Saturday, January 18, 2014

The Drive To Do Good Work When No One Is Looking

Himmelstein and Woolhandler have hit it out of the park again.  This time in a seemingly modest letter to the editor in the weeks JAMA.  Although their main point was the perversity of pay-for-performance incentives in the final paragraph they make this observation:

"We fear that pay-for-performance incentives will undermine the mindset required for good physician practice (ie, the drive to do good work even when no one is looking)." (p.304)

This is a quote that everyone should pay attention to because the so-called financial incentives that have been with us for years are highly problematic and have not contributed a thing to the quality of medical practice.  The first consideration is whether or not what Himmelstein and Woolhandler say is true?  It has  clear validity.  Physicians are trained to do the right thing in terms of individual patient care.  That involves dedication and adhering to professional standards like continuity of care.  Physicians as a group are driven people.  They will work past the point of exhaustion especially f they feel that they need to follow up a certain problem for a certain patient.  Many primary care physicians are working 24/7 in order to keep up on paperwork, labs, prescriptions, and test results.  It is not possible to "call it a night" if there is a critical lab value that needs to be evaluated or a patient that needs to be informed.  There are plenty of decisions in medicine that have a high inherent degree of uncertainty and that can lead to sleepless nights searching for an answer.  The electronic health record (EHR) has increased that burden.  I have colleagues that tell me that every morning when they walk into clinic they may have as many as 200 lab results waiting for them in the EHR.  If they do not have assistance with triage, they have to personally view these results and do something about them.  They may also have to respond to direct e-mails if they are affiliated with a plan that allows patients to directly e-mail their physician.  That is before they begin seeing a full schedule of patients and seeing urgent problems that day.  Responsibility to the patient is what drives this process not trivial financial rewards and penalties based on illogical outcome measures.

Why all of the confusion?  I think the problem can be analyzed at several levels.  At the antiphysician level there are obvious forces in play that want to coopt the professionalism of physicians and make it seem like without a business and government bureaucracy that physicians would be out of control.  In other words the public needs the government and big business to control physicians.  This was the essential rhetoric that led to the managed care era in the first place.  Physicians were greedy, did too many procedures for financial gain, and needed to be managed by a large and expensive business administration.  The managed care theory completely ignores two basic facts.  The most obvious fact is that three decades of managed care has not controlled health care inflation or increased health care access as promised.  That is compounded by the fact that many of these companies currently own all of the means of production including the doctors, hospitals, labs, and MRI scanners.  But nobody has accused them of being greedy.  The second fact is that independent review of the problem from the 1990s showed that the political theory that physicians were engaged in massive overutilization was so inaccurate that they closed the government review program down.  How often does a government program get shut down?

The assault on the professionalism of physicians reached new heights in the 1990s when the federal government under the Clinton administration decided that fraud conducted by physicians was also a major driver of health care inflation.  Armed with billing codes and guns, it was an era where FBI agents were trained to enter offices, get the records and determine of the documentation matched the billing code.  There were several high profile cases where the Department of Justice decided that "fraud" had occurred.  The fraud they were referring to in one of the landmark cases was attending physicians not writing an extensive enough note in addition to whatever note the resident physician wrote.  That led to about two decades of excessive and unnecessary documentation.  Right up until the point that the federal government could grant this power to managed care companies and give them the power to deny payment and demand reimbursement back based on their interpretation of whether the documentation supported the service provided.

Throughout all of this heavy handed political intervention physicians have continued to do the right thing and maintain their professionalism but the rhetoric has taken a heavy toll.  It is not unusual for a person to enter a health facility owned by a managed care company and see some credos from the physician code of conduct plastered on the walls in an elevator.  When that happens the company is generally sending a clear message: "It is only because of us that medical professionals behave in a professional manner."  Nothing is farther from the truth.  There is no carrot or stick that keeps a neurosurgery residents on call in a hospital every day and night for a year at a time.  There is no carrot or stick that keeps psychiatrists treating aggressive patients who continually threaten and may try to assault them.  There is no carrot or stick that keeps physicians going to the point of exhaustion and to the point that they realize that they are clearly spending much more time caring for other people than their own families.  A typical response from people with those scenarios is either: "You don't have to do that." or "You don't have to work that hard."  Physicians currently waste 20-30% of their time on bureaucratic nonsense dreamt up by people who themselves do not provide direct patient care.  In the time left they are expected to see more patients and do more documentation than physicians have ever had to do at any point in history.

I don't mean to suggest that there are not people in all walks of life who do the right thing at work.  As I am typing this I recall my father describing one of his coworkers sometime back in the 1960s.  My father was a railroad engineer.  He drove locomotives of both the diesel and steam variety.  The railroad world in those days was a parallel universe that I would occasionally get a glimpse of.  Massive buildings smelling of diesel fuel storing massive, loud, pulsating diesel locomotives.  They just let them run 24 hours a day.  My father and a group of his railroad colleagues were rebuilding the porch on our house one day and they were all talking very positively about a car knocker who worked for the same railroad.  A car knocker is a rail car repairman.  They keep trains rolling.  All of my father's co workers and my father all praised this person because at work he always seemed to do the right thing.  The quality that they seemed to admire the most was that he would always attend to part of the task even when he lacked a certain skill that required him to pass part of the job on.  That was apparently a rare trait and that was my first observation of the quality of doing good work and how it was widely admired on its own merit.  No financial reward or employee of the week parking spot.  Just the mostly hidden admiration of your coworkers.

To me the most perverse aspect of physician "incentives" is that they are already earned by physicians.  In the managed care world, groups of physicians are subjected to a "holdback" of anywhere from 5-15% of their gross billing.  The idea is that unless the entire department makes the productivity expectation (an arbitrary number set by an administrator) the money is lost.  If the financial target is made money "awarded" was already earned.  There are no bonuses only penalties.  A lot of the incentives for reporting various measurements or complying with the next bureaucratic hoop are in the 1-2% range.  That is important only for the bean counters described in the letter.  Make no mistake about it - they will exert whatever pressure they can on physicians to "make" that 1 or 2%.  And make no mistake about it - twenty years that bean counter job would not have existed.

And there still would have been physicians in hospitals all night long trying to do the right thing and not caring who knows about it.

George Dawson, MD, DFAPA

Himmelstein DU, Woolhandler S. Physician payment incentives to improve care quality. JAMA. 2014 Jan 15;311(3):304. doi: 10.1001/jama.2013.284475. PubMed PMID: 24430325.

Additional Clinical Note 1:  Some health care settings have really taken the incentive programs to an absurd level.  Mandatory pep rallys for the employees, employee of the week or month awards and the expectation that everyone generates a huge amount of false enthusiasm for these rewards.  At the same time the confidential 360 degree performance review is the latest tool that administrators can use against physicians.  In this exercise a panel of anonymous multidisciplinary coworkers rate physicians on a number of non-medical standards (like how well they support corporate standards).  The performance review is basically a combination of subjective impressions that have very little to do with the physicians medical competence.  It is all part of the corporate beauty contest that is passed off as medical quality.  Most managed care companies seem to not know the meaning of the word.

Additional Clinical Note 2:  Governments and managed care companies are never at a loss for measuring things that are completely irrelevant to the provision of quality medical care.  The best example is patient satisfaction surveys.  If you are ever handed one as you leave the hospital or clinic the first question to ask yourself is whether the questions seem familiar.  If they do it is because your were scripted.  That means that the person you just talked with covered a number of bullet points corresponding to your survey.  It is designed to maximize the score on that satisfaction survey.  These results are often linked with incentives.  But nobody ever talks about the fact that you could have received the best care in the world and may not been able to complete a survey when you left the hospital.   


  1. Excellent post on a vexing and unfortunately complex topic. One the one hand, most physicians have strong internal motivation to pursue clinical excellence. For this majority, external incentives are at best irrelevant, and at worst plainly insulting — and as Himmelstein and Woolhandler suggest they may paradoxically sap motivation; see the "overjustification effect", Wikipedia has a good discussion.

    On the other hand, not everyone in medicine, or any other field, aspires to the highest standards of professionalism. A parallel thread in our history has been a kind of exceptionalism borne of hubris. For decades we physicians claimed we were unaffected by on-call sleep deprivation, too smart to be swayed by billions in drug company marketing, and that we knew better than our patients what was best for them, even when the question wasn't technically medical. America's cynicism toward authority may have started with police and politicians in the 1960s, but it's not terribly surprising that doctors were swept in, too, along the way. In part, we brought it on ourselves.

    Like you, I am insulted by the smarmy faxes from drug benefit managers who "as fellow members of the health care team" just want to assure I'm prescribing correctly. I'm offended that Medicare decides my medication management is more valuable than my psychotherapy even when it isn't, and that Medicare will cut a percentage or two from my reimbursement next year because I didn't participate in their arcane and silly quality-assurance system in 2013. But at the same time I also feel our actions speak louder than our words. Protesting that we are not afforded the respect we deserve means little unless we both earn and insist on that respect. We can't have it both ways, claiming to be above reproach and review, yet acting in ways that call for it. It only takes a minority of physicians to tarnish the profession as a whole. If we don't police ourselves, others will be happy to do it for us.

    1. Excellent points Steven. I take a different view of the threshold problem that you describe. I don't see it as a question of respect but a question of no evidence. For example when Public Citizen has a political initiative suggesting that there are "not enough" complaints about physicians to state medical boards, where is the evidence? Medical boards currently have an extremely low threshold for investigating physicians and the standard is lower than any legal standard. In the state where I practice all complaints are accepted as legitimate no matter how implausible and physicians are really never cleared. Old complaints are kept on file. I don't know of many other professions where that occurs.

      The body of my post describes the history of how physicians have been smeared in this country for political and economic gain. The evidence cited at the time has all but evaporated and the rhetoric lives on. It is clear to me that the necessary policing of physicians as "necessary because physicians can't do it themselves" is a convenient gambit to concentrate the political power with politicians and managed care. The evidence hits me in the face every day when I see ads for these companies talking about how they care for people and treat people. That rhetoric needs to be aggressively countered because it is one of the major factors that had led to the current overregulation of physicians.

  2. If your options was pay for output or pay for outcomes, which payment/pricing model would you choose. It could be argued that current pay for procedure approach (regardless of how necessary, expensive, or risky) leads physicians do to provide extra, unnecessary, wasteful, and potentially harmful care despite their proclaimed motivations. I can see how pay-for-outcomes can sap intrinsic motivation for clinical excellence, but I'm not sure that having extrinsic motivation for clinical excellence will dampen patient outcomes (they might improve them). The devil is in the details as defining what is an "outcome" and what "measures" are used to indicate them can be a political and sloppy process often done by people who don't care much for patients.
    Still, for me, given the choices, I'd take outcome-oriented payment structures over process-oriented almost every time. A socialized approach that paid doctors based on the health of their communities and encouraging tax payers and voters to support more efficient community health initiatives to reduce costs i think would be even better.

    1. You are correct in saying that the devil is in the details and that is one of my main points. If you are a physician dealing with even moderately complex care the outcome variable you are interested in is adequately treating the patient's problem. In real life this is what happens if you are employed by or contract with a managed care system:

      1. You are "incentivized" for productivity. Many companies will hold back part of your salary unless you see enough patients. The productivity units are arbitrary and arbitrarily increased every year. Part of your work if you happen to teach or lecture is not included in the productivity expectation. A substantial part of your work is an increasing paperwork burden because you are doing the job of many people the corporations have fired (stenographers, billing and coding staff, etc). Even though that may be as much as 2-3 hours a day - you don't get paid for that either.

      2. The outcomes variables themselves are totally arbitrary business related variables rather than variables based on medical outcomes.

      The example I have used on this blog many times is the managed care length of stay variable for psychiatric hospitalizations. In the hypothetical case of an emergency hospitalization for psychosis it may take a month for that person to be stabilized. The Medicare based length of stay is 8 days. Managed care companies especially those who own the hospitals want those people out in 4-5 days. In many cases the person is acutely symptomatic and at high risk for complications.

      This is a good example of using an outcome variable that is demoralizing and totally at odds with clinical excellence. In fact most outpatient psychiatrists view inpatient services as being essentially worthless because their patients leave the hospital as symptomatic as they were when they entered.

      Welcome to the world of business outcomes being mapped on to medical illness and biological variability.