Sunday, December 17, 2017

Less Is Less - A World of Difference Between Psychiatry and Cardiology

I read Lisa Rosenbaum's opinion piece in this week's New England Journal of Medicine (1). She discusses both sides of the rationing coin. On the one hand, we don't want to reduce resources to the point that people do not get necessary care. On the other hand there are forces including financial incentives and the inability of physicians to tolerate the diagnostic uncertainty of not performing the necessary tests that lead to both increased cost and in some cases unnecessary risk to the patient. She provides an example from her personal medical history on forgoing a recommended test with no associated adverse outcome. A lot of the article is written from her perspective as a cardiologist or cardiology fellow. I can recall, the think tank studies from the 1980s suggesting that coronary artery bypass surgery was overutilized. There are many studies that suggest that medical treatment of coronary artery disease provides similar outcomes. Today we hear the same arguments about the treatment of the atrial fibrillation epidemic and the equivalence of rate control versus rhythm control. The options are presented as a coin toss to many patients. But in both cases it is much more than that. Anytime population based averages of care are applied across large populations there will be a significant number of people who do worse than the norm and may have done better with the other option. My concern has always been, the implicit pressure by healthcare companies to make money by exerting pressure in the direction of the least expensive option right up to including no care at all.

Dr. Rosenbaum discusses the "less is more" movement and the Choosing Wisely campaigns to reduce unnecessary care. She discusses the early role of the Dartmouth Atlas in pointing out the lack of correlation between cost of care and outcomes - a notion that has been discredited (2) but it was the mantra of administrators for nearly two decades.  She concludes that these movements resulted in the idea that "less care is better care or that more care is harmful."  She reviews more recent data that higher spending is associated with better outcomes. She includes recent research on unnecessary admissions and how Medicare beneficiaries discharge from the emergency departments (ED) of hospitals with the lowest admission rates were 3.4 times as likely to die with a week than similar patients admitted to hospitals with higher admission rates - even though those same patients were healthier.

She discusses overdiagnosis in cardiology. Unlike psychiatry, cardiology has a considerable array of biochemical markers, electrophysiological studies, and imaging studies that are very useful in the diagnosis and management of their patients. She illustrates the trade offs involved in considering false positives for troponin and how liberalization of the cut-off values leads to better diagnosis and treatment rather than overdiagnosis. In the area where I currently practice, the entire landscape for diagnosing and treating suspected acute coronary syndrome (ACS) has changed significantly. Nobody tries to guess if chest pain has a cardiac origin or not.  Middle-aged patients are generally admitted and tested with troponin levels and an exercise stress test the following morning if the troponins are negative. If the stress test is negative they are sent home. In most acute care metropolitan hospitals there is ample intensive care and telemetry space to accommodate all of these admissions. The cost of that overnight admission to cardiology exceeds the cost of a week long admission to an inpatient psychiatric unit with a psychosis diagnosis.

What is the parallel process on psychiatry? A patient in crisis presenting to an ED of a metro hospital in crisis has no similar guarantee of cautious screening. In the majority of cases they will never see a psychiatrist. In most cases the assessment and screening is done by nonphysicians. In addition, diagnoses and syndromes are generally secondary in the discharge process. The only way that patient gets admitted is dangerousness to self or others. That could be due to an acute intoxication, an emotional overreaction, a mood disorder, a developmental disorder, a neurodegenerative disorder, or a psychosis. The only thing that counts is the dangerousness. There are no biochemical markers or imaging markers of dangerousness. There is significant disagreement in many cases among clinicians, patients, and their families. If a person is admitted either voluntarily or on a legal hold - in any case they will typically find themselves sitting on a psychiatric unit until somebody determines that they are no longer dangerous. Hopefully they will see a psychiatrist and other skilled professionals like trained psychiatric nurses, social workers, and occupational therapists - but there is no guarantee. The issue in an acute dangerousness based psychiatric hospitalization is not a question of overdiagnosis - but whether the patient will get the correct diagnosis and an adequate medical evaluation and discharge plan.  The driving force for that is rationing. The cost of an overnight stay on a cardiology unit with telemetry, blood tests, and an exercise stress test in the morning easily exceeds the payment for complex psychiatric care. I would say that complex psychiatric care is the equivalent to treating a person with a psychosis, extreme mania, or life threatening catatonia or depression. In general we are not worried about the issue of overdiagnosis. People flee psychiatric units if they are given the opportunity and they don't really care if they get diagnosed or not. Psychiatrists cannot present them with an array of options because there aren't any.

When I saw the term overmedicalizing in Dr. Rosenbaum's title - I wondered if she was aware of its Szaszian origins? Szasz was apparently so enthralled by a form of psychiatric treatment that was totally subjective and more akin to a literary critique that he suggested society has an interest in using psychiatry as a way to exert social control over certain subgroups.   The logical conclusion is that mental illness is not a disease and calling something an illness is strictly a power play.  For some reason society and its unholy alliance with psychiatrists is seeking to exert power over a subset of society for unclear reasons. I doubt that Dr. Rosenbaum is using the Szaszian definition. She is probably referring to any number of situations where non-disease is treated as disease.

There are many problems with Szasz - not the least of which is how he would end up treating any number of severe mental conditions. More modern authors on what is and is not a disease seem confused about the imprecise definition, especially in the absence of gross pathology. There is no family member affected by schizophrenia, bipolar disorder, depression, alcoholism, or addiction that doubts for a moment that these are diseases. They generally don't doubt that psychiatrists, at least until very recently were the physicians most interested in treating these problems.

Dr. Rosenbaum's theme does not seem to apply to psychiatric practice. There are no expensive tests to overutilize. Stays on inpatient units are capped by ridiculously short lengths of stay that do not reflect the severity of illness.  Even then - admissions to psychiatric units are generally under the control of emergency physicians.  This is part of the oversimplication, that she referred to. That oversimplification characterized all inpatient stays by diagnosis related groups (DRGs) and suggested that all inpatient stays could be kept to a certain number of days or cost. Nothing else was necessary. This led to three outcomes that led to very subpar care. The first outcome was the deterioration of inpatient services. Rationing does not maximize state of the art care and practically all inpatient units are essentially observation services waiting for people to become less dangerous. Dialogue with patients on acute care units is essentially focused on that issue. Addressing the psychiatric disorder is a consideration only as it applies to dangerousness.  I have had many utilization reviewers tell me that they would no longer pay for the treatment of extremely ill people because they did not seem to be dangerous. 

The second outcome was splitting off addiction treatment. At some point, a large number of detox admissions were directed to psychiatry because medicine units no longer did detoxiification. Then at some point to capitalize on the DRG payments, psychiatric units not longer did detox. patients with addiction were sent from the ED to a county detox unit. The only time they came back is if they experienced seizures or delirium tremens.  The overall rationale is saving the insurance companies money.  They don't cover people at county detox units.

 The third outcome is that patients with severe psychiatric disorders are sent to jail rather than inpatient units. This has resulted in county jails becoming the largest psychiatric hospitals in the United States at a time when psychiatric beds per capita here are among the lowest in the world according to OECD data. All of these changes are associated with a tremendous lack of quality and would be a national embarrassment - if they were not viewed as cost effective by the businesses and governments in charge. The American Psychiatric Association and other district branches still incorporate the cost effective rhetoric when in fact, psychiatry left cost effective in the rear view mirror thirty years ago.

Psychiatrists don't have expensive procedures to order.  In psychiatry less is less (or no) time seeing a psychiatrist.  Less is no time being treated in a medically supervised and therapeutic inpatient or detox unit when you need it.  Less is no psychotherapy that might work for you.  Less is no case management services.  Less is no public health nursing.  Less is not taking the best medication because a pharmaceutical benefit manager says you will have to pay full price for it.  Less in no available child psychiatrist when they are needed.  Less is not getting your blood pressure checked in a public clinic because there are no blood pressure cuffs.

Less in psychiatry is obviously far less than any other speciality.

That brings me to the last concept in the article illusions of value. There is no greater illusion of value than current psychiatric care and that is not because of psychiatrists. To give a clear example, I am an excellent diagnostician - both medical and psychiatric illnesses. I can figure out what is wrong with people and come up with a plan to address all of those issues. I can't do it in a 15 minute appointment. I can't do it if I have to type up all of my encounters like a stenographer or waste my time supporting horrible electronic health record software. In the case of people with severe problems, I can't do it without staff people who can get the patient to the appointment to see me and make sure that the person follows up with all of my recommendations. Without all of that infrastructure on the outpatient side, I will end up seeing about 60% of the people who are scheduled and the average person coming back will tell me they are taking their prescribed medication half of the time. Almost all of that supporting infrastructure has been eliminated in the past 30 years and managed care organizations have set up psychiatric services based on the prescription of a medication. Even if you have a severe problem. Show up 3 or 4 times a year, have the psychiatrist ask you a few questions, and get enough refills until the next appointment.

Psychiatry is actually a paradigm that the rest of medicine should look to in terms of less is less. In her final sentence Rosenbaum describes "less is more" as an aphorism that is "better suited to telling coherent stories than to the complex decisions faced by doctors and patients." I could not agree more. My only qualifier would be that the administrators are always telling their coherent stories that make it seem like they know more than physicians know about medical practice. They do a great job of selling it and convincing people that a symptom checklist and an antidepressant prescription constitutes optimal care. 

That is the only way that the current abysmal psychiatric services offered by large health care corporations could get a pass.

George Dawson, MD, DFAPA


1: Rosenbaum L. The Less-Is-More Crusade - Are We Overmedicalizing or Oversimplifying? N Engl J Med. 2017 Dec 14;377(24):2392-2397. doi: 10.1056/NEJMms1713248. PubMed PMID: 29236644.

2:  Sullivan K.  The rise and decline of the Dartmouth Atlas.  The Health Care Blog, September 25, 2016


  1. The irony is that treating the chronically and persistently mentally ill in jails is far more expensive than treating them correctly (according to the paradigm you so nicely laid out) in a psych hospital. But of course the tax dollars go to the private prison industry in a goodly percentage of cases. And the "patients' rights" groups are somehow not bothered by incarceration of the mentally ill, since they don't really believe the ARE mentally ill. Ignorant voters have to share some of the blame for this.

    1. It is all about the cost shifting. Remember when that was a term to suggest that private insurance would have to adsorb the other costs in the system that Medicare or Medical Assistance did not cover. The real costs shifting is getting the mentally ill patient out of the ED without incurring any additional cost. That can be to county detox, to jail, to a line at the Mission, or even a bus ticket out of town. Anything is fair game to prevent admissions or counter the myth that no more psychiatric beds are necessary. Nothing lost is pure profit for these companies.

      You don't see the problem studied or even considered for study. It would be an easy matter to look at all admission/nonadmissions after 5 years on the Social Security mortality database and look at mortality outcomes. I recall a German study some time ago in the Br. J Psychiatry that did that and the numbers were not very hopeful. There should be a massive nation wide effort to look at the numbers and design interventions to improve them instead of keeping a system in place that maintains a steady state of seriously ill people.

  2. In my opinion, society as it presents today is so characterologically compromised from top to bottom, such Axis 2 pathology is a quantifiable amount of what presents as alleged Axis 1 disorders. This comorbidity, at best, is in fact so hopelessly and pervasively confounding that it defies the "standards of care" that do not account for this role in symptomatology. There are several etiologies to how this came to be, but, for me at least, the primary factor was enduring the pervasive effort of managed care in the late 1980s to the early 2000s to eliminate any ability to diagnose and be reimbursed to treating legitimate Axis 2 issues to the patient presentation.

    And now the consequences have come home to bite us all in private parts. It is hilarious and hideous simultaneously to watch all these "treatment resistance cases" be overdiagnosed and overtreated when biology is the last intervention for personality disordered presentations.

    So, the less is more crusade, how laughable that is to me as a 7 year Locum Doc watching all my colleagues just dig a ditch instead of molding a strong foundation of recovery for such patients.

    Let's have a moment of brutal candor, what percentage of patients in CMHC clinics are either addicts from the community, or now addicts from the treatment process. Sorry to write here Dr D, but it is approaching 50% in my travels, if not more.

    We are all lemmings in some form, whether in the office, our neighborhoods, our political travels, or navigating the Web. Again, hilarious and hideous simultaneously...

    But, who is really laughing? Not me.

    1. Joel - I couldn't agree with you more about the conflation of Axis I and II! Bipolar spectrum, anyone?

  3. James O'Brien, M.D.December 21, 2017 at 7:53 AM

    It can always get worse. Collabo-care is the next phase in the destruction of private practice, cheered on by many in the academic psychiatry community. 15 minute med checks are simply not demeaning enough top down control freaks.

    If the future of psychiatry is helping a NP hand out scripts for SSRIs that have an NNT of 10-14 when you know a hell of a lot more than 1/10 patients will develop side effects that lead to discontinuation, I fail to see how that assembly line of nothing for people you never see is rewarding to either patient or doctor.

    I still maintain that if this is the future of psychiatry, interested medical students should do a flex internship then earn a Ph.D. in psychology, skipping residency. It does not take three years to learn psychopharm, and the reality is most psychotropics are handed out by doctors and surrogates who have only a superficial understanding of the meds they are giving. I no longer shake my head at patients who are given Xanax and Adderall at the same time for something that is not GAD or ADHD.

    I totally agree with Dr. H about Axis 2, and all of these developments also reflect the Cluster B (mostly narcissistic and sociopathic) takeover of healthcare by aggressive admins whose only real talent is smelling fear and weakness in doctors.

    1. If there are no significant changes I agree with you on the collabo care scenario. I think that MCOs are currently strained by the fact that psychiatrists won't work for them, hence all of the notifications that you are now a designated provider if you happen to work for any business that has a contract with them. It makes it seem like these organizations have a massive pool of psychiatrists. In fact most psychiatrists when called will tell patients they did not even know they were on a treatment panel.

      Only legislation that forces all MDs to work for these organizations can result in the model of permanently dwarfed psychiatry.

      On the issue of the admins responsible for the takeover. There are so many of them - I don't think they are inherently PDOed. At this point it is just easy work. The true evil geniuses were the people who set this up in the first place. Plenty of evidence that they resented physicians and used standard American business practices to take over and dominate medicine. Standard manipulation and leveraging with no accountability to science or ethics.

  4. To tie in this post with your last Dr D, how much of psychological distress is related to under-diagnosed if not just flagrant mis-diagnosed somatic disorders that are not effectively worked up per the endless intrusions of now Obamacare to minimize responsible and appropriate somatic assessment?

    I still see cardiac "issues" in the over 50 y.o. crowd that are "ruled out", but yet, if one is lucky enough to get records (and that is really a joke with the lack of coordination between EMRs by different providers, await the "hit #2 if your EMR doesn't speak analogue code" voice message), even without a sophistication with cardiac evals, it can appear a bit obvious the work up was a tad limited because the patient had psychological complaints.

    And yet, still haven't seen SSRIs, or even Benzos, treat A fib effectively...

    Oh, and Diabetes, geez, how many of us providers have seen patients come into our offices and be more pleasant and comfortable because their DM mgmt has been effective?

    Again, the real sabotage has been by the APA and DSM 5, eliminating the Axial 1-5 system, you folks want to talk about real collusion crippling the nation, yeah, let's start there!!!

    Happy Holidays to you and readers here, thank god I have off for 5 days total...

    Joel Hassman, MD

  5. "Happy Holidays to you and readers here, thank god I have off for 5 days total..."

    Happy Holiday to you as well. I also have my first 2 week block of time off in the past 8 years. I am usually scrambling for a day here and there based on the call schedule.

    Agree with you about medical diagnoses. Plenty are missed and plenty are unaddressed for one reason or another. Hence my posts about CPAP/APAP - most of the patients I see get inadequate follow up, see it as an option and not a long term danger to their health, and have their insurance companies actively working against them (if you haven't used the machine enough we want it back rather then let us help you use it correctly). Hypertension is another issue. When I do BP checks and see the patient is still hypertensive and taking starting doses of antihypertensives my response is:

    "You have to be really aggressive with your clinic/doctor on this blood pressure issue. Get a home device, check your blood pressure every day, bring those measurements into the appointment, and tell him/her that you know the current guidelines and you want your blood pressure in that range."

    I end up giving people a long list of free medical advice that is not in the E&M bullet points, but what can you do? You cannot treat people for depression when they are having physical illnesses and are actively symptomatic.

    I am less enamored with the multiaxial system. I just list everything and prioritize the less stable conditions.

  6. I get your perspective about listing everything, but, if you document at least in the evaluation all the issues including the psychosocioeconomic stressors in Axis 4, and what is the presenting GAF you can best apply within 10 points, it is on the record and gives a baseline. Plus, listing stressors reminds you and anyone else reading it who is a provider to ask, "so, what about these marital/ occupational/ financial/ legal issues that played some role in coming into treatment?"

    You do it, but, with the Axial system gone, who goes that extra mile these days???

    Amazing how my blog logo continues to be applicable: you can't medicate life!

    And let the Festivus for the rest-of-us commence!

    Joel H

  7. "On the issue of the admins responsible for the takeover. There are so many of them - I don't think they are inherently PDOed. At this point it is just easy work."

    Based on my small n sample on the phone in the past couple weeks, I'm going with disorder not traits. I would like to get MCMIs.