Big Data - What Is It Good For?

Big Data and Data Science have been buzzwords in science and industry for over a decade.  A Medline search shows over a thousand current references to Big Data in healthcare.  A good starting point is consider what is meant by Big Data and then discuss the implications.  A quick scan of the references shows that they vary greatly in technical complexity.  A standard definition from Google is: "extremely large data sets that may be analyzed computationally to reveal patterns, trends, and associations, especially relating to human behavior and interactions."  These techniques developed because of the widespread availability of digitized data and the ease with which sets of behavioral choices (in the form of mouse clicks) can be collected on web sites.  In many cases specific data collection paradigms can be used to elicit the information, but there is also a wealth of static data out there as well.  In health care, any electronic health record is a massive source of static data. Financial, real estate, educational records and records of all of these transactions are also a significant source.

Most  Americans logging in to set up a Social Security account online (ssa.gov)  in the past couple of years would be surprised at what it takes to complete the job.  After the preliminary information there is a set of 5 security questions.  Four of those questions are about your detailed personal credit history - home mortgage information and credit card history.  When Social Security was initially set up in 1936 there was widespread concern that Social Security Numbers would become national identifiers.  At one point Congress had to assure the electorate that the number would not be used for that purpose.  Since then the SSN has been used for multiple identification purposes including credit reporting.  At this point it seems that we have come full circle.  Congress invented the SSN and told people it would not be an identifier.  They mandated its use as an identifier. Congress authorized and basically invented the credit reporting system in the United States.  The federal government currently uses the credit reporting system to quiz taxpayers wanting to set up a Social Security account online.  In the meantime, large amounts of financial, legal, and health care data are being collected about you under your SSN in data systems everywhere.  At this point the full amount of that data and the reasons why it is being collected for any person in the US is unknown because it is all collected without your knowledge or your consent.  It is impossible to "opt out" from this data collection.  The federal government does have an initiative to remove SSNs from health records, but there are so many other identifiers out there right now, this effort is too little and too late.

Additional sources of data include your online foot print including sites that you may have visited and what you seem to be interested in.  A visit to Amazon for example and a quick look at an expensive digital camera may result in that same camera with a link to Amazon in the margin of very other web page you see for the next two weeks.  Expensive digital cameras of a different brand than the one you originally looked at may start showing up.  You may notice product ads showing up in your Facebook feed that you mentioned casually to your friends during a conversation there.  The conversation could be as generic as bicycle seats and suddenly you are seeing a flurry of ads for bicycle seats.  Any number of web sites encourage to sign in with other accounts and then share your account information with them.  All of this data provides companies with what they need to fuel their predictive algorithms to sell you a product.  It provides the major advertisers in this space like Google and Facebook with a huge revenue source because based on the scale and personalization of these ads - they are effective.  Big Data seems to be very good for business.  But is there a downside?        

That brings me to a current resource on the nefarious uses of Big Data written by an expert in the field.  Cathy O'Neil is a PhD in mathematics.  Her PhD work was in algebraic number theory.  She started work as an academician but subsequently worked for a hedge fund, work as a data scientist for several firms and currently heads the Lede Program in Data Journalism at Columbia University.  I am familiar with her work through her blog MathBabe.  Her newly released book Weapons of Math Destruction takes a look at the dark side of Big Data specifically how data collection and biased algorithms can be good for administrators, politicians and business but bad for anyone who falls under the influence of those agencies and their work.  In the introduction she leads of with the example of teacher assessments.  I was familiar with a scattergram that she had posted on her web site showing that year to year teacher assessment scores were essentially uncorrelated or random.  In the book she describes the human toll in this case a teacher fired because of this defective algorithm.  In another example later in the book, an experienced teacher scored a 6 out of 100 on a "value-added" teacher evaluation.  Only tenure kept him from getting fired.  The scoring algorithm was opaque and nobody could tell him what had happened.  The next year he scored the 96 out of 100.  But the algorithm was so flawed he knew that score was no more legitimate than the last one.  With the politicized environment surrounding teaching the proponents of teacher "accountability" like this variation since it fits their ideas about the system retaining incompetent teachers that need to be weeded out.  In fact, the algorithm is defective and like many is based on erroneous assumptions.

I personally know that physicians are subjected to the same processes as teachers, but so far it is less technologically advanced.  O'Neill points out that there is nothing magical about algorithms.  That they frequently incorporate the biases of the people who design and contract for them.  Opacity and a lack of correction by feedback is another feature.  I worked for the same employer for a number of years when physician "accountability" measures were put in place.  The "algorithm" for salary went something like this RVU Productivity + Outside Billing + Citizenship = Pay.  RVUs were the total number of patients seen according the the biased government and managed care billing schemes.  Outside billing was any consulting work done outside of the clinical work that was billed through the department.  Citizenship included teaching and administrative duties as well as any Grand Rounds or CME lectures that were done.  In other words apart from the subjectively based billing scheme all of the inputs are almost totally subjective and influenced by all kinds of pseudoaccountability measures along the way.  For example, in parallel with the teacher ranked on the algorithm, I was told one year that I had achieved the top rank in terms of documentation in a group of about 25 physicians.  The next year - making no changes at all in terms of that documentation - I was dead last.  My conclusion, like the teacher in the example was that the rating scheme was completely bogus and with that kind of a scheme who cares about the results?

The number of based algorithms applied to physicians has eerie parallels to those mentioned in WMD.  Here are a few that I picked out on the first read:

1.  The algorithm is based on faulty data - the teacher evaluation algorithms were based on a faulty interpretation of data in the Nation at Risk report.  The report concluded that teachers were responsible for declining SAT scores between 1963 and 1980.  When Sandia Labs reanalyzed the data 7 years later they found that an great expansion in the number of people taking the test was responsible for decreased average score but subgroup analysis by income group showed improved scores for each group (p. 136).  The only reason that teachers are still being blamed is political convention.  I posted here several years ago that the top ranked students in the world in Finland are taught by teachers who are assumed to be professionals and who are not critiqued on test results.

The parallel in medicine was the entire reason that medicine is currently managed by the government and the healthcare industry.  It was based on criticism in the 1980s that doctors were lining their pockets by performing unnecessary procedures and that work quality was poor.  That should sound familiar because that criticism has been carried forward despite a major study that showed it was completely wrong.  The massive Peer Review Standards Organizations (PRSO) in each state in the 1990s conducted rigorous reviews of all Medicare hospitalizations and concluded that there was so little overutilization and so few quality problems that it would not pay to continue the program.  The only reason that managed care companies exist today is by political convention.

2.  An effective teacher like an effective doctor is too complex to model - When that happens only indirect measures or "crude proxies" (p. 208) can be used to estimate effectiveness.  In medicine like teaching - the proxy measures are incredibly crude.  They generally depend on diagnosis, poorly account for comorbid illness, and the outcome measures are heavily influenced by business rather than medical decision making.  The best examples are length of stay parameter and readmission parameters.  Every physician knows that there are set payment schedules based on the supposed ideal length of stay for a particular illness.  The business influence in the discharge decision is so malignant these days that non-physician case managers are present to pressure physicians into discharging patients.  If the discharge beats the length of stay parameter - the hospital makes money.  I sat in a meeting at one point and asked the obvious question: "OK - we have completed the discharge checklist - do we know the outcomes?  How do the patients do when they are discharged by this process?  How many of them die?"  Dead silence followed.  Most people would be shocked to hear that what passes for evidence based medicine is often a checklist that has no meaning in the real world.  Making the points on the checklist is good for advertising though.

3.  There is a lack of transparency in the overall process - The teachers in WMD who were blindsided by the algorithm were never told how that conclusion was reached.  I encountered the same problem in a managed care organization when it was clear to me that administrators with no knowledge of psychiatry were telling us what to do.  In some cases, "consultants" were brought in to write reports to confirm the most recent administrative edicts.  When I asked my boss if I could talk with the people sending out the edicts I was informed that there was a "firewall" between clinicians and upper management.  This lack of feedback is another critical dimension of algorithms gone astray.  If you are writing an algorithm biased toward a business goal - why would you want feedback from clinicians?  Why would you want any humanity or clinical judgment added especially in the case of psychiatric care?  Let's just have a dangerousness algorithm and leave it at that.  Those are the only people who get acute treatment, even though it is patently unfair relative to how the rest of medicine works.             

Big Data is good for science.  We can't do elementary particle physics or genomic analysis very well without it.  Big Data is also good for business is much different ways.  There are clearly people out there who cannot resist buying items online if the Amazon algorithm shows it to them enough times across a number of web pages.  Big data in business can also come up with billing algorithms that have less to do with reality than making a profit.  Similar programs can be found for employee scheduling, performance analysis, and downsizing.  The problems happens when the business biases of Big Data are introduced to science and medicine.  Those techniques are responsible for an array of pseudoquality and pseudoaccountability measures for physicians, hospitals, and clinics.

Unfortunately physicians seem to have given up to the political conventions that have been put upon us.  Some administrator somewhere suggests that quality care now depends on a patient portal into an electronic health record and a certain number of emails sent by patient to their physician every month.  Across the country that will result in hundreds of millions of emails to physicians who are already burned out creating highly stylized documentation that is used only for billing purposes.  Terabytes of useless information that nobody will ever read again - the product of a totally subjective billing and coding process that started over two decades ago.  Is there any data that email communication is tied to the effectiveness or technical expertise of the physician?  I doubt it.  I worked with great physicians long before email existed.

It is about time that somebody pointed out these manipulations provide plenty of leverage for the management class in this country at the expense of everyone else.  It is well past the time that doctors should be confronting this charade.  


George Dawson, MD, DFAPA




References:

1.  Cathy O'Neill.  Weapons of Math Destruction - How Big Data Increases Inequality And Threatens Democracy.  Crown Publishing Group.  New York, NY, 2016.  I highly recommend this book for a look at the other side of Big Data.  It is written in non-technical language and is very readable.

Attribution:

The photo at the top is a Server Room in CERN By Florian Hirzinger - www.fh-ap.com (Own work (Florian Hirzinger)) [CC BY-SA 3.0 (http://creativecommons.org/licenses/by-sa/3.0) or GFDL (http://www.gnu.org/copyleft/fdl.html)], via Wikimedia Commons"   href="https://commons.wikimedia.org/wiki/File%3ACERN_Server_03.jpg"><img width="512" alt="CERN Server 03" src="https://upload.wikimedia.org/wikipedia/commons/thumb/d/d7/CERN_Server_03.jpg/512px-CERN_Server_03.jpg"/></a>

  

Psychosis Idealized

I thought I would provide a counterpoint to the New York Times editorial entitled "Medicating A Prophet" written by Psychiatrist Irene Hurford (1).  The opinion piece is available free online and I encourage anyone interested in the topic to read the article rather than accepting my summary here as adequate.  I will say from the outset that I am not a stranger to any of the issues that Dr. Hurford discusses either clinically or personally.  The bulk of my career was spent treating people with severe mental illnesses and addictions.  Once you have worked in that setting, it is clear that many people who are severely ill need involuntary treatment and that is one of the decision points that she addresses.

In her essay, Dr. Hurford describes an early call experience during her residency.  She was asked to assess a man in the emergency department (ED) who had been delusional for 30 years.  The delusions were religious and grandiose in nature.  He was a college graduate but was homeless living on the street in Philadelphia.  He also had AIDS and the complication Kaposi sarcoma.  His reason for being in the ED was "to preach".  Dr. Hurford encourages him to come in for voluntary treatment but he refuses.  At that point she ponders involuntary treatment but in the essay decides to discuss the patient's right to psychosis. Later we learn that she made the decision but has decided to analyze that decision in retrospect based on factors that she has encountered since.

One of those factors was the influence of a professional colleague who based on her own experience with psychosis and that colleague's mother's experience suggested that thoughts about living "in psychosis" and outside of psychosis need to be challenged.  She basically states that the problem may be within the beholder rather than the identified patient. Following that logic, it makes sense to show up in an ED to preach while ignoring serious health problems.  It also makes sense to make decisions about the person's "in psychosis" experience knowing so little about them.  In my experience, nobody in the ED calls the parents or family of a 50 year old street person, to get a clear picture of how the psychosis has truly affected him.  When I have treated these people on an inpatient unit and made those calls, I have never heard that the patient was well served by the psychosis.  Not a single time.  In many cases, family members were surprised to hear that person was still alive.

Dr. Hurford advances a number of other arguments that I call into question.  She uses a very loose definition of insight as a "failure to accept an alternative view of reality".  She turns this around to suggest that anyone who does not accept this premise (implicitly the treating physician) also lacks insight.  I don't think that you can practice psychiatry and not be comfortable with alternative realities.  I would suggest a more appropriate definition of insight as a decision-making process.  Can I accurately assess how I am doing in the world?  Am I making decisions in my best interest?  Are those decisions consistent with my ability to survive?  If I realize that I am not doing well can I get help?  Pretty basic decisions.  Not a question of lifestyle choices.  To have a lifestyle you have to live.  That is the kind of insight that I am used to dealing with.

Dr. Hurford discusses a case of a young patient with a psychotic disorder who stopped taking his medications and started using cannabis on  daily basis.  He dropped out of college and became progressively incoherent and then mute.  She is concerned about traumatizing the patient by "enforcing" treatment even though he cannot "eat, sleep, and talk."  I don't follow the logic that some treatment intervention - even basic detoxification from cannabis is somehow more traumatic than not eating, sleeping, or being able to communicate.  How is that a preferred alternative existence?

At that point she digresses to a very brief overview of the usual comments about mental illness being only peripherally associated with violence and the lack of evidence that forced treatment led to fewer hospitalizations, arrests, or a better quality of life.  She cites a meta-analysis of three randomized-controlled studies of more than 700 people.  There are a lot of reasons why meta-analyses are not superior to the actual trial data.  There are also a lot of reasons why truly clinical samples with these problems cannot be ethically randomized or included in the studies.  There are also reasons why I would expect the entry points into these studies to be highly variable as well as the treatment resources that are involved.  In Minnesota, we have 87 counties and the rule is that there are 87 interpretations of the commitment act for involuntary treatment.  There are two corollaries operating here.  The first is that the courts will be very liberal in terms of dropping commitments until something bad happens.  At that point the pendulum swings back in the direction of more frequent commitment.  The second is that only the wealthiest counties in the state can afford to provide adequate resources to treat the severely mentally ill.  Even then there is no assurance that the counties that can afford it will actually provide the care.  Some currently function like managed care companies and ration the care.  They can end up rationing care and commitments in order to save the county money.  The lack of evidence that forced care does anything may be more of an indictment of the lack of quality or consistency in delivering care and interpreting the law  and rationing care more than anything.  I have personally treated many times the number of people with forced treatment than in the meta-analysis and there is no doubt that the outcomes were better than with no treatment.

The outcome variables cited by Dr. Hurford are also dreadfully lacking compared with what can be seen routinely in clinical settings.  They include very adverse outcomes in encounters with the police including getting shot, dying from a treatable illness, suicide, loss of relationships with spouses and children, loss of a job and income, and acute loss of life due to poor insight and judgment.  In Minnesota, all that takes is going outside in the winter time without adequate protective clothing and you are dead or in the Burn Unit with frostbite.
     
Right now we are in the midst of a sweeping cultural change that idealizes psychosis and suggests that hallucinogens and cannabis are therapeutic drugs.  That will put the next generation or two of people with psychoses, mood disorders, and substance use disorders at risk for chronicity and every possible negative outcome.  A point that should not be lost on anyone is how no care for psychosis is "cost-effective" care when the total impact on the patient is ignored.  My point in writing this rebuttal is really advice for the people in these generations.  Ask any psychiatrist treating you or your family member where they stand on this issue.  

Especially if you value psychotic symptoms a lot less than your psychiatrist does.



George Dawson, MD, DFAPA





References:

1.  Irene Hurford.  Medicating A Prophet.  New York Times.  October 1, 2016.

Cancer Care In Psychiatry - Yes It Happens

Great Case Records of the Massachusetts General Hospital in the New England Journal of Medicine this week.  A psychiatrist is presenting a case of complicated disorder and coordinating care with oncologists after a cancer diagnosis is made.  It must have taken an editorial change in the Journal to get psychiatry more front and center in this prominent medical journal.  This article has a lot of meaning for me, because the bulk of my career was spent on these issues.  When you are the inpatient doc or the psychiatrist staffing the case management teams - either Assertive Community Treatment (ACT) or some other case management model a lot of this important work falls to you because there is nobody else out there.  Contrary to the popular hype about collaborative care - people with severe mental illnesses generally avoid medical clinics and non-psychiatric physicians.  There are always some exceptions, but as I have said many times there are reasons that people do not go into psychiatry.  Talking to people with communication problems, irrational thought processes, and atypical social behavior are high on the list.  Some of the best primary care physicians recognize this and like the oncology clinic described in this paper give people with severe mental illnesses wide latitude in terms of appointments and treatment schedules,  but they can only do so much.

What is needed on the front end is a psychiatrist with strong medical skills to figure out the problem and get other medical staff involved.  I don't mean looking at PHQ-9 scores and suggesting medication adjustments.  I mean actually talking with the patients.  Very frequently a person with impaired judgment due to a psychiatric disorder will be fully cooperative in one setting but not another.  Consider a patient who is acutely admitted with very high blood pressure (260/140).  He has a history of schizophrenia and hypertension requiring moderate to high doses of two different antihypertensives.  In this case the patient has gone off of both medications and maintenance antipsychotic medication.  He is agitated and paranoid.  It is impossible to determine if he is also delirious due to the presence of cognitive disorganization from acute psychosis.  He will not allow testing or physical examination beyond the blood pressure determinations and eventually stop cooperating with that.  The inpatient psychiatrist makes an assessment and swings into action.  He tries to establish rapport with the patient to convince him that this is a medical emergency.  At the same time he has contacted the hospitalist service and there is agreement that he needs to go to an acute care bed as soon as possible.  The hospitalist reminds the psychiatrist that they cannot touch the patient unless he consents "or it is assault".  His advice is to call when he is ready to cooperate or call when he gets obtunded by encephalopathy or a stroke and they will treat him acutely.  In this case the psychiatrist eventually convinces the patient to check out the ICU and walks him over there.  Once he is in the bed - he is fully cooperative with all recommended measures including a complete physical exam.

In addition to the lack of a logical progression to the care of severe medical problems there can also be obstacles at the level of presenting the diagnosis to the patient.  I have presented diagnoses of cancer and diabetes mellitus to patients with psychiatric disorders only to be told that the diagnosis is impossible.  "I could not possibly have diabetes doctor, because I don't have a pancreas." comes to mind.  In terms of evidence, holding up an image of a lung or brain tumor may get the response: "I don't think that is my x-ray doctor.  That is somebody else's x-ray."  Those responses and the lack of ability to cooperate can be very frustrating for primary care physicians and specialists.  These patients are always very easy to get rid of.  All it takes is a comment like: "If you want me to treat you - you are going to have to stop smoking.  If you can't stop smoking - I can't treat you."  At the other end of the spectrum I have had Internists coach me over the phone on what to do for a patient, because they knew the patient would never come into their office or pick up a prescription.

The most frustrating cases are the ones that I saw too late.  The mistakes of informed consent as in:  "Mr. Smith you have colon cancer and need to have surgery to have the tumor removed.  At this stage you have an excellent chance for recovery but we have to operate in the next few months."  Mr Smith has schizophrenia.  He smiles and seems to understand everything.  He just wants to get out of the clinic and away from doctors back to his home where he will be much less anxious.  He never returns until two years later when concerned relatives call the police because of their concern about him.  The police find him alone at home.  The house is in disarray.  There is blood everywhere.  Mr. Smith is emaciated and has lost 35 lbs.  He is brought to the hospital and admitted to inpatient psychiatry.  He is seen by the oncologist who originally consulted on his case.  He now has widely metastatic colon cancer and no chance for survival.  That whole sequence of events can be prevented by a psychiatrist willing to discuss these potential outcomes long before the clinical picture worsens.

An infrastructure that allows for continued outreach and rapport building is also useful.  I had many patients with terminal cancer diagnoses admitted to my inpatient unit, not only because they were mentally ill and medical services would not admit them, but because there was no place else for a mentally ill person with terminal cancer to go.  Pulling all of the necessary resources and teams together with an initial acute admission potentially saves lives.  This paper is a good example of that, but acute care psychiatrists may still be held to the "acute dangerousness" standard for care and these admissions are actively discouraged.    

In the case of the NEJM article, the patient is a 63 year old woman with a history of lung cancer, chronic obstructive pulmonary disease (COPD), hypertension, hypercholesterolemia, asthma, insomnia, and restless leg syndrome.  She has also had two previous cancer diagnoses.  A previous diagnosis of lung cancer had been treated surgically with lobectomy and adjuvant chemotherapy six years earlier.    She had a past history of stage IIa estrogen-receptor positive and progesterone-receptor positive, HER2/neu negative invasive ductal carcinoma of the left breast.   The breast cancer was treated with lumpectomy, whole breast irradiation, and chemotherapy.  The patient had a previous psychiatric diagnosis of Attention Deficit-Hyperactivity Disorder treated at times with various stimulants and modafinil.  She was a admitted to the inpatient psychiatry unit for treatment of depression with electroconvulsive therapy.

In the process her psychiatrist comes up with a list of 13 factors that affect cancer care in patients with severe mental illness and cancer and strategies to approach them.  The factors follow initially what is known about cognitive and social behavioral deficits in the population with severe mental illness.  That would include an inability to understand the diagnosis or treatment.  In many cases, the patient is unable to provide informed consent due to cognitive deficits.  System wide deficits are identified at the level of the provider, the health care system, and society and culture.  Any physician who tries to provide medical or psychiatric care to these populations has seen most of these deficits.

In addition to the factors affecting cancer care there is a separate table of Differential Diagnosis of Depression in a Cancer Patient that every psychiatrist working in these settings needs to be aware of.
      
The psychiatrist in this case provides psychotherapy focused on the patient's understanding of the illness and their decisions to cooperate with care.  That included but was not limited to biological interventions for depression.  ECT and lurasidone were the main identified treatment modalities.  An enhancing mass was noted in the right breast on chest CT scan to follow up on previous cancer treatment.  That was subsequently diagnosed as ductal carcinoma of the  right breast.  In this case, the radiation oncologist modified the radiation treatment to maximize the flexibility of treatment for the patient.  She was able to complete all 5 treatments without difficulty.  

From a psychiatric standpoint she was discharged as improved after 19 days taking lamotrigine, gabapentin, quetiapine, and modafinil.  She had received 6 ECT treatments.  But as importantly, she had follow up oncology care, identification of a new cancer diagnosis, and coordination of that care also occurred in that setting.  This is a very compelling study at that level and a clear departure from the rationed inpatient care that people have come to expect when psychiatric units are run by business people.


George Dawson, MD, DFAPA


References:

1:  Irwin KE, Freudenreich O, Peppercorn J, Taghian AG, Freer PE, Gudewicz TM. Case 30-2016. N Engl J Med. 2016 Sep 29;375(13):1270-81. doi: 10.1056/NEJMcpc1609309. PubMed PMID: 27682037.

2: Shtasel DL, Freudenreich O, Baggett TP. CASE RECORDS of the MASSACHUSETTS GENERAL HOSPITAL. Case 40-2015. A 40-Year-Old Homeless Woman with Headache, Hypertension, and Psychosis. N Engl J Med. 2015 Dec 24;373(26):2563-70. doi: 10.1056/NEJMcpc1405204. PubMed PMID: 26699172.

Supplementary:

From the New England Journal of Medicine earlier this year:

New England Journal of Medicine Discovers Assertive Community Treatment. link

The Reality Of Burprenorphine Therapy

It is increasingly popular for politicians and healthcare businesses to discuss their ideas about how to end the opioid epidemic that they started.  One of the common themes is widespread availability of both buprenorphine maintenance therapy and naloxone opioid antagonist therapy for acute overdoses.  I am certainly not opposed to either and in fact work in an addiction treatment environment where these are two of several medication assisted therapies used to treat addictive disorders.  I am skeptical of the idea that broad prescribing of these therapies in either primary care clinics or some treatment settings will ever occur.  Naloxone will be more readily available because there is a movement to create easy access without a prescription.  That will never happen with buprenorphine.  Last week - an article in JAMA backs up my skepticism (1).

The JAMA article looks at 3234 buprenorphine prescribers in the 7 states with the most buprenorphine prescribers.  In their introduction the authors talk about the policy initiatives to increase the maximum patients per prescriber from 30 to 100 patients after a year.  The average monthly patient census per month varied from 7 - 22 patients and a median monthly patient census of 13 patients.  The duration of treatment episode was 53 days.  This illustrates that the monthly census was well below the allowed limits and the duration of treatment was well below the recommended maintenance guideline of 12 months.  They cite evidence that novice prescribers wanted more access to substance use counselors or other prescribers with more experience as potential limiting factors.

The authors of this article do not offer other explanations for the low rate of buprenorphine prescribing.  I have a few.  I really do not like stigma arguments.  To me stigma seems like an excuse for not being able to overcome societal biases toward a particular problem.  I don't see how you can train to be a physician and not have most of these biases wrung out of you.  With addictions and mental illnesses there may be a stronger bias based on personal experience.  Some physicians may have come from a family where the the father was an alcoholic or a heroin addict living homeless on the street and everybody was used to that idea.  Some physicians may have come from families where the father was still drinking and dying of cirrhosis and the familiy opinion was that he "has a right to drink himself to death" rather than get treatment that he did not want in order to stop drinking.  Other physicians may have come from families where father and his father both had severe alcoholism.  Grandfather drank himself to death by the time he was 50.  Father got treatment for his alcohol problem and was in stable recovery for years.  All of these personal experiences and the reactions to them will affect how a physician approaches alcoholism and addiction.

Those biases are all part the the inevitable decision-making process that leads physicians down specific career paths.  I have lost count of the number of times that another specialist told me that they really liked psychiatry and were considering the residency except for certain features of the field.  A couple of examples include needing to try to predict suicide and aggression and live with the consequences or dealing with a certain diagnostic group like patients with severe personality disorders.  People are less specific about addictions, probably because as medical students and interns we all see the severe effects.  Most of the acute care hospitals where physicians train have 30-50% of their admissions based on the acute effects of alcohol or drug use.  That includes many admissions for acute hepatitis, hepatic encephalopathy from cirrhosis, acute alcohol poisoning, acute overdoses on addictive drugs, and various psychiatric morbidities like delirium and psychosis from the acute effects of addictive drugs.  It is less obvious but addictive drugs and alcohol are also overrepresented as reasons for admission to surgical trauma units and burn units.  Most interns and residents see these effects first hand and develop both short term and long term perspectives on these problems.

This seems like another case of managers and politicians not appreciating the intense interpersonal aspects of medicine.  Physicians are all not foot soldiers just waiting for the next assignment from a policy maker.  Physicians have probably carefully selected the type of practice they want to be in and there are more than the technical aspects of the speciality that were considered.  It takes a unique skill set to treat people with addictions.  Treating and maintaining an opioid addict in treatment long enough with buprenorphine maintenance for them to realize any benefit is a very unique skill.  Being affiliated with other buprenorphine prescribers is also a necessity to provide cross coverage for patients.  Speciality care centers for addiction seem like an idea to me that does not get a lot of consideration.  Trying to run a buprenorphine maintenance program in a practice environment that is rationed to the degree it currently is does not seem feasible to me.  Adding buprenorphine maintenance as just another task for a busy primary care physician practicing primary care medicine is not likely to work.  It should be obvious that these physicians have  more than enough to do right now.

There is a lot more to it than increasing the maximum numbers of opioid addicted patients on buprenorphine maintenance and trying to treat as many people as possible.  The data from this paper illustrates that.  There is also the issue of the preventing the pool of opioid users from increasing while trying to treat those who are currently dependent on these drugs.  That seems like the best long term option to me.

Addressing this complicated problem takes more than a licensed buprenorphine prescribing physician sitting behind a desk who is willing to prescribe it.  It takes better infrastructure including managers who are enlightened enough to get that physician the kind of resources they need to do the work.  I never hear politicians or policymakers talking about that.


George Dawson, MD, DFAPA


Reference:

1: Stein BD, Sorbero M, Dick AW, Pacula RL, Burns RM, Gordon AJ. Physician Capacity to Treat Opioid Use Disorder With Buprenorphine-Assisted Treatment. JAMA. 2016 Sep 20;316(11):1211-1212. doi: 10.1001/jama.2016.10542. PubMed PMID: 27654608.


        

An Excursion into the Psychiatric News - Blurred Lines Between Business and Professional Organizations

Psychiatry Eclipsed

Before anyone says that this is me going off the rails again - consider one small factor.  For over 30 years I have been paying the American Psychiatric Association (APA) significant amounts of money in annual dues.  Last year it was about $935.  During some of those years, I thought it might be useful to also donate to their political action committee and I gave them significantly more money.  All the time, I was expecting something to reverse the inexorable deterioration in the practice environment  and the rationing of mental health services.  During that time, I witnessed first hand the deterioration of psychiatric services in the state of Minnesota to the point that there is now a mandate that county sheriffs have priority in admitting their mentally ill prisoners to state psychiatric hospitals.  The psychiatrists in the state have no say in who gets admitted to these facilities or the severely rationed number of inpatient beds in the state.  The reason for professional organizations as I understand them is to speak for and advance the profession, support its members and advocate policies that benefit the people that interact with the profession.  In the case of psychiatry that is the patients that we treat, their families, and the larger society.  All I have to do is pick up a copy of the Psychiatric News to doubt that these mandates are very relevant anymore.

I will say in advance that in my assessment the APA does a fair job in terms of education and professionalism.  I have criticized them in this area in the past for not keeping the treatment guidelines up to date and relevant.  Subsequent to that there was a new guideline published.  Access to the educational materials is not contained in the membership dues.  A subscription to Psychiatry Online or the CME Journal Focus are additional charges as are CME credits for reading articles in the American Journal of Psychiatry.  There is also a CD version of courses and presentations at the APA Annual Meeting that is available for a significant cost.  The educational and professional materials are definitely available and some of them are first rate - but they do come at a price.

My biggest problem with the APA has been the total lack of rigor in countering the deterioration of the practice environment and in many cases seeming to directly participate in initiatives that are counter to the interests of psychiatrists and their patients.  Thumbing through the September 16, 2016 edition of the Psychiatric News provides some ready examples.

On page 1, there is a story Everett Appointed head of New SAMHSA Office.  The story is all about APA President Elect Anita Everett, MD assuming a new position as chief medical officer at SAMHSA - the lead federal agency for mental health and substance use treatment.  A direct quote from Dr. Everett: "Having a psychiatrist as a member of the leadership team at SAMHSA will enable psychiatrists to join other mental health and public health professionals in guiding the federal component of the nation's behavioral health systems."  My emphasis on behavioral health.  As far as I am concerned SAMHSA is a pro-managed care government bureaucracy - like most of them.  Secondly, there are plenty of psychiatrists out there who have been chief medical officers for managed care companies and I would challenge anyone to tell me why they are necessary and what they have accomplished.  Managed care companies tell psychiatrists what to do.  They are not interested in a reasonable practice environment, reasonable inpatient settings of even professional standards.  They are interested in cheap, rationed care by overworked clinicians.  I don't doubt Dr. Everett's qualifications or good intentions.  I don't think I am going out on a limb too far to say that she is going to be severely restricted by the current bureaucracy with a strong managed care bias.  That is not good for psychiatrists and it certainly is not good for patients.

The other story on page 1 seems worse - Are Psychiatrists prepared for Health Care Reform?  Yes and No.   I really can't think of a more nauseating term in the medical literature than health care reform.  I have been hearing those hot little words for the entire length of my career.  I heard them from the Clintons back in the days when Hillary Clinton headed up the health care reform efforts during the first Clinton presidency.  Some students of the topic like to recall that for one reason or another the initiative worked on by Hillary Clinton was not successful.  I think that depends on the standard.  There certainly was no expected global program, but it did make managed care a household word and set managed care as the predominant bias in all further discussions of health care reform.  Like most history - people seem to have forgotten this and the Clinton administration (and all that followed) as having a strong managed care bias.  The article suggests that psychiatrists need to get on board with the collaborative care model - another managed care rationing technique.  In the span of 3 decades psychiatry has gone from protesting managed care rationing (especially because it affects us and our patients the most) to suggesting you really have to get on board with this.  The usual buzzwords like further workforce development and merit-based payment reforms are evident.  When professional standards are abandoned what is merit-based payment reform? In all likelihood it has to do with rationing techniques rather than quality medical care.  Paragraph after paragraph in this article read like a managed care playbook.  Maybe the only way to see through all of this pro management rhetoric is to have actually worked in one of these systems of care.  Try working in one with a manager who is reimbursed to extract the maximum amount of productivity while not providing resources to physicians in the system.  In that case I believe the management buzz word is creativity.  In a rationed environment there is often an audacious statement about creativity as a solution rather than additional personnel.  Most reasonable people would be shocked at what constitutes merit-based payment or the hold back procedures before you can get to that level.  Just another in a long line of meaningless cliches flowing from health care reform.

As you might imagine I was a little tense and clammy as I went on to page 2.  There I was an editorial piece by APA President Maria A. Oquendo, MD.  It was title Why 'Physician Heal Thyself' Does Not Work.  I was mildly optimistic that she might come to the same conclusion that I have about physician burnout - it is not a disease it is just bad management.  Dr. Oquendo began  with a description of the recent suicides of a psychiatric resident and a medical student.  She presents the epidemiology of physician suicide and suicidal ideation.  She points out for example that suicide is the second leading cause of death for physicians between the ages of 24 and 35.  She discusses the stigma of a psychiatric diagnosis and the gap between problems and who gets treated.  Her solution is self identification of depression and excessive alcohol use.  There seem to be other factors that are operative.  She quotes a six fold jump in PHQ-9 scores during internship - using that as a metric for depression.  I can't help but think how physicians and trainees are more isolated now than ever.  No matter what the setting we had great teams when I was an intern and resident.  We took care of one another and we had attending physicians who cared.  I addressed some of that in my previous burnout article.  Nobody discusses what it is like to train in a managed care and rationed environment today compared with medical care as usual in the past.  During my last stint in a hospital I did not see well developed teams anywhere.  Most of the senior physicians who did a lot of the teaching and tended to view themselves as affiliated with residents had been replaced by hospitalists.  Entire teaching services had been replaced.  Non-medical management has left many medical institutions very arid places with few personnel and limited collegiality.  That is exactly the wrong environment for depressed and stressed physicians.   Training programs everywhere can help residents by making sure they build collegiality and that team factor in all of their rotations.  They need to provide highly motivated faculty who have the interests of trainees in mind as a priority.  The teams I am referring to here are teams of physicians, not teams that contain administrative staff telling physicians what to do.

The article most directly related to managed care hegemony was "Medical Necessity in Psychiatry: Whose Definition Is It Anyway? by Daniel Knoepflmacher, MD.  The title is of course purely rhetorical.  Like many things in medicine today medical necessity has nothing to do with medicine.  It is a pure business definition designed to give the appearance of legitimacy to what is a pure business driven decision.  The decisions are made by people with no appreciation of human biology or its complexity.  They are people who seem to think that a lot of meaningless business metrics somehow apply to the practice of medicine.  At the worst (and most probable) they are simply rationing to make a profit.  I would call them nerds but I really don't think that they are that smart.

In the article, Dr. Knoepflmacher makes that point.  There is not even a standard business definition of medical necessity.  Companies can basically say and do whatever they want.  He traces the history of the term and how various groups define it today.  Interestingly one of the largest managed care companies states that it is for payment purposes only.  He points out the overemphasis on acute or crisis care rather than professional guidelines or standards.  I would argue that in psychiatry, managed care companies do a very poor job of addressing acute care by using only a dangerousness metric.  The term cost effectiveness is incorporated into some of the definitions in the 1960s.  The acclaimed Mental Health Parity and Addiction Equity Act of 2008 lacks any definition of medical necessity or a more useful definition of medical appropriateness.  That may explain why this legislation has had negligible impact.  Dr. Knoepflmacher's thesis can be best summarized in the sentence:

"Without universal medical necessity criteria for mental health care, clinicians and their patients are saddled with a concept highly susceptible to abuse by insurers."

I would take it a step further.  The abuse has been institutionalized at this point.  Clinicians find themselves abused at every fork in the road.  Any time a psychiatrist refills a medication for a colleague or because the treatment setting has changed they are subjected to abusive prior authorization processes that are in place purely to harass physicians into giving up and patients to the point that they are paying out of pocket instead of using the insurance they have paid for.   In that case Congress is directly responsible for erecting two multibillion dollar industries and inserting them between the physician and their patient.  I would also propose a much better limit than arbitrary medical necessity criteria.  It should be apparent that any managed care company can get around legislation and rules that they lobbied to pass.  I propose that physicians recommend a course of treatment to patients and that they are totally removed from the payment process.  No more wasting time with insurance company employee-reviewers.  No more conflict of interest in favor of big business.  The physician recommends treatment.  The insurance company tells the patient if they will pay for it.  Other than civil action by the patient, the only oversight should be a panel of physicians carefully screened for conflict of interest at the state level to mediate disputes (sorry no insurance industry insiders).

Highlighting these four articles creates a portrait of what is wrong with the APA.  Like other professional organizations it has clearly bought into the pro-management zeitgeist that is generally sold by American businesses and government.  The general idea is that there are business managers that know more about what you do and can tell you what to do - irrespective of your professional training and experience.  That idea is a mile wide and an inch deep.  Anyone with middle school analytic skills should have come to the same conclusion as Dr. Knoepflmacher - about 20 years ago.  His article is there now as a necessary reminder that there is a much better way to do things.  Instead of affiliating with these outrageous business practices - they should be actively resisted at every level.  That should include the practice and training environments.  There is nothing worse for physicians and patients than wringing the humanity out of medical practice.

And there is nobody better at doing that than current healthcare business managers.


George Dawson, MD, DFAPA



Attributions:  The graphic is all me.  It is supposed to represent a progressive overlap by government and business interests with the profession.  There are psychiatrists that work in the overlap areas and some who work just in the black and gray zones.  The field is still plodding along as though it is an autonomous profession.  

Is Melatonin A Benign Supplement?

Before I get too carried away the answer to the question is most likely yes - it is a benign supplement.  That is based on all of the studies and reviews I have about melatonin used to treat insomnia and other sleep disorders.  Nobody seems to report any major side effect from even very high doses taken for periods of time.  Those same reviews have a lot of qualifiers.  As an example some will say that doses in the 4-6 mg per night range are unlikely to cause major side effects.  Others will suggest that even though melatonin use is widespread, there have not been any large clinical trials to assess long term safety.  There probably never will be.  When people ask me about the safety of melatonin that is exactly what I tell them.  Individual use of over-the-counter melatonin varies greatly.  The average self-administered dose that I encounter these days is 10 milligrams.  The range of outliers is to the high side.  Nobody uses the 1 mg or 3 mg tablets but there are people using 10-20 mg.

My concerns come from a number of perspectives.  First there is the issue of normal excretion of melatonin and pharmacokinetics.  Second there is the role of melatonin in other systems in the body. And third there is the question of the toxicology of melatonin.  In the case of supplements, the presumed toxicology is often based on the exposure of large numbers of people to the supplement and no alarming or immediate side effects.  Melatonin is ubiquitous in some tissues in the body and this gives the impression that it may be a benign compound.  On the other hand it is a metabolite of 5-HT (serotonin) and this has additional implications.  5-HT is a highly politicised molecule with unquestionable roles in human physiology and pharmacology.

When I start to think about the general approach to this problem - the first thing I look at is what the physiological levels of melatonin might be in both blood and CSF.  The published medical literature is scant. Normal levels are often determined as part of an experimental intervention like peripheral sympathectomy.  In one study (1), the normal diurnal fluctuations were noted with a peak ranges of peak plasma levels varied from 122-660 pmol/L, and the peak CSF levels from 94-355 pmol/L.   In this study the authors demonstrated that bilateral T1-T2 ganglionectomy abolished both the diurnal rhythms and resulted in lower levels of melatonin suggesting that the peripheral sympathetic nervous system has input into melatonin secretion.

Other studies have looked at the effects of melatonin supplementation on plasma and CSF levels.  This is always important because average natural melatonin secretion in humans is about 0.15 mg per day.  As I previously indicated supplementation is many times that.   An early study (2) showed that oral melatonin supplementation led to rapid increases in blood and CSF melatonin.  The authors also compared the ratio of CFS/blood melatonin and noted it was constant before and after supplementation suggesting that blood is the source of CSF melatonin.  Recent studies (4-6) suggest that melatonin is secreted directly into the third ventricle from the pineal recess or that there may be central or peripheral compartments of melatonin for different effects.

There was a recent comprehensive review of the pharmacokinetics of melatonin.  The authors reviewed 22 high quality studies looking at oral or IV  dose ranges of 0.1 mg  to 100 mg.   Cmax (maximum plasma concentration following the dose) ranged from 72.1 (10 ml/h; 0.02 mg, IV) to 101,163 pg/ml (100 mg, oral).  Tmax (time to peak plasma concentration) ranged between 15 (2 mg) and 210 min (10 mg).  T1/2 (half-life) ranged from 28 (0.005 mg, IV) to 126 min (4 mg, oral).  The response to melatonin for insomnia is unpredictable and no dose response curve has been determined. Of course there are a wide range of available melatonin preparations from pharmacies and health food stores.  Even in the research, it is often difficult to say what preparation was used, but the bioavailability is quoted as 15%.

The sleep research on melatonin shows that (as most clinician know) it has weak sleep effects but variability does exist.  Most people will notice very little effect.  Some will claim that they were "knocked out" and strongly sedated even the next day - an effect that is hard to reconcile with the short half-life.  The lack of a dose-response curve makes it difficult to determine dosing but a review in an authoritative sleep text (8) says that most studies use a dose of 4-6 mg and that the timing of the dose (1-3 hours before sleep) may be an important factor.  I have been advised myself by a sleep expert that a 1 mg dose of melatonin 3 hours before sleep may be optimal because it entrains the circadian rhythm of that individual and have seen that work in many cases.  The weak effects on sleep EEG to me suggest a strong placebo component.

 Getting back to the side effect issue - I was surprised to find a few pages on melatonin in last week's New England Journal of Medicine in their Clinical Implications of Basic Science Research section (9).   These authors discuss the implications of the MTNR1B variant as a risk factor for Type 2 diabetes mellitus.  This variant leads to an overexpression of MT2 melatonin receptors on pancreatic beta cells.  Melatonin in this case leads to a G-protein-coupled receptor (GPCR) initiated cascade that inhibits insulin secretion.  They reviewed the epidemiology of diabetes and the facts that Type 2 diabetes is typically a combination of insulin resistance in muscle and fat cells, increased hepatic glucose output and no compensatory insulin release.  They reference an experiment where the daily melatonin (4 mg dose) for 3 months reduced first phase glucose induced insulin secretion in an oral glucose tolerance test compared with baseline before melatonin secretion.  They point out that there has been some epidemiological work that associates shift work with susceptibility to diabetes mellitus.  They conclude that typical doses of melatonin of 4 mg/day or less are not likely to be associated with decreased insulin release and talk about the application of melatonin in jet lag with dosing for a few days.  They also point out in the 3-month long study that overall glucose tolerance was hardly affected and baseline glucose levels returned after 2 hours of dosing.  That may not be as assuring to psychiatrists who are seeing their patients take 5-10 mg indefinitely along with other medications that may cause metabolic syndrome.

To me this is also further evidence of just how crude many of our clinical trials are that look at the issue of metabolic syndrome and diabetes in clinical populations.  Whenever I discuss consent for atypical antipsychotics and the metabolic issues, I always tell people that weight gain and increased appetite is common.  I also tell them that I have seen people develop diabetes from these medications who were rail thin and never gained an ounce.  The 150 risk alleles for Type 2 diabetes mellitus, the thousands of associated intracellular signalling proteins, and the 300 GPCRs expressed in pancreatic beta cells probably has something to do with that.

Biological complexity rarely leads to simple answers.

I will probably encourage a lot of insomniacs from taking melatonin when there has been no clear benefit.  I will probably also discourage the practice of taking high dose (> 5 mg) of melatonin for the same reason.                


George Dawson, MD, DFAPA




References:

1:  Bruce J, Tamarkin L, Riedel C, Markey S, Oldfield E. Sequential cerebrospinal fluid and plasma sampling in humans: 24-hour melatonin measurements in normal subjects and after peripheral sympathectomy. J Clin Endocrinol Metab. 1991 Apr;72(4):819-23. PubMed PMID: 2005207.

2: Young SN, Gauthier S, Kiely ME, Lal S, Brown GM. Effect of oral melatonin administration on melatonin, 5-hydroxyindoleacetic acid, indoleacetic acid, and cyclic nucleotides in human cerebrospinal fluid. Neuroendocrinology. 1984 Jul;39(1):87-92. PubMed PMID: 6205317.

4: Leston J, Harthé C, Mottolese C, Mertens P, Sindou M, Claustrat B. Is pineal melatonin released in the third ventricle in humans? A study in movement disorders. Neurochirurgie. 2015 Apr-Jun;61(2-3):85-9. doi: 10.1016/j.neuchi.2013.04.004. Epub 2014 Jun 26. PubMed PMID: 24975205.

5: Leston J, Harthé C, Brun J, Mottolese C, Mertens P, Sindou M, Claustrat B. Melatonin is released in the third ventricle in humans. A study in movement disorders. Neurosci Lett. 2010 Jan 29;469(3):294-7. doi: 10.1016/j.neulet.2009.12.008. Epub 2009 Dec 11. PubMed PMID: 20004701. 

6: Skinner DC, Malpaux B. High melatonin concentrations in third ventricular cerebrospinal fluid are not due to Galen vein blood recirculating through the choroid plexus. Endocrinology. 1999 Oct;140(10):4399-405. PubMed PMID: 10499491.

7: Harpsøe NG, Andersen LP, Gögenur I, Rosenberg J. Clinical pharmacokinetics of
melatonin: a systematic review. Eur J Clin Pharmacol. 2015 Aug;71(8):901-9. doi: 10.1007/s00228-015-1873-4. Epub 2015 May 27. Review. PubMed PMID: 26008214.

8:  Krystal AD.  Pharmacological Treatment: Other Medications.  in Kryger MH, Roth T, Dement WH.  Principles and Practice of Sleep Medicine, 5th Edition.  Elsevier Saunders, St. Louis, 2011. p. 916-930.

9: Persaud SJ, Jones PM. A Wake-up Call for Type 2 Diabetes? N Engl J Med. 2016 Sep 15;375(11):1090-2. doi: 10.1056/NEJMcibr1607950. PubMed PMID: 27626524. (to full text link).


Attribution:

Serotonin metabolism graphic is from VisiScience.

Hospitalists.....

I was a hospitalist before the word was fashionable.  It was July 1988 and I had just completed a 3 year post residency stint at a community mental health center as part of a public health service scholarship payback.  For one of those years I commuted another 300 miles to keep a community hospital psychiatric unit open.   I headed for the hospital where I did my rotating internship in Internal Medicine, Pediatrics and Neurology.  It was the only real metropolitan inpatient treatment setting I had known at that point.  In my residency program, the interns were split up into two groups and each group worked at one of the major county hospitals in the Twin Cities.  It was a unique setting at that time because psychiatrists provided almost all of the medical coverage.  They had to be able to diagnose and treat a lot of common medical problems, write for all of the patient's medications, attend to acute medical problems and do the appropriate diagnosis and triage.  I had a wide range of medical problems admitted directly to me ranging from gunshot wounds to delirium.  Any psychiatrist working in these conditions realizes that the term "medically stable" is a relative one.  I had many patients admitted to my service with severe medical problems only because they also had a severe psychiatric disorder and were symptomatic at the time.  In many cases I had to rapidly assess them and transfer to medicine or an intensive care setting.

I had excellent back up by consultants and many of them to this day are some of the best physicians I have ever seen.  But they really did not want to hear from me unless I had a very specific probable diagnosis and most of the evaluation was done.  There are not too many places in psychiatry where jobs like that exist anymore.  If anyone asks me about similar positions - I actively discourage them from accepting a similar job.  With this arrangement the work is far too long and all of the medical care is provided for free - psychiatrists do not get any extra credit for it.

In those days there were six of us covering 3- 20 bed wards, five days a week.  The ads for psychiatrists these days often speak of "psychiatric hospitalists" - but every one of them specified no medical coverage.  They also tend to leave out the part that it is basically a rapid triage and discharge position and the job is to either maintain or cooperate with high discharge rates.  The only thing they have in common with the Internists and Family Physicians who have come to be designated as hospitalists is that they work 7 days on and 7 days off.  A schedule that very few people question.

I naturally picked up this week's copy of the New England Journal of Medicine to see what the two perspective pieces on hospitalists (1,2) had to say.  I was also interested because my brother is an Internist and over the years we have discussed the issue at length.  The initial essay by Wachter and Goldman documents the rapid rise of hospitalist care as a medical specialty.  Since 2003 the number of hospitalists has increased 5-fold to 50,000.  That makes hospitalists the largest speciality within Internal Medicine.  They cite the growth of managed care, Medicare DRG payments, and possible evidence as reasons for the growth of the field.  I am always skeptical of the term efficiency especially when it is combined with the term quality.  I guess it is difficult for some people to accept the fact that managed care and Medicare DRG payments are rationing mechanisms that are tied to quality only by the tenuous thread of government and healthcare company rhetoric and advertising.  The other critical question is efficiency for who?  It certainly is more efficient to administer a group of physicians who work 7 days on and 7 days off and happen to all be in the same chain of command.  It is a lot easier to get them to accept the role of rationing care in the interest of the hospital or health care group than the patient's personal physician who may see their part of their role as patient advocacy.

The authors have an interesting take on the deficiencies of the model.  They talk about the 7- days-on, 7-day-off model as implying that during the off period the physician is literally off and suggests that time might be better spent contributing to key institutional programs.  To me - this schedule seems more conducive to burnout and anyone who works it needs the off time to fully recover.  I have never seen a study on the cognitive efficiency during the 7-days-on, but my conversations with hospitalists suggests that by day 6 it starts to plummet.  With hospitalists supplanting specialists and subspecialists as inpatient attendings they suggest that trainees have less exposure to basic and translational science.  Although not stated in the article, the model involves eliminating whole blocks of specialty care.  I worked at a hospital where an entire Neurology service was eliminated by hospitalist care.  When I questioned that decision I was told: "We have an Internist who is interested in strokes."  Changing neurologists from attendings to consultants with hospitalists as the primary physicians for neurological problems changes the entire nature of care.  It also changes the associated nursing care when staff have no ongoing interest in the care of complex neurology patients.  The authors also note that hospitalists do not seem to have focused on investigating common inpatient illnesses.  They suggest possible remedies - but these seem like major problems that will only get worse with the increasing business rather than academic emphasis in medicine.

Gunderman points out that as opposed to the usual delineators of speciality care - patient age, physician skillset and body system hospitalists are delineated only by patient location.  He doesn't make it explicit but what is the relationship between location and his list of putative benefits? Looking at length of stay for example - that could logically follow as a concentrated effort in the location, but is that a clinical effort or an administrative one?  He points out that the increasing number of hospitalists per se,  cannot be taken as evidence of benefit and that perverse incentives exist.  I agree with the most perverse being the low reimbursement incentive for high volume practice. Seeing complex inpatients with a high frequency of initial and discharge assessments may reduce the volume necessary for productivity demands.  When I was a psychiatric hospitalist, this dimension was manipulated in a number of ways.  I was initially told, I was responsible for a set number of inpatient beds.  At some point there was a great deal of pressure for me to start running outpatient clinics because they would be more "interesting" than just seeing inpatients.  I resisted that and had significant leverage because nobody else wanted to do my job.  I eventually did run a Geriatric Psychiatry and Memory Disorder Clinic for many years while continuing inpatient work.  That clinic was eventually closed by administrators because they claimed our productivity was not high enough to work with a nurse.  The neurologist and I needed all of the collateral data that she collected to do our work.  The expectation was that we would see complex dementia patients and do everything that the nurse in our clinic did - so we closed.  In over two decades of political wrangling around inpatient productivity the current consensus is that covering 10-12 inpatient beds is a reasonable approach.  At one point I was covering 20 beds with the help of an excellent physician assistant but at the cost of doing no teaching.

The critical aspect of Gunderman's thesis is his emphasis on the physician-patient relationship exemplified by this sentence:

"The true core of good medicine is not an institution but a relationship - a relationship between two human beings."

He points out that physicians being affiliated with institutions creates significant conflicts of interest,  isolates hospital staff from the rest of the medical community and that naturally leads to less expertise in the entire community.  It also creates the illusion that an institution rather than the relationship is the core of medical care and it is not.  Government-business constructs like Accountable Care Organizations have a similar effect.  I have experienced this first hand many times as I dealt with the iterations of hospitalists consulting on my patients.  In one case I talked with a young hospitalist about a patient with Type 2 diabetes mellitus.  The patient had a trace of renal insufficiency and was on metformin - a medication that is risky in that context.  The hospitalist advised me to call the primary care Internist taking care of the patient because "He has been doing it a long time and probably knows more about it than I do."  In addition to the relationship - there is clear expertise associated with caring for people with multiple complex medical problems for years in an outpatient setting - compared to a few days as an inpatient.  The medical industrial complex does not adequately value that expertise.                             

I think that there is room for hospitalists and psychiatric hospitalists.  They have to be focused on the needs of both the patient and the patient's outpatient physician.  There have to be clear goals for the hospitalization and one of those goals is what the patient's personal physician would like to see accomplished.  Since making the transition to strictly outpatient care - it is clear that the hospitalists no matter who they might be don't have much control over who gets admitted to the hospital and what happens there.  They are having less to say about when a person is discharged.  This is probably more true for psychiatry than medicine and it results in a large number of psychiatric outpatients not being able to access needed care.

And I can't help but notice that inpatient hospital medicine is still a far better resource than inpatient hospital psychiatry.



George Dawson, MD, DFAPA




References:

1:  Wachter RM, Goldman L. Zero to 50,000 - The 20th Anniversary of the Hospitalist. N Engl J Med. 2016 Aug 10. [Epub ahead of print] PubMed PMID:27508924.

2:  Gunderman R. Hospitalists and the Decline of Comprehensive Care. N Engl J Med. 2016 Aug 10. [Epub ahead of print] PubMed PMID: 27509007.