Safe or Out?  

April 17, 2014

CHW Logo
The only apparent sign of spring around here is the fact that baseball season has begun.  (And perhaps the Brewers’ 11-4 record is a sign of just what an unusual spring it has been.)  Now, I’m not a member of the “baseball as a metaphor for life” school of thought, but it does seem that baseball is in some ways a lot like medical care.  Both involve a large team of people with specialized jobs who spend most of their time in what I would term “anticipatory inaction” (i.e., standing around and waiting for something to happen).  And when they do go into action, the goal is to get someone to home.  (One could also add the unexplained ridiculous prices – a $9 Miller Lite is even less defensible than a $25 Tylenol.)

They also share the concept of being safe, but here they diverge.  In baseball, “safe” is a result – one branch of a dichotomous outcome, the opposite of “out.”  But in healthcare, “safe” is much more complex.  It does involve good outcomes: one measure of safety is the absence of harm caused by the care provided, such as wound infections or pressure ulcers.  However, while such outcomes are an important measure of safety, and are necessary, they are not sufficient.  No wound infections can result from safety or from luck.  Safety is as much about the structures and processes that create a safe environment.  A car without airbags or seatbelts is unsafe even if you drive it many miles without getting injured.

I was thinking about this recently as we at Children’s increase our commitment to providing the best and safest care.  We participate in something called Solutions for Patient Safety, a collaborative of almost 80 children’s hospitals around the country.  The hospitals share data and best practices to try to eliminate patient harm.  When I have seen the data, on most measures we perform better than most, so it would seem that we are a “safe” hospital.  But even if we have a rate of 0, that by itself does not mean we are safe.  I used to have this argument with my son: I’d catch him riding his bike without a helmet, and tell him that wasn’t safe; he would argue that he hadn’t gotten hurt.  To which I would respond, “Yet.”

So while I feel good about our record, we can’t get complacent.  That’s why we are embarking on an effort to increase education among all of the staff and providers on what it takes to be a high reliability organization – the kind of place where not only do harmful events not happen, but they can’t happen.  It means having systems in place that make it easy to do the right thing and hard to do the wrong one.  Redundancy, double checks, alerts, standardized approaches, checklists, etc.

It also means having an environment and culture of safety.  This is an environment where people continually seek to improve by asking questions, raising issues, and intervening when there is a problem.  There are many elements critical to a functioning culture of safety, one of which is what we call “just culture.”  This means treating errors, when they do occur, as a system flaw, not a character flaw.  If I order the wrong medication, or forget to give someone a tetanus shot, it doesn’t mean I’m a bad person.  In a just culture, people are more willing to acknowledge errors, and more willing to point them out.

Creating a just culture and an environment of safety are challenging in practice.  For one thing, it requires overcoming the rather natural tendency to blame on the one hand, and to be defensive on the other.  It’s one thing for an organization to say “we’re not blaming you,” but only after an extended period of acting that way will people feel confident in the attitude shift.  In hospitals in particular, there is a traditional hierarchy that must be overcome.  It doesn’t matter how many posters we put up asserting “You have our permission to speak up,” it is never going to be easy for a nurse six months out of school to tell the chief of surgery that he or she is about to use the wrong instrument.  And it’s even harder if that physician makes it difficult by their response.

In recent years, physician performance has been defined around six core competencies.  Among these are medical knowledge and patient care, which are somewhat easy to define and measure.  Another key one, professionalism, has been shown to be a key element in creating the kind of atmosphere that promotes safety.  Much has been written about disruptive behavior –up to 5% of providers are estimated to be regularly disruptive.  But professionalism, and unprofessional behavior,  is much harder to measure.  It can range from an attitude of dismissiveness to frank verbal or physical abuse.  To paraphrase Justice Potter Stewart, it may be that you know it when you see it.  But the more objectively we can define it, the easier it is to identify when there are problems, to create actionable solutions, and monitor progress.  It allows for fair processes to hold people accountable and ultimately improve, which is the goal.  As difficult as it is to live out a “just culture” when it comes to things like medication errors, it’s that much harder when the “error” is a behavior.  It’s one reason we have moved away from the older language of “problem physicians” to “disruptive behavior.”

Not long ago, I was approached by my medical director and patient care manager with a concern about how I might be making the emergency department unsafe.  They told me that a nurse had indicated, in a survey, that “Dr. Gorelick is difficult to work with.  He is hard to approach with questions and dismissive of the nurses.”  I was absolutely floored.  I don’t consider myself to be a difficult person, and I pride myself on my respect for nurses and indeed all the members of the care team.  (My mother is a nurse, after al!)  But while I heard “Dr. Gorelick is a difficult person who disrespects nurses,” what was actually said was quite different.  It was about my actions, and a specific set of actions at that.  It caused me to reflect, and I realized it was true.  As I’ve written about before, I find it increasingly difficult for whatever reason (I’m sure it’s not aging…)  to get back on task when I’m interrupted.  So if I’m entering an order or writing a note and someone stops me to ask me a question, I do sometimes get snappy.  I worry that it will at least set me back in the task I was doing, and may even increase the risk of an ordering error.  In the parlance of our Mood Elevator, I go to the lower floors of judgmental and irritated.  But by being hard to approach, I was creating my own safety hazard.  The next time a child needed immediate attention, nurses might hesitate to seek me out, putting that child at risk.

As Warren Buffett said, “It takes 20 years to build a reputation, and five minutes to ruin one.”  My reputation was also at risk, and I needed to fix it.  The first step was acknowledging the problem, and owning it.  The problem was not the nurses’ inability to prioritize, it was my reaction to being interrupted.  Rather than being irritable, I needed to be curious – why does this person want my attention now?  If it turns out to be something that could wait, let’s have a respectful discussion about the risks of unnecessary interruptions, which could lead to a constructive solution (e.g., some way to let me know I’m needed soon but not immediately).

This sounds easier than it is.  I’m sure I still seem less than approachable at times in the emergency department.  As much as I’ve tried to objectify this and focus on my behaviors, it still feels like a questioning of my character.  And no doubt people’s perceptions are colored by my prior actions: even a slight hesitation in responding could be perceived as being “difficult” again.  But if I want to restore my reputation – and more importantly, contribute to the kind of environment that allows us to provide the best and safest care – I have to be accountable for my role.  Professional behavior is like an RBI, allowing our patient to get home safe.  Sometimes we have to attend batting practice.

Medical Homelessness  

April 11, 2014

CHW Logo
With millions of people newly insured thanks to the Affordable Care Act, a lot of people will be seeking medical care for the first time in a while.  While many of these will be people with problems needing specialty care due to problems that have gone untreated, many others will be in search of primary and preventive care (most of which is not subject to cost sharing).  For several years there has been talk of a shortage of primary care physicians – seems that will only become more acute.

There has also been a lot of talk about increasing the supply of physicians selecting primary care specialties.  The pendulum swings back and forth.  But, as nicely summarized by Pauline Chen,  there has been no major, sustained trend toward more physicians going in to primary care, despite a variety of initiatives such as medical school expansions, dedicated primary care training tracks, loan repayment programs.  With many subspecialists enjoying shorter hours and much higher salaries – the recent release of data on Medicare payments to physicians was particularly eye-opening – simply creating more of the same doctors and asking them to pretty please go into primary care is simply not going to work.  We need to, as Steve Jobs said, “Think Different.”

1.  Expand the role of non-physician providers.  New York recently became the latest state to expand the ability of nurse practitioners to provide care independently by removing the requirement for a written practice agreement with a physician.  Nurse practitioners are already beginning to fill important holes in primary care in a number of underserved communities – including two inner-city clinics resulting from a partnership between Children’s Hospital of Wisconsin and Marquette University School of Nursing.  This model is very likely to be an important part of meeting the national primary care gap.  We need to commit to increasing the number of NPs, and to removing the unnecessary barriers to their ability to practice to the full extent of their training.

2.  Make primary care providers more productive.  Fields like education and medicine, being relatively reliant on the human element, typically have much smaller increases in productivity than industries like manufacturing that are amenable to automation.  That appears to be changing, with numerous actual or promised innovations such as remote sensors and other forms of telemedicine, better care management, and newer therapies that require, frankly, less skill.  By simply embracing innovations that promote efficiency – not to mention a shift in focus from medical care to maintaining wellness –  we should be able to get by with relatively fewer providers.

3.  Rethink how we train physicians.  The current model of medical education – four years of college, then four years of medical school with two full years of basic sciences, followed by a minimum of three years of residency – is neither long-standing nor the international norm.  In most European countries, university and medical training are combined over a period of six years, and tuition is often free or far lower than in the US.  In the 1970s and 80s, a number of medical schools in the US experimented with alternative pathways, including 6 year combined undergraduate and medical degrees, or 3 year medical school.

The 3 year alternative is attracting new attention.  One important motivator is to allow students to graduate with lower debt, thus decreasing the barrier to entry into lower-paying primary care fields.  At the Medical College of Wisconsin, a Community Based Medical Education Program is being implemented, with satellite campuses in rural parts of the state and a 3-year curriculum.  The idea is that by attracting students with an interest in primary care in a rural setting, allowing them to decrease their tuition burden by 20-25%, and providing residency training programs in those communities, it is much more likely that they will choose to practice primary care in those areas.

I certainly hope this works.  But I think we need to use the opportunity to re-examine what is required for medical education.  In designing the community-based program, there has been a lot of thought to how to provide the basic science education (e.g., remote learning, offering some classes on local campuses, etc.).   I think the real question is how much basic science do you need, especially for a clinical (as opposed to research) career.  The opportunity to do less microanatomy and biochemistry, with earlier clinical exposure, may be even more appealing to those with an interest in primary care.

This borders on heresy in the world of medical education.  Without thousands of hours of basic science instruction, doctors will simply be “technicians,” no different than nurse practitioners!  Hmmm…..

Drumroll, Please

April 4, 2014

CHW Logo
The March 31 deadline for open enrollment in plans offered on the healthcare exchanges has now passed.  The original goal was 7 million enrollees.  And the actual number was … 7.1 million.  Whew!

Some of these are folks that had previously had insurance, and were simply exchanging one plan for another.  But a recent study from the Urban Institute Health Policy Center shows that the number of uninsured adults in the US decreased by at least 5.4 million since enrollment began in October 2013.  This represents a drop of 2.7 percentage points, or a relative decrease of 15%.

I say at least because the data come with 2 caveats.  First, it does not reflect enrollments in the last couple of weeks of March, when activity surged.  Second, it does not show the effect of other provisions that have been shown to increase coverage, especially the provision allowing young adults to remain on their parents’ insurance until age 26 (that number is estimated to be about 3.1 million).

This is a huge improvement in America’s shameful status among developed countries in providing healthcare coverage for its citizens.  The impact could have been even greater.  The drop in uninsured was less than half as big in the 24 states that opted not to expand Medicaid compared with those that did, leaving millions more uninsured.

BUT – I’m not giving 3 cheers yet.  The stated goals of the Affordable Care Act were to expand coverage (good start but more work to do), and to decrease costs (still more work to do).  In a previous blog I talked about how the ACA has likely already had a positive impact on healthcare spending.  However, now that there are millions more Americans with coverage, demand may start to increase, potentially reversing some of those gains.

Nevertheless, while the jury is still out, I think 8.5 million Americans with insurance coverage they didn’t used to have is something to celebrate.  I’m sure those people are celebrating.

Experience Matters

March 28, 2014

CHW Logo
In healthcare, experience matters.  And I don’t mean years of training or tenure – I mean patient experience.  For years we’ve resisted calling healthcare a “service industry,” and felt that “quality” was all that mattered.  If by quality we mean clinical effectiveness and outcomes, I’d argue it’s important but not enough.  Which is why the Institute of Medicine defines quality as having other domains beside effectiveness: safety, timeliness, efficiency, equity, and patient-centeredness.  This last dimension of quality includes incorporating  patients and families into the decision-making process and considering their preferences.  And patients are pretty clear that one of their preferences is to be treated with courtesy and respect.

There are now numerous sites where people can post ratings and comments about physicians, such as   The vast majority of comments are not about technical skill or knowledge, they are about listening, wait times, courtesy, bedside manner.  These things matter, especially as individuals are paying more of the very high cost of health care out of their own pockets, and demanding value for what they spend.  Not only consumers, but professional organizations such as the Robert Wood Johnson Foundation are leading efforts to make the health care experience more patient-friendly.

At Children’s Hospital of Wisconsin, we have long measured patient satisfaction and experience, but haven’t always been as focused on it as we should be.  This is changing, and with people at all levels of the organization starting to pay attention to experience, satisfaction is improving.  In some of our areas, like imaging, the emergency department, and the Surgicenter, our scores are now well above the national average for pediatric hospitals.  But we have a ways to go.  Eventually, we need to rethink many of our assumptions about how healthcare is different from other industries.  People used to think that cars, computers, and airline travel were different, too – too complex for the average person to evaluate on their own.  Yes, computers are complex; it takes as much education and training to build and program computers as it does to become a doctor.  But somehow even someone like me who has never taken a computer science class in his life can buy and use one without any specialized help.  Of course, when things don’t go right I seek expert assistance, but even there, I shop for that service the same way I do for everything else, judging them not on whether they can fix the problem (I expect that), but on how long it takes, whether they are nice or rude, how well they explain things, and how much it costs.

If you think experience doesn’t matter, watch this video comparing health care to the airline industry.  It’s funny and shameful at the same time.

Sold to the Highest Bidder

March 21, 2014

CHW LogoImagine if William Shatner were your doctor.  Crazy, no?  What about buying your health care from him?  After all, he has been the face of, where you can bid on hotels, flights, cruises, etc.  Almost anything but healthcare.

Until now.  Along comes MediBid™, an online auction for medical care including physician services, hospital stays, durable equipment, etc.  Sounds like something from The Onion, but it’s real.  According to their Web site, “MediBid offers cash paying patients seeking care the ability to find the medical practitioner who best fits that patient’s set of criteria.  Criteria that can include price, location, time-to-treatment…. even professional credentials.  Unlike traditional network sources, your healthcare decisions are in your hands, not someone else’s.”

Prospective patients enter a request for a procedure such as MRI or knee replacement, and providers may then submit bids for that procedure.  MediBid charges fees for both requestors and bidders, but otherwise does not get involved in the transaction.  They do not handle any payments, and make it clear that they do not do any background checking or verification of a provider’s credentials.  So let the buyer beware.

It sounds radical, maybe even a little creepy.  And I’m not sure this is truly taking off; one article from October 2013 says that since its founding in 2010 MediBid has connected 1800 providers and patients.  But Star Trek communication devices and talking computers seemed far-fetched in the days when William Shatner played Captain Kirk.  Let’s not laugh this one off just yet.

Getting to the Healthiest Kids – Update

March 14, 2014

CHW LogoThe link between socioeconomic status and child health has long been appreciated at least intuitively, but the details of that link are becoming clearer.  Two recent studies in the journal Pediatrics provide specific details about aspects of poverty that adversely affect the health of children.  Such knowledge, in turn, can inform policies aimed at making kids healthier.

1.  A study from Cornell University examined the relationship between income inequality and child abuse.  Looking across all 3142 US counties, there was a linear relationship whereby the counties with the highest levels of income inequality had the highest rates of child maltreatment.  The effect was independent of other factors such as absolute levels of poverty and education.  This confirms other studies that have shown that not only income but income inequality affects the health and well-being of individuals.

2.  Researchers at American University reported that, among households at below 300% of the federal poverty level, food prices are associated with rates of overweight children. Specifically, higher local prices of fresh fruits and vegetables correlate with higher weight and body mass index (BMI); conversely, lower prices for soft drinks correlate with higher weight and BMI.  The authors exploited the fact that the database they used tracked children over time.  Not only was the relationship between fresh food price and weight true when looking across children who live in different areas, but it also held when they examined children who had moved over time; such children’s weight trajectory was different depending on food cost differences between old and new areas of residence.  Again, this bolsters our previous understanding about how lack of access to fresh foods can adversely affect child health.  Fortunately, the 2014 farm bill contains some provisions that may mitigate this, including increased support for organic farming, ability to use food stamps at farmer’s markets, and programs getting schools to grow their own food.

Decisions, Decisions

March 10, 2014

CHW LogoI consider myself somewhat  risk averse.  For example, my brother-in-law is an entrepreneur who is borrowing and putting a good bit of his savings to start a new craft brewery; as much fun as that sounds, I don’t have the stomach for that kind of gamble, and I suspect the majority of people don’t (that’s why most of us are not entrepreneurs).  On the other hand, lots of people buy lottery tickets, which also seems like a pretty risky gamble.  So what’s the difference?

The answer may lie in prospect theory, as explained in a recent JAMA article.  Building on work by behavioral economists Daniel Kahneman and Amos Tversky (and laid out in more detail in Kahneman’s book Thinking, Fast and Slow), prospect theory is a framework for understanding choice.  There are several basic tenets.  First is that people tend to simplify complex choices by focusing on key differences, while ignoring similarities.  For example, when deciding on a vacation destination, if choice A will cost $2000 and require 8 hours of travel, while choice B costs $5000 but only 7 hours of travel, the difference in travel time will likely not even be considered in the decision, which then becomes simpler.  Second, choices are made with regard to a reference point, not on the absolute value.  Two candidates for a position will see a salary offer of $50,000 very differently if one is currently making $40,000 and the other is making $60,000.  This may seem obvious.  In health care, treatment outcomes are going to be weighed against that reference point, which may be a patient’s current state of health, or a remembered past state of health.  Difference in reference points explain some of the difference in risk tolerance between individuals.  A corollary to the concept of the reference point, one that is backed by a good deal of experimental evidence, is that people tend to prioritize preventing a loss over achieving an equal gain.  This is called the endowment effect – we feel the loss of something we already have more than the regret at not getting something we only hoped for.   In one interesting experiment, people were offered the chance to buy a mug a coffee mug, or given the mug and then given the chance to sell it.  People were willing to pay on average about half as much to acquire mug they didn’t have, compared with what they would accept to part with the one they had gotten for free.  It appears that a bird in the hand is quite literally worth two in the bush.

The third element of prospect theory is that, after simplifying and framing the choice, people consider the desirability and probability of each outcome and estimate the expected value of each choice.  However, there are many cognitive inconsistencies and biases that creep in.  For example, we tend to place more weight on proportional than absolute differences.  A difference between winning  $1 and $3 is not treated the same as the difference between $98 and $100, even though each results in someone being able to buy one additional cup of coffee the next day.   Conversely, very small probabilities are difficult to deal with intuitively.  We tend to either ignore them completely (as in, my chance of winning the lottery went up 10-fold, from 1 in 10,000,000 to 1,000,000 – it’s still nearly impossible so I’m still not buying a ticket) or blow them out of proportion (as in, Did you see that the size of the pot is at a record $700 million – how could you not buy a ticket!)  Prospect theory only helps us understand decision-making.  It doesn’t actually make it rational.

These insights into how we, and our patients, make decisions may help us in practice.  There are several studies showing that parents and physicians differ in the values they assign to various short- and long-term outcomes, presumably because of different reference points.  Parents and patients also make different choices when the decision is framed in terms of a gain or a loss, for example, chance of survival versus chance of dying.  This all suggests a need to see the world through someone else’s eyes if we want to help them come to the best decision for them.  We need to understand the filters – theirs and ours.

It also helps us potentially make better business decisions.  Prospect theory suggests that in the face of a small chance for a large gain, we tend to reject choices with a much greater degree of certainty for something that is less lucrative.  For  example, if offered $2 in cash or a $2 lottery ticket, most people would take the ticket.  We don’t see either choice as a loss, and from our reference point the high likelihood of the small gain may not be worth forgoing the small chance to hit it big.  But a forgone gain IS a loss (I gave up a free cup of coffee).  This argues for a rigorous process for providing an objective calculation of the actual expected value of  each of the possibilities in a business decision.  As another example, “cutting one’s losses” may be the best strategy for avoiding further losses.  But we know that, because of the endowment effect, we tend to overvalue what we already have, which makes it emotionally difficult to let go even when that would make the most sense.

In  the meantime, while starting a brewery may not be in my risk comfort zone, the Old Bust Head beer is pretty tasty – trying one is no gamble at all.  (Alas, it’s only available in Virginia for now….)


Get every new post delivered to your Inbox.

Join 302 other followers

%d bloggers like this: