Starting With Curious

When even a child of the 60s and 70s like me is connected via texting, Twitter, Facebook, etc., it’s safe to say that the digital revolution is over, and digital won.  The last holdout, it seems – the analog equivalent of the tsarist White Army – is health care.  Sure, we have robots for surgery, telemedicine for remote diagnosis, and an increasing use of electronic health records by hospitals and providers, but it still feels very old school.  Sort of like education: even if you take notes on an iPad, attending a lecture is still a very traditional method of learning.  Health care is still very provider- and facility-centric.  To paraphrase the NY Times columnist Thomas Friedman, the world of health care has still not gotten very flat.  Yet there are some steps toward leveraging new modes of communication that have the potential to disrupt the model of health care delivery.

Many people today use texting as their primary means of communication, but it is seldom used in medicine.  Expanding its use would not only be more patient-centered, but may really improve efficiency.  We recently ran a trial of texting appointment reminders to patients in our sickle cell clinic, which has a traditionally high no-show rate.  The day after text reminders were sent, only 1 of 14 patients did not come for their appointment.  Anyone who has ever dealt with a teen knows that this is a group where texting may provide a particular opportunity to vastly improve provider-patient communication.  Providers and staff appear to be open to the idea of greater electronic communication, but some barriers are apparent.  One is reimbursement: under current payment systems, electronic communication is a service for which it is difficult to collect a fee.  Another concern is confidentiality, though some progress has been made to ensure that texts or other electronic messages are secure, such as a HIPAA-compliant messaging app.

Texting of course, while seemingly radical by the conservative standards of health care, is not exactly a disruptive technology.  But some providers are pushing the envelope with “virtual clinics,” where almost all interaction is electronic.  Check out, for example, these examples in New York and Minnesota.  Experience is limited, especially regarding outcomes, but these experiments suggest that at least two elements of the triple aim – lower cost and better experience – can be met.

Medicine is clearly moving toward being a true team sport.  Many of the proposed innovations in care delivery and payment depend on a collaborative approach, with individuals from multiple health professions playing a role.  Perhaps the clearest example is the patient-centered medical home.  Which all sounds great, until you start to ask different people what they mean by a “team.”  Many physicians have a, well, physician-centric concept of a team, where the doctor serves as “quarterback” or “captain.”  And while teams generally need leadership, you can imagine that a nurse, social worker, or pharmacist might have a less enthusiastic embrace of a team where she or he can, by definition, never be the leader.  It’s not that physicians are trying to be uncollaborative.  However, medicine has traditionally been structured hierarchically (e.g., doctors write “orders,” not “requests”).   And this is reinforced throughout the training not only of doctors, but other health professionals as well.

To move toward a truly comprehensive, collaborative team approach, this approach needs to be embedded into the education of all potential team members, right from the beginning.  By the time medical and nursing students have completed their four years of school, patterns of thought and behavior – cultural constructs – are already developed.  This is why there has been an increased interest in interprofessional education, defined by the WHO as occurring when “students from two or more professional learn about, from, and with each other to enable effective collaboration and improve health outcomes.”   MCW’s Community Medical Education Program is exploring the possibility of incorporating interprofessional education into the curricula at the regional campuses, in conjunction with the other health professional schools in those communities.

Dr. Melanie Dreher, Dean of the Rush University School of Nursing, recently gave a presentation sponsored by the CMEP on interprofessional education.  She noted some of the dominant cultural constructs about medicine as a hierarchy that need to be revised, and the potential for interprofessional education to do so.  I’ll call out two.  The first is what I mentioned above, the idea of a team revolving around a single leader (typically a physician).  She offered the counterexample of situational leadership, where leadership of the team is flexible depending on the issue being addressed.  For example, if the issue is one of polypharmacy, then a pharmacist might assume the role of leader; if it is one of support services, perhaps the social worker.  An analog is the Orpheus Chamber Orchestra, a group in which the role of conductor is shared and rotated among all members, depending on the piece to be performed.  As they see it, it is not the conductor who is the center, nor even the musicians – it is the audience.  The other cultural construct is that of how we measure contributions to the team.  It can’t be by RVUs generated, or papers published, or referrals received, which ignores equally important roles played by many of the members.  Dr. Dreher offered the basketball example of Shane Battier of the Miami Heat, whom Michael Lewis referred to as the “no stat All-Star.”  He is considered one of the best defensive players in the league. But he has few points, rebounds, or assists –the conventional measures of success – and is therefore undervalued compared with high scorers, although statistics suggest that there is a stronger correlation between Battier’s playing time and the team’s record.

A true team approach means we need to fully utilize and recognize the unique knowledge, talents, and approaches provided by every member of that team.  Engendering that attitude, and teaching the actual skills of teamwork (which are separate from the skills inherent in being a doctor, nurse, or pharmacist) means we need to break down the silos in which health professionals learn.  It’s not a terribly new concept: when I was in medical school, the physician assistant students and medical students took several of their preclinical classes, such as anatomy, together.  It fostered a sense of collegiality and mutual understanding.  But is has not become widespread.  As we evolve toward a more collaborative model of health care, the way we educate health professionals must adapt as well.

I introduced this analogy at a Pediatric Grand Rounds on Population Health Management and Payment Innovation, as a way to conceptualize the changing role for providers in the evolving health care environment.  My grandfather was a cab driver in New York City in the 1940s and 50s.  He and most other drivers at the time were owner-operators.  Although there were some regulations around fares, etc., he was more or less on his own – he paid for his gas, insurance, upkeep, etc., and whatever he had left after paying expenses he kept.  What regulation there was, was enforced by the requirement for a medallion (of which there were a limited supply) in order to operate legally.  Within the constraints of government regulations, though, they set their own practice.  They could pick the neighborhoods they operated in, the hours they worked, the routes they drove.

How is this relevant to healthcare?  The medallion was the equivalent of a medical license (of which there are also, in practice, a limited supply), and he was in essence a solo practitioner.  Sometimes cabs would come together into a cooperative of several owner-operators, sort of like a group practice, but still controlled by the drivers themselves.

As the costs of obtaining a medallion and operating the cab (e.g., fuel prices) go up, it gets harder for an individual driver to make a living as an owner-operator.  Moreover, the system itself is inefficient and wasteful.  Lots of drivers want to be in the nicer neighborhoods where the tips are better; fewer want to drive long routes out of town with no way to recoup the time for the return trip.  Cabs circle around waiting for people who need a ride, wasting gas.  There is no way to efficiently match the size of the vehicle with the number of people in a party: a group of 6 might have to wait some time for a large enough car to come by.  So a system of leased cabs, managed by a cab company, evolved.  The company had a phone line, so people in need of a ride can call.  A dispatcher sends the closest, most appropriate size vehicle.  When the rider pays her fare, it actually goes to the cab company, not to the driver.  The driver gets a fee (he’s essentially an employee), the company pays the expenses (including not only the vehicle operating costs but also the salaries for the dispatchers), and keeps whatever is left as profit.  The company, not the individual driver, sets the guidelines about where and when the cab goes.  And a company that does a good job of minimizing costs benefits financially.  They will do that by paying as little as possible for a unit of service (cheaper insurance, less expensive gas, decreasing payments to the drivers), and by eliminating waste (e.g., not having drivers circling around, sending the right cab to the right place at the right time).

This is where the health care system is heading.  We are moving away from managing the medical care of an individual to managing the health of a population.  As specialists, we have been like the owner-operator cab drivers of my grandfather’s era.  We are used to focusing on providing a service to an individual patient and getting paid for that service, according to our own preferences and individual guidelines.  In the future, providers (and I use that term to include not only physicians but also hospitals) will be like the drivers in a cab company.  Payment will not go directly to the provider, it will go to the health manager (e.g., an ACO).  Like the cab company, this manager will use “dispatchers” to direct the right care to the right place at the right time.  Payment to providers will not be the usual fee-for-service: it may be a salary, or some modified payment for services but with shared risk or shared savings.  And of course, the health manager will attempt to minimize costs by pressuring payments to providers and eliminating unnecessary utilization.

Sounds bleak?  What provider in her right mind wants to be a cab driver working for a large company, being told what to do by a dispatcher?!  The hope lies in shifting our thinking from being an individual provider to being part of a system.  I, myself, might just drive the cab, but I have to be a fully-participating member of the company, with a voice and a stake in figuring out the dispatching protocols, minimizing variation and waste, and sharing in the profit (and risk) of doing a good (or bad) job.  In an era when health spending needs to come down from unsustainable levels, those who are only providers can only lose, in the form of lower reimbursement for fewer services.  But those who also participate in the function of managing the fleet can reap the benefits and thrive.  That doesn’t mean that someone who is now an allergist or a surgeon will stop doing what he or she does.  A cab company doesn’t survive without the cabs and great drivers.  It does mean that we need to figure out how to operate in the larger system, how to work with the managers and payers and primary care providers (and most of all, the riders) to reduce costs, provide a better service, and improve health.

One more note.  As first noted in a 1960 article by Theodore Levitt in the Harvard Business Review, railroads largely went bankrupt because they saw themselves as being in the railroad business, rather than in the transportation business.  As cars and buses flourished, demand for rail services went down.  We not only have to make the shift from being cab drivers to being a cab company, we need to be a transportation company.  Or, in our terms, we need to shift from being providers to being managers, and not just medical managers but healthcare (or better, health) managers.  When alternatives to hospitals spread, we’ll be better poised to bring our expertise to whatever those alternatives are.

“Don’t it always seem to go

You don’t know what you’ve got ‘til it’s gone…”

“Defensive medicine” has long been a favorite rationale for our health care costs.  Indeed, in a recent survey of physicians in JAMA, trial lawyers were assigned the greatest responsibility for reducing health care costs:  60% of respondents thought they had a major role, followed by insurers (59%), pharma and device manufacturers (56%), hospitals (56%), and patients (52%).  In contrast, only 36% of physicians thought physicians had a major role in cost control.

I’ve never been a huge fan of the defensive medicine hypothesis.  First, several studies have suggested that only a very small percentage (around 3%) of total health care spending can be attributed to malpractice.  Second, poor outcomes and allegations of malpractice can arise from errors of commission as well as of omission.  A fatal allergic reaction to an antibiotic prescribed unnecessarily for a viral illness could as easily open a provider up to a liability claim as not doing an unnecessary CT scan for a child with trivial head injury, so I would imagine someone fearful of a lawsuit would hesitate to do both.  Most important, in my own experience (admittedly skewed by working in academic centers), overuse appears to be driven far more strongly by factors such as the desire for more complete information before making decisions (more on that in a subsequent blog), and by financial incentives.

A recent study, though, provides what appears to be compelling evidence of the effect of fear of malpractice on use of diagnostic tests.  An article in the August issue of Health Affairs linked physicians’ responses regarding their levels of malpractice concern as reported in the 2008 Health Tracking Physician Survey to Medicare Parts A and B claims for the patients they treated during the study period, 2007–09, focusing on use of various diagnostic tests for complaints of chest pain, headache, or lower back pain.  The investigators found higher rates of imaging by physicians who had reported high vs. low concern about malpractice on the survey, although the magnitudes were modest.  This would suggest that defensive medicine may be a significant contributor to overall healthcare costs.  However, the study undercuts the assertion that moves to limit malpractice claims would help – there was no association between diagnostic testing and various measures of actual state malpractice risk, only with the physician’s reported subjective level of concern.

Which raises several questions.  Where does the fear arise from, if not from the actual climate of the state within which the physicians practice?  Is it truly a fear of liability, or is that just an expression of an underlying tendency to place a premium on diagnostic certainty?  Providers tend to fall into one of two camps – “test minimizers” and “risk minimizers” – based on their tolerance for uncertainty.  In either case, if individual provider attitudes override public policy, how can we address this factor in the effort to control costs?  At the very least, it suggests that the doctors have at least as much responsibility as the trial lawyers.

This song about burnout on the job was quite popular in 1977 (original version by Johnny Paycheck; subsequently also recorded by Dead Kennedys).  While many of you are not old enough to have been assaulted by the recording on AM radio, the sentiment probably isn’t at all foreign.  The phenomenon of burnout among medical professionals has been the subject of both serious research and discussion in the lay press.  A 2012 study in JAMA Internal Medicine revealed high levels of self-reported burnout among physicians, especially in “front-line” specialties such as family practice and emergency medicine, where over half of physicians reported some form of burnout.  (Fortunately, both primary care pediatrics and pediatric sub-specialties had below average rates.)  Also, physicians had higher rates of burnout than the general population.  (There are studies showing similar statistics for nurses, but I haven’t been able to find any studies specifically dealing with burnout among advanced practice providers.  I think we can assume it’s fairly similar.)

Burnout is defined as “a syndrome characterized by a loss of enthusiasm for work (emotional exhaustion), feelings of cynicism (depersonalization), and a low sense of personal accomplishment.”  Effects of burnout include symptoms of depression and/or anxiety, loss of empathy and objectivization of patients and co-workers, unprofessional behavior, and high rates of error.  It can also lead people to leave the profession.  Thus, burnout is a problem for the physicians, for their patients, and ultimately for the system.

While a good bit has been written about the prevalence of burnout, there seems to be little data on what can be done to prevent it.  It doesn’t appear to correlate with hours worked, income, or satisfaction with work-life balance, but data are limited.

In the meantime, try this two-item screening tool for identifying burnout:

How often do you agree with the statement “I feel burned out from my work”?

Never

Less than once a month

A few times a month

Once a week

A few times a week

Every day

How often do you agree with the statement, “I have become more callous toward people since I took this job”?

Never

Less than once a month

A few times a month

Once a week

A few times a week

Every day

If, after that, you need a little pick-me-up, watch this – a good reminder of our value of health.

The teen had been referred to the ED for evaluation of a mediastinal mass found on a chest X-ray; he and his mother both knew his primary physician was worried about cancer.  We performed a CT and a slew of lab tests.  The lab tests were pretty non-specific and not overly concerning.  As we reviewed the test results with them, the patient said, “Yeah, but I’m gonna die, aren’t I?”  At that particular moment, when we didn’t even have a diagnosis, much less a prognosis, I was more focused on whether he had laboratory signs of blood stream or organ involvement.  He couldn’t have cared less about a white count or liver enzymes – he wanted to know if he was going to live.

Providers often find themselves focusing on things that are at best of indirect interest to our patients.  Nobody cares what their hemoglobin A1c or HDL level is (except for a few hypercompetitive types who brag about their biometrics after their screening), except to the extent that they are indicators of what truly matters – mortality or morbidity.  People don’t want a healthy blood pressure, they want a healthy life.

Why have we not paid more attention to these patient-centered outcomes?  One thing is, we like to measure what we can affect.  I may be able to improve someone’s pulmonary function with an inhaled steroid; the link between the prescription and their overall quality of life is more tenuous.   It’s also more difficult to measure something as abstract and multifactorial as “quality of life.”  At the very least, tools for determining patient-centered outcomes can be more time consuming to use, and if not properly developed may be less valid and reliable than “hard” measures such as lab results.  There is a developing science around measurement of patient-oriented outcomes.  The Patient Centered Outcomes Research Institute (PCORI), created under the Affordable Care Act, serves as a promoter and funder of efforts to advance such measurements.  But, while there are a growing number of tools for measuring both disease-specific and generic quality of life across the age spectrum, such tools are not incorporated into routine clinical care.

There has been much discussion of “meaningful use” of the EHR.  But if you look at many of the measures, one can ask, meaningful to whom?  Documenting BMI, keeping INR in a certain range for patients on warfarin, prescribing steroids for patients with asthma – I doubt very much that our patients care about these in and of themselves.  If we want the EHR to be truly meaningful, we should think about incorporating patient-centered outcomes into the record.  This would promote discussion of these outcomes with patients and families, and allow us to determine how effective our care is in improving them.

It’s pretty clear that by looking only at conventional measures of effectiveness, we are missing a big part of the picture, the part that matters most to the people we are trying to help.  A conventional measure of success of ED treatment of acute asthma is discharge to home (with a prescription for systemic steroids), and no unscheduled return visits within a set period (usually a week).  By that measure, the ED where I used to work did a good job.  But when we looked at patients “successfully” discharged to home, we found that nearly half had either prolonged symptoms, substantial missed school or day care, or significant activity limitation.  While we were congratulating ourselves for keeping these children out of the hospital, they were sitting at home coughing and wheezing and being miserable.

If we are going to act in the service of children and families, we need to understand and measure what is important to them – to have a mutual understanding of what constitutes “success.”

I recently recounted for someone the story of how, during the oral exam for my surgery rotation in medical school, one of the senior attending physicians made me cry.  Literally.  He belittled and humiliated me, and what was worse, he seemed to be enjoying himself.  Now, I don’t think there are very many true sadists in the healthcare profession, but we have to admit that, in the grip of the intensity of what we do, we sometimes lack a certain amount of sensitivity and civility.  We are not always respectful.

Sometimes it’s directed toward patients.  Not necessarily as overt rudeness, but more subtly, such as keeping people waiting without apology, not introducing ourselves or addressing people by name, etc.  Other times it’s a lack of respect for our colleagues.  A recent study from the Journal of General Internal Medicine revealed the extent to which physicians criticize other physicians to patients, often with inappropriate language (e.g., “This guy’s an idiot!”).   The situation is no better when it comes to relationships among other members of the healthcare team.  While perhaps 5% of physicians can be described as chronically disruptive, disruptive behaviors are well documented among both physicians and nurses.

I confess to being guilty myself.  As time goes on, I find it harder to get back on task when I am interrupted.  It shouldn’t be a surprise that interruptions are very common in the emergency department.  But I find that when a nurse “interrupts” something I am doing because there is something that she interprets as more urgent, I sometimes – especially if my perception of the urgency is different – react in a manner that is curt or dismissive.  It makes me come across as difficult to approach, which has a negative impact on our ability to work as a team, and ultimately perhaps on the effectiveness and safety of our care.  (And needless to say, I am not always correct in that relative perception of urgency.)

Even little things, things we may not ourselves perceive as disrespectful, can damage the team relationship.  For example, in a recent survey, only 4% of physicians agreed with the statement that nurse practitioners should be paid the same as physicians when providing the same services.  Now, this is certainly a complex issue, but consider what it sounds like to a nurse practitioner to hear that she or he should be paid less for doing the same thing.

Fortunately, a little respect goes a long way.  When wait times in the ED are long, trying everyone’s patience, I’ve seen the simple act of apologizing and recognizing the inconvenience to the family melt the icy stare that greeted me when I walked in the room.  Thanking a nurse for alerting us to a monitor alarm, telling a physician assistant “nice job” when she picks up an exam finding we missed, talking candidly but with respect to a consultant with whom we may disagree on a management plan – these are all ways of living our value of integrity.

Two years ago, I was having trouble finding time to get over to employee health for my flu shot.  One day I went to Walgreens for something, and realized I could get it right then and there, so I did.  I admit, I felt a little tainted, like I was aiding and abetting the enemy.  But it was quick, cheap, and convenient – hard to argue with that for a simple flu shot.

A recent article in JAMA talks about the “convenience revolution,” the innovative options for treatment of low-acuity conditions, including store kiosks and retail clinics, employer-based clinics, “clinic in a car,” and e-visits.  These are examples of the kinds of “disruptive innovations” than Clayton Christensen talked about in “The Innovator’s Prescription.”  It’s easy to be dismissive of these as a “doc-in-the-box,” but we need to be mindful of why they have been growing.  For many conditions, especially in pediatrics, it’s frankly difficult to demonstrate much of a difference between providers and settings in the quality domain of effectiveness.  Let’s face it, most of the time kids with simple illnesses are going to get better no matter what.  So from a consumer perspective, if effectiveness is the same, then the other domains, especially efficiency (i.e., cost), timeliness, and patient-centeredness (i.e., service) are going to be more important.  There is no difference in how well protected I was against the flu by my Walgreen’s shot than by one administered by MCW Employee Health.  But I didn’t have to make an appointment or walk across campus to get it – I could do it when I happened to need to go to the drug store to buy razor blades anyway.

OK, perhaps a flu shot doesn’t matter.  But even something as simple as a sore throat – will the provider at Target perform a strep test correctly?  (Answer: Probably – it’s really not that hard.)  Will she be able to recognize the rare case of epiglottitis or diphtheria? (Answer: Is that a serious question?)  And what happens if everyone starts going to retail clinics for routine things, and only come to their pediatrician or Children’s for things that are more complex?  Or worse, those “convenience clinics” start dealing with some of the more complex things, where there may be a difference in effectiveness or safety.

The answer to those questions is, we can’t let them.  I don’t mean we lobby for restrictions to prevent it – that probably won’t work.  I mean we have to outcompete them.  We must acknowledge the importance of those other domains – service, cost, timeliness – and provide a similar level of convenience.  What kind of “disruptive innovation” can you think of in your area.

And by the way, last year, employee health came to me.

This week’s post will be short, as I am living out our value of Health, specifically, “I am an active participant in my own health and lead a healthy lifestyle.”  In other words, I’m on vacation, and I mean really on vacation, as in no cell phone and no email.  (I wrote this last week.)  I’ve written previously about the work showing a link between rest and rejuvenation and increased quality, productivity, and satisfaction with work.  As anyone who has tried to schedule a meeting over the summer knows, many people do take vacation.  But are you really gone?  Try it out.

While you’re checking out, check out this item about how giving is the key to getting.  Specifically, organizational psychologist Adam Grant, at the Wharton School of Business, has written a book describing how framing work as altruism can lead to improved productivity and satisfaction as well.  In his view, there are 3 kinds of people: takers (those who always seek advantage), matchers (people who give in expectation of a return in kind), and givers (those who give – especially of their time and attention –  without expectation of immediate gain).  While most people are matchers, givers are overrepresented at both extremes of success – they can be taken advantage of, but they are also among the highest achieving leaders.

We speak of the gift of time.  Try giving it to yourself and to someone else.  It could be the gift that keeps on giving.

In a recent post, I talked about the growing interest among our patients and families – driven by increased cost sharing – in understanding what care is costing them, and the difficulty in answering that question.   A recent study in Pediatrics illustrates the problem.  Residents and attending hospitalists at Children’s Hospital of Philadelphia were surveyed about their knowledge of costs, charges, and payments for a range of common tests and treatments.  The respondents were not confident in their knowledge: only 13% of attending and 3% of residents rated their understanding of hospital finances as at least moderately adequate.  The large majority of both groups expressed a desire for knowing more.

It turns out they need to.  The median percent error was 75% for attending and 80% for residents.  Only 15% of estimates were within 25% of the actual value, and the magnitude of errors were similar for costs, charges, and payments.  However, both groups tended to overestimate costs, while underestimating charges and payments.  Arguably, the latter two are what matter most to patients who will have to pay at least a portion of the bill.  In addition, there appeared to be a certain tendency toward the middle: overestimates were more common for the least expensive items, while the costs and charges for the most expensive items tended to be underestimated.

If our patients and families are becoming more price-sensitive, and are asking more about what services will cost them, we need to prepare providers to deal with that.  Moreover, we know that making clinicians aware of costs can reduce unnecessary “routine testing” and other resource utilization.  This study shows that we have a long way to go in terms of getting accurate information into the hands of providers.  One of the advantages of an electronic health record is the ability to embed that information into the ordering process.  Interestingly, the Pediatrics study was done at an institution using Epic; it’s not clear whether they include any price information to the ordering provider.

Transparency, anyone?