't even be met by the best of hospitals?
(And this is even without addressing the very real clinical argument that the drugs may not be very useful overall, and should not even be considered the standard of care.  Once again the “Center’ industry…no doubt supported by the pharmaceutical folks who make these drugs…has promulgated a standard that may not even reflect the best practice in patient care.  This is similar to what happened with the standards for cardiac resuscitation over a decade ago, where a perfectly useful and inexpensive drug called lidocaine used for irregular heartbeats was magically replaced by a much more expensive and less useful agent called amiodarone, and nobody could figure out if there was a connection between the maker’s kind and generous support of the American Heart Association and its’ Clinical Investigators, and the change in standards.  But I suppose this is all cyclical and really nothing new.  At one time I understand it was popular to be a “Leech Center” as well.)
The most recent “quality benchmark” has been proposed by the Department of Health and Human Services.  HHS has set a goal of getting patients who will be discharged home in and out of the ER in a total time of two hours.  They’re even talking about it being used as a reimbursement factor.  Let’s leave aside for the moment that this is a totally abstract number, and that even the fastest doctor in our ER group did an experiment one month and couldn’t get her patients out in less than 120 minutes no matter how hard she worked.  Given unlimited resources of staffing, scores of empty patient rooms, and idle x-ray and laboratory machines just begging for something to do, you could probably meet the goal.  Nobody has that, and the two hour turnaround for all discharged patients often becomes a totally unobtainable measure in any busy ER that can’t be met without tricks of accounting (and there are ways to do that, some legitimate, some not so much).
But the practical issues beg the question of whether meeting this goal represents a measure of quality, at least as defined in medicine.  Fast turnaround times are possible, as long as the patient has a straightforward outpatient problem, sufficient ancillary resources are available to permit rapid lab and x-ray evaluation, and the physicians and nurses are not otherwise burdened with an  excessive number of patients under care or even a single critical patient which can take them ‘out of the system” for a prolonged period of time.  But that’s most often not the case.  As an emergency physician, the nature of the work means I’m not fully dedicated to a single patient at a time, and I can’t focus on moving that one patient through as quickly as I can.  I’ve got up to ten or more patients to juggle at a time, all with different issues, different levels of severity, and different agendas and expectations that need to be resolved.  Statistically speaking, the conflict is really one of modeling ER flow.  Throughput goals are based on a linear model of patient arrival, departure, provider workload, and client movement through the department.  Patients being the subjective and unpredictable creatures that they are, those with “boots on the ground” recognize that the ER actually works on a non-linear model.  Patient show up when they choose or need to, have varied complaints, require different levels of clinical and social interventions.  Workload for physicians and nurses follows a non-linear pathway as well.  So turnaround times mean something, or nothing, depending on who you ask; and as a result it’s hard to have a coherent dialogue with between two mutually exclusive perspectives.  And while shorter turnaround times no doubt promote customer satisfaction, it’s an open question whether that is equivalent to quality care, which brings us back to where we started.
Then there’s the Nursing Complaint.  As the power of Nursing Administration (who represent the majority of "critical" employees) within the hospital structure grows, the relative power of the physician ebbs.  It's fine when the paternalistic (at best) or authoritarian (at worst) physician-nurse relationships of the past become collaborative in nature.  But the pendulum has swung so far that the relationship is often adversarial, with the doctor considered guilty until proven otherwise.  Nurses are considered rare prizes and hard to get, and hospitals commit to specific staffing ratios and promotional schemes to keep and retain them; physicians are thought of as interchangeable commodities, especially in this era when most ER docs work not for themselves, but either as hospital employees or for an outside staffing group whose bottom line is profit margin and keeping administration happy.  So who becomes your more valuable asset when conflict occurs?  Especially in the ER, where physicians provide no referrals to the hospital, do not expand the financial footprint of the facility, and don't generate outside revenue for the institution?
(I'll be the first to admit I have little to no internal filter at work.  But I always watch myself when I talk to nurses, especially if I disagree with their assessment or they've made a blatant error in care, because I know who's going to win that argument.  It's not me.)
Don't misunderstand me.  It's not that doctors don't act like jerks from time to time. They do, myself included, and some of us more often than others.  And we've all done things in a moment of confusion of exhaustion that good nurses have caught and helped us correct. When these behaviors become routine, we should probably get called out on it, and in practice we do.  But since nurses are "owned" by administration in a way that doctors aren't, they're routinely protected despite similar behaviors or a frank lack of knowledge. 
(Want to know something else that really annoys me? Anonymous complaints.  We're professionals.  I'll sign my name to my concerns if you'll sign yours as well.  Otherwise, we both lose the opportunity to confront each other in search of the truth, and to find accommodation between our views.  Or better yet, just ask what I'm doing and why I’m doing it in real time.  I really don't mind questions, and an ensuing discussion if you've got a different point of view. But let's act like adults here, okay?)
Finally, we come to the Billing Questions, the subject of the most recent Nastygram to collect dust on my electronic desk.  It concerned a lab test for hepatitis, and I was asked by the billing office why I ordered the test.  The way the billing process works is that anything ordered needs to link up to the discharge diagnosis according to established criteria (the “criteria” industry is yet another moneymaker.)  If the test does not relate in a predetermined way to the discharge diagnosis, no payment for that test is forthcoming.
I went back to the chart to find out what occurred.  The patient presented with a probable cellulitis, or skin infection.  He had also come in with instructions from his doctor to be tested for HIV, the virus that causes AIDS.  He wanted this test done, and clinically when I test for HIV I also test for hepatitis, which is also transmitted through blood and body fluids (and which you are much more likely to contract from exposure than HIV).  They run together in high-risk individuals, so it always made sense to me think that if you’re at risk for one, you’re likely at risk for the other. And wouldn’t you like to know in advance rather than be surprised one day when you wake up looking like Chiquita Banana? Isn’t that good medical care?
But according to the Billing Lords, with a diagnosis of cellulitis the test not indicted.  So here’s the dilemma.  We’re told that medicine is now all about customer service, and I’m doing what both customers (the referring physician and the patient) want done.  I’m probably helping to secure a referral base by honoring that doctor’s request, and hopefully the patient will provide a glowing report back to him or her.  But in the effort to enhance customer service and provide quality care I've run up a charge that can’t be billed, and in getting these extra tests I’ve probably increased my throughput times.  I’m caught either way I turn.  And none of these issues which come to haunt me in the Nastygrams involve the actual provision of care for the patient’s skin infection, which is to prescribe some antibiotics, advise the patient to elevate and apply warm compresses to the leg, draw a marker line around the area of redness to make sure it doesn’t get any worse, and wish the patient the very best of the afternoon. 
In reading back through what I’ve written, I seem like nothing more than an angry old coot.  There’s truth in that.  I am highly frustrated with the current practice of medicine for any number of reasons, but for the record I don’t object to being judged in my work.  I simply want the standards to be fair and to mean something to clinical practice and patient care, which should be my bottom line.  And I want the playing field to be level, to know that my word means something, and to know that the institution for whom I work is actually on my side.  I’ve been in places where the hospital declared open war on physicians; fortunately, that’s not the case where I practice now, and I hope the tide doesn’t turn until I’m ready to hang it up for good.  But that spectre is always out there, and all it takes is one new administrator in just the wrong place to tip the balance…as it inevitably will.
Thanks for reading.  I’m sorry it’s taken so long today, but I’ve been busy writing other paragraphs.  I really appreciate your patience and understanding.  Please be sure to fill out your customer satisfaction survey.  Please make sure you rate me as a “9” or “10,” because anything else is considered a failure.  If you like me, the name’s Dr. Rodenberg.  If you don’t, it’s Dr. Smith.  Because we don’t have any of those here in our ER.