Saturday, February 22, 2020

The Best Advice I Got In Medical School

It has been a long time but when you get to be an old man you can obsess about what you currently know, what you used to know, and how you got here.  I got some life-changing advice as an undergrad but not much good advice in medical school or residency. I can say without a doubt, the best advice I got was:

“If you are sure you are going into psychiatry, take as many medical electives as you can. Don’t take any psychiatry electives because you will be doing that for the rest of your life.”

I did not have to think too much about it because I enjoyed most aspects of medical education and training. The only two negative rotations I had in my training were based primarily on the staffing patterns at the time and they were not major medical or surgical rotations. They also did not seem to be very interesting. Practically all of the medical and surgical residents I worked with were outstanding in many ways. I felt like an integral part of the team and I was happy to do the necessary work.

As a result of the advice I took endocrinology, cardiology, renal medicine, allergy and immunology, neurology, infectious disease, and neurosurgery in addition to the required general medicine rotations. I took a little flak from the Dean. There was some concern that there were not that many spaces available in medical electives. At one point it was suggested that I should limit myself to two or three medical electives. I prevailed and got what I wanted largely because the specific rotations were at a public hospital and the local VA hospital. 

One of the aspects of medical training that is not discussed enough is camaraderie. When you are a medical student, your role is often ill-defined. The role generally depends on the staff you are working with, the institution, and the general culture within the medical school. At the hospitals where I spent most of my time medical students were an integral part of the team. On day one – you are assigned patients and admissions. You were expected to report on patient progress and write progress notes. You learn communicate with everybody in the hierarchy ranging from the intern to the resident to the attending physician. You are supposed to learn how to research and study the specific problems that your patients had and in some cases do a special report. Examples would include a chart review I did on gram-negative meningitis at the VA medical center and presentation on anaphylaxis on my allergy rotation at Milwaukee County Hospital. Both of those studies went extremely well.

But camaraderie is more than knowing the chain of command, hospital systems, and how to get the work done. A key component is the educational quest that everyone is on. Doing rounds with five or six different people at all levels of training ranging from novice to world expert is experience that you don’t get into many places. Some of the results can be stunning. I did a consult on a patient with possible spontaneous bacterial peritonitis (SBP). I wrote up the consult form and prepared to present to the attending physician that afternoon. When he walked in the room from about 10 feet away, he asked everyone else on the team what the problem was with the patient’s leg. I had been focused on abdominal, systemic, and laboratory findings. Nobody could answer the question. The attending physician who happened to be an expert in streptococcal infections, pointed to a rosy rash on the patient’s left shin and suggested that it was a form of streptococcal cellulitis. He did the necessary tests to confirm that diagnosis at his lab.  One of the many processes that must be attended to in these rounds is the pattern matching aspects of diagnosis. It was vaguely implicit in my training and I realized only later when teaching a course in avoiding diagnostic errors - that these rounds are the place to ask experts: “What are you seeing that nobody else is?” All experts including psychiatrists recognize certain patterns and can make more rapid and more accurate diagnoses than people outside their specialty.

A lot of people reading this may have a hard time believing that what you learned in medical school is relevant to a specialty that you practice your entire life. After all the aren’t these specialties updated at some point and doesn’t your knowledge base become dated? It is surprising how the basic approach to the patient that is unique to each specialty does not change much. There is still relevant review of systems, specialty specific diagnoses, and laboratory testing. Working with specialists for even a month gives medical students and residents a clear idea of how to approach patient problems in a systematic manner. Even though there have been radical changes in some specialties like cardiology, most medical specialties change slowly at the mechanistic level typically with some pharmacological innovation. A clear example relevant to psychiatrists is the endocrinology of metabolic syndrome and diabetes mellitus. Over the course of my career that has resulted in increasingly complex pharmacotherapy ranging from insulin, metformin, and sulfonylureas to an additional five classes of drugs and more complicated insulin preparations.

A unifying concept that I learned on all those medicine specialty rotations is that it is important to still know about these mechanisms and medications even if your specialty involves another bodily system and you are prescribing an additional treatment. No matter what specialty service I was on there was never the idea that we could focus only on a specific bodily system and ignore the rest. On all of those rotations including neurosurgery, I was often the person focused on what was going on with the patient’s brain.

Learning medicine and neurosurgery on all of these rotations was quite exciting. I am much more likely to retain information if I am excited about it. I was excited right up until 11 PM on the last day of medical school.  I was doing renal medicine at the time and the senior resident was going to be a rheumatology fellow. We finished rounding about 6 PM and he noticed we had 4 or 5 additional consults. He was the kind of guy that you really like working with. He had a great sense of humor and was always engaging. He could even engage an introvert like me. I remember him saying: “Look I know - this is your last day but you could really help us out by doing some of these these consults. The new team is coming in tomorrow and I don’t want to leave all of these consults behind.” He threw in a couple of politically incorrect jokes for good measure and I headed off to do two consults. We came back and met with the attending physician who was considerably older than I am right now and finished them all by 11 PM. I really did not want to say goodbye to that team. But I headed off by foot across the golf course sized county hospital grounds to my apartment on 89th St.

The knowledge gained in that fourth year of medical school was a springboard for the next 30 years. I continue to read about all those medical specialties and remember what happened in 1982. I continue to research all the medical problems and medicines that my patients are taking. I continue to wish at times that I was still on that renal medicine team back at Milwaukee County Hospital.

I didn’t get a lot of good advice in medical school but for all those reasons the advice about what to do in my fourth year was the best.

George Dawson, MD, DFAPA

Sunday, February 16, 2020

Medication Checklist-Current Version

I have posted past medication checklists on this blog in the link below will take you to the current version. I developed this over the last 10 years seeing outpatients who have been treated with various psychiatric medications over the previous 5 to 50 years. During a comprehensive evaluation a history of past psychiatric treatment including hospitalizations, past medications, past psychotherapy, and other biological therapies needs to be discussed. Ideally a patient will recall ineffective therapy from the past that can just be restarted. In my current practice setting that is relatively rare. People with chronic mood disorders, anxiety disorders, and insomnia ever happened men treated with multiple psychotherapies and medications. They are typically seeing me because those past therapies have not worked. It is up to me to come up with a newer and safe approach based on my past history.

There are varied responses to the question about what medications have been prescribed in the past. If it has been a long time since the last episode of treatment many people say they can’t recall the name of the medication at all. In some cases people admit that they never really studied the name of the medication, they just took it out of the bottle at the correct time. There are some people who will get a month-long prescription from a physician and never take a single pill. When people have been treated with multiple medications the responses are more varied like “I have taken all of them”, “I have taken all of the SSRIs”, or “I’ve taken all the SSRIs and SNRIs”. Closer examination often shows that many people take anywhere from 3 to 5 antidepressants over any 20-year span. There are people have taken the same antidepressant for 20 or 30 years and wanted it to be changed.

There are many other questions pertaining to best use of medications including diagnostic clarifications. The commonest problem I see is people misdiagnosed as having bipolar disorder and then not treated in a standard way for bipolar disorder. There are also people who have bipolar disorder who do not receive standard treatment. In this era of direct to consumer advertising, many people are treated with aripiprazole or brexpiprazole who would not have been treated with dopamine receptor blocking agents in the past. That opens up an entire new category of potential side effects and comorbidities.

All these reasons make the history of medication use extremely useful in a psychiatric evaluation.  Formal versions of medication history such as the Antidepressant History Treatment Form (ATHF) have been used in research for 20 years as a standard way to document whether or not a patient has received an adequate trial of an antidepressant (dose x duration). This form generally requires collecting a lot of collateral information especially with regard to the dose. Checklist approach I am using is focused on getting the general name and class of the medication. If I think additional information is required I will try to get the necessary collateral information. But generally I am looking for class effects, especially if it is apparent that the patient cannot tolerate a particular group of medications.

Over the years I have been using it this form has been useful. It is essentially like the memory testing paradigm where you proceed from spontaneous recall to categorical recall to list prompts. This is definitely a list prompt but for much larger universe. That is often why I direct people to focus on a particular section of the list. As an example if I am seeing a young person and they have only taken one or two antidepressants and cannot recall the names, I advise them that they can probably find it in the top half of the antidepressant column. The clinical problem can be managed to subsections of the list.

A disclaimer is in order. This list is for the purpose of discussion among clinicians and possible scientific use. It is not been validated from a psychometric perspective. It is not commercial or for-profit like everything else on this blog. There is no guarantee that it will cue accurate memories of past medications. Like everything in clinical psychiatry, collateral information-in this case from pharmacies is the gold standard. Apart from its clinical use, this list can also function to illustrate the universe of medications that are applied to psychiatric disorders. There can be useful for trainees and since this list is updated, anyone who wants to take a look at current FDA approved medications. The list is generally compiled for convenience. I wanted all the medications to be listed on a single sheet - front and back. I wanted the list to be easily readable even by geriatric patients. I wanted the list to be disposable - in many cases the patient wants to take it with them in order to do future research.

That brings me to the topic of research. Most medical centers have been are large electronic health records for about 20 years at this point. Those EHRs vary significantly in their research capabilities. The obvious study of a list like this would be to see how accurate patients can recall their medication history spontaneously and with this list or a something similar. Optimal membership on the form can also be debated. I eliminated tiagabine from the anticonvulsant section as a misadventure in the treatment of anxiety from about 20 years ago. I did that in order to make room for beta-blockers and orexin antagonists. A colleague pointed out that I don’t have the old amitriptyline-chlordiazepoxide or amitriptyline-perphenazine combination medications. I started practicing over 35 years ago it was extremely common to see those medications being prescribed and thankfully that does not happen anymore. Some of the older medications on the list are of historical interest but also because older patients may have taken them.

The list can be downloaded from the link below. Let me know what you think in the comments section. Please restrict those comments to the utility of this list.

George Dawson, MD, DFAPA


1: Sackeim HA. The definition and meaning of treatment-resistant depression. J Clin Psychiatry. 2001;62 Suppl 16:10-7. Review. PubMed PMID: 11480879. Full Text

Medication Checklist:


Monday, February 3, 2020

Adventures In Vaccine Reactions

This area of erythema appeared on day 3 - only on the arm where the Pneumovax was placed

This post is about vaccinations for old people.  It is a mix of science and anecdotes because that is all there is out there right now. A lot of the information I elicited by posting my experience getting the Shingrix Zoster Vaccine and the Pneumovax 23 Vaccine. It illustrates the concept of biological heterogeneity that I post about on this blog. Many people can seem confused about that, particularly the idea that practically every biological system in the human body has significant heterogeneity and the immune system is no exception.

I don’t work on Fridays anymore so last Friday afternoon I headed down to a primary care clinic to get some problems checked out. The primary care doctor was clearly knowledgeable, reassured me that I had no major problems, and asked me if there is anything else that he could do for me. At that point the conversation went something like this:

Me:  “Well I keep being told every other visit that I need a second Pneumovax vaccination and I need the second Shingrix vaccination. Is it possible get those today?”

MD:  “That should not be a problem”

Me:  “Is there any problem with getting them on the same day?”

MD:  “No you can get them both on the same day.”

At that point he pulled up the EHR and verified that I had a Pneumococcal Conjugate Vaccine (PCV13) on 1/11/2016 and a Pneumococcal polysaccharide vaccine (PPSV23) on 9/8/2010 and a previous Shingrix vaccine on 10/30/2019. He explained the rationale for the second PPSV23 vaccination.

MD: “Did you have any reactions to the first vaccinations?”

Me:  “Just a little pain at the injection site but nothing major.”

The nurse came in the room and asked me if I wanted the shots in the same arm or different arms.  She gave me the Shingrix injection in the left arm and the PPSV23 injection the right arm. I noted the time was 2 PM.  I have a history of anaphylactic reactions and an anaphylactic reaction to a second dose of anti-rabies duck embryo vaccine when I was in the Peace Corps. I forgot to bring an EpiPen, but there is a coffee shop just down the street. I spent an uneventful 40 minutes there (to get through that window) and then headed home.

About six hours later I started to get more intense muscle pain in both arms that eventually extended into the back and down the back. A short time later I started to get a headache. By 9 o’clock that night I was taking acetaminophen for those symptoms. I also noted that I was feeling physically ill and fatigued. The next morning I did not feel any better and continued to take the acetaminophen and added naproxen.  By 1 o’clock the next day - nearly 24 hours after the injection I started to get intense chills. It took my temperature and it was normal. I had put on outdoor clothing and sat next to the fireplace. I was shaking. It reminded me of when I had malaria back in 1975. In those days I had a cheap sleeping bag and crawled into that but eventually crawled across the floor into a tub of hot - water dragging the sleeping bag with me.  I had better resources now. I developed some tachycardia and felt very physically ill until about the 60 hour mark. At that point the chills stopped and the tachycardia resolved.

In the meantime I had solicited a number of medical and nonmedical opinions. One of the best internists I know told me that he advises patients take Shingrix vaccination but that they should plan on being out of commission for 2 to 3 days. He also does not recommend any other vaccinations with it because of the severity of the reaction. Several other primary care physicians have given me similar advice. I posted my experience on a listserv for psychiatrists and my Facebook site and several people had similar experiences. The physical illness caused by the immunizations was intense enough to take sick leave from work.

When I left the doctor’s office I was given “Vaccine Information Statements” that were both CDC documents.  One was entitled “Pneumococcal PolysaccharideVaccine - What You Need to Know” and the other was “Recombinant Zoster (Shingles)Vaccine: What You Need to Know”.  For the shingles vaccine the risk of reaction to the vaccine was listed as mild to moderate arm pain in 80% of people and side effects that prevented 1 of 6 people from doing regular activities that included fatigue, muscle pain, headache, shivering, fever, stomach pain, and nausea. The side effects were supposed to resolve in 2 or 3 days. By comparison the pneumococcal vaccine states that “less than 1/100 people develop a fever, muscle aches, or more severe local reactions”.

At a theoretical level it is interesting to consider the differences between both vaccines. In order to generate an immunological response, vaccinations need to create inflammation at the injection site. That typically requires an adjuvant. Adjuvants like aluminum compounds result in local cell death that leads to release of IL-1 family cytokines and danger-/damage associated molecular patterns (DAMPS). That in turn leads to T cell response and immunity (3). 

The adjuvants for these vaccines are significantly different.  The PPSV23 adjuvant is a standard alum based.  The Shingrix adjuvant is an ASO1B that is described as “ f 3-O-desacyl-4’-monophosphoryl lipid A (MPL) from Salmonella minnesota and QS-21, a saponin purified from plant extract Quillaja saponaria Molina, combined in a liposomal formulation. The liposomes are composed of dioleoyl phosphatidylcholine (DOPC) and cholesterol in phosphate-buffered saline solution containing disodium phosphate anhydrous, potassium dihydrogen phosphate, sodium chloride, and water for injection.” By comparison Zostavax the original shingles vaccine was live attenuated varicella-zoster virus without an adjuvant.  The Zostavax vaccination is much less effective, leading the CDC to recommend both doses of Shingrix, even if there is a significant non-allergic reaction to the first injection. 
Flu-like symptoms are an area great interest to me into a large extent they are cytokine mediated.  The top flow sheet at this post gives common cytokines that can cause flu like illness.  Given the complexity and sheer number of cytokines the scope of inflammatory cytokines triggering these reactions is not known.  Of the known cytokines from the Shingrix inflammatory reaction (4) it is likely that IFN- γ is a cause. The other interesting aspect of this response is that it is hypothalamically mediated.  The preoptic anterior hypothalamic area (POAH) has networks for heat production (shivering) and dissipation (vasodilation) (5) and both were affected in my situation. This area is described as being sensitive to a number of compounds including cytokines.  Other  cytokines may explain the associated flushing and sweating.

At the time this was posted I have been seen in the Urgency Room last night because my health plan was concerned about tachycardia and flushing.  The ED doc thought it was all a post vaccine reaction and did not do any further testing.  He said to come back if I got a rash, bloody urine, or a temp of 100.5.  I had the chills for about 8 hours at that point. I went to bed at midnight and woke up at 3AM sweating.  By 7AM, the tachycardia and flushed sensation and appearance were gone. Things were going well until this afternoon when I noticed the redness on the arm where I had received the Pneumovax (see photo).  I was seen in Urgent Care this time.  They had an alternate explanation for the inflammatory response, if the needle is withdrawn too quickly some of the vaccine and adjuvant gets deposited in subcutaneous tissue and leads to a broad inflammatory response.  They recommended an antibiotic (cephalexin) and prednisone just in case. I declined the prednisone and could not get the antibiotic filled anyway.  The pharmacy I use closes at 7PM and I got out of Urgent Care at 7:07.

What are the lessons from this adventure in vaccinations?  First off, as I have said repeatedly on this blog there are no guarantees in medicine. Everything is a probability statement.  In the case of Shingrix, I am taking a vaccination that may be 90% effective in preventing shingles but the trade off is that in clinical trials as many as 1 person out of 6 gets disabling side effects for at least 2-3 days. I have heard from some people where it lasts for 10 days.  Second, even though those are significant side effects shingles is potentially far, far worse. In the people I have treated, that includes months of disabling neuropathic pain, eye complications, facial nerve complications, and in some cases extensive testing just to find out what the problem was.  As an example, a 79 year old man had extensive testing for severe abdominal pain that was thought to be a malignancy that was ultimately diagnosed as shingles.  In the case of Pneumovax it is pneumonia, meningitis, sepsis and death. Case closed.  My memorable case there was an eldelry patient who presented to the emergency department with "agitation".  As the intern on Neurology I was called to see her.  By the time I arrived she was unresponsive.  Upon examination she had pus running out of her ear and nuchal rigidity. A lumbar puncture showed pneumococcal meningitis.  She survived a complicated course including acute respiratory distress syndrome (ARDS) but became completely deaf from the meningitis. Third, it is probably reasonable to not get both vaccinations at once. The physicians writing to me at this point have all been affected by complications they have seen in their patients who got Shingrix and they have seen it as a vaccine that has more than the ordinary amount of adverse effects compared with typical adult vaccinations.  Finally, what about the issue of repeat Pneumovax vaccinations?  It only comes up in cases where a person has COPD or asthma (me) and gets a vaccination before the age of 65 on that basis. Clearly, the arm rash here seems to indicate more of an inflammatory response on the Pneumovax side than the Shingrix in my case. Was it because of a robust antibody response after the initial vaccination?  I don’t have the answer to that question but if there are any immunologists reading this post – I would be very interested in your comments.

That is it for now.  Hopefully I am heading into work tomorrow morning.

George Dawson, MD, DFAPA


1.  SHINGRIX (Zoster Vaccine Recombinant, Adjuvanted), suspension for intramuscular injection.  FDA Package Insert

2.  PNEUMOVAX® 23 (pneumococcal vaccine polyvalent) Sterile, Liquid Vaccine for Intramuscular or Subcutaneous Injection.  FDA Package Insert

3.  Muñoz-Wolf N, Lavelle EC. A Guide to IL-1 family cytokines in adjuvanticity. FEBS J. 2018 Jul;285(13):2377-2401. doi: 10.1111/febs.14467. Epub 2018 May 3. Review. PubMed PMID: 29656546.

4. Cunningham AL, Heineman TC, Lal H, Godeaux O, Chlibek R, Hwang SJ, McElhaney
JE, Vesikari T, Andrews C, Choi WS, Esen M, Ikematsu H, Choma MK, Pauksens K, Ravault S, Salaun B, Schwarz TF, Smetana J, Abeele CV, Van den Steen P, Vastiau I, Weckx LY, Levin MJ; ZOE-50/70 Study Group. Immune Responses to a Recombinant Glycoprotein E Herpes Zoster Vaccine in Adults Aged 50 Years or Older. J Infect Dis. 2018 May 5;217(11):1750-1760. doi: 10.1093/infdis/jiy095. PubMed PMID:29529222.

5. Swaab DF. Autonomic Disorders in Handbook of Clinical Neurology, Vol. 80
(3rd Series Vol. 2) The Human Hypothalamus: Basic and Clinical Aspects, Part II,
2004, Elsevier, Amsterdam, pp351-370.

Supplementary 1:

Did not make it into work as expected on February 4. Went to see my primary care MD instead.  The erythema (redness) had not gone down and actually extended farther down the arm (I drew a line around it the night before).  No systemic symptoms but more redness.  Needed to know if the antibiotic needed to be changed and if prednisone was a good idea.

His conclusion - not an infection but inflammation from a reaction to the vaccine.  He has only seen a couple of reactions like this with Pneumovax and pointed out that it is given to immunocompromised people every ten years starting at a younger age.  He said to keep taking the cephalexin for a week but no prednisone: "I give prednisone to people with life threatening problems - not a red arm.  More medicine is not necessarily good medicine."

Work tomorrow. 

Supplementary 2:

New CDC Guidelines on Pneumococcal Vaccinations:

New Pneumococcal Vaccine Recommendations for Adults Aged ≥65 Years Old

PCV13. PCV13 vaccination is no longer routinely recommended for all adults aged ≥65 years.
Instead, shared clinical decision-making for PCV13 use is recommended for persons aged ≥65 years
who do not have an immunocompromising condition, CSF leak, or cochlear implant and who have
not previously received PCV13 (Table 1).

CDC guidance for shared clinical decision-making. When patients and vaccine providers engage
in shared clinical decision-making for PCV13 use to determine whether PCV13 is right for the specific individual aged ≥65 years, considerations may include the individual patient’s risk for exposure to PCV13 serotypes and the risk for pneumococcal disease for that person as a result of underlying medical conditions (Box).

If a decision to administer PCV13 is made, it should be administered before PPSV23 (5).
The recommended intervals between pneumococcal vaccines remain unchanged for adults without
an immunocompromising condition, CSF leak, or cochlear implant (≥1 year between
pneumococcal vaccines, regardless of the order in which they were received) (5). PCV13 and PPSV23 should not be coadministered.

ACIP continues to recommend PCV13 in series with PPSV23 for adults aged ≥19 years
(including those aged ≥65 years) with immunocompromising conditions, CSF leaks, or
cochlear implants (Table 1) (2).

PPSV23 for adults aged ≥65 years. ACIP continues to recommend that all adults aged ≥65 years
receive 1 dose of PPSV23. A single dose of PPSV23 is recommended for routine use among all adults aged ≥65 years (1). PPSV23 contains 12 serotypes in common with PCV13 and an additional
1 serotypes for which there are no indirect effects from PCV13 use in children. The additional
11 serotypes account for 32%–37% of IPD among adults aged ≥65 years (22). Adults aged ≥65 years
who received ≥1 dose of PPSV23 before age 65 years should receive 1 additional dose of PPSV23 at
age ≥65 years (2), at least 5 years after the previous PPSV23 dose (Table 1) (5).

Note:  In the above case of the vaccine reaction I received 1 dose of PPSV23 before the PCV13 and one dose after by a period of 4 years.  That is not consistent with the guideline but they  were no coadministered and there was a period of at least one year between injections (PPSV23 in 2010, PCV13 in 2016, and PPSV23 in 2020). 

Thursday, January 30, 2020

Warren v. Dinter - More Implications Than Malpractice Law

I have few comments about this case that is written about from the perspective of malpractice law and the implications for informal consultation.  In this case, Richard Dinter, MD was a hospitalist employed by one of the main hospital systems in Minnesota.  By way of background, over the past 20 years most hospitals in the United States are now staffed by hospitalists who are generally internal medicine or family medicine trained specialists dedicated to admitting people to hospitals, treating them while they are there, and discharging them. In the overall history of medicine this is a fairly recent movement.  Before hospitalists, this function in most hospitals were covered by rotating assignments of staff from these departments or in the case of some private practices - the practitioners would see anyone from their practice admitted in the AM and see their clinic patients in the afternoon.  The hospitalist movement seems to have developed to create more efficiency in the system but an undoubted byproduct has been less personal care by physicians who don't know the patient as well, more control over physician practice by business entities, and an increase in the amount of care delivered by proxies or physician extenders.

In this case Susan Warren presented to a clinic in Hibbing Minnesota with abdominal pain, fevers, and chills. She had an elevated white blood cell count. The nurse practitioner seeing her called the hospital system where the call was randomly assigned to Dr. Dinter. The question was whether or not the patient should be hospitalized.  Hospitalization was not recommended and the nurse practitioner did not seek the hospitalization on her own. The patient was discharged home and died of sepsis. The family sued both the physician and nurse practitioner.  In Minnesota nurse practitioners can practice independently without physician supervision.  Reversing prior precedents, the Minnesota Supreme Court ruled that a physician-patient relationship was not necessary for a medical malpractice claim. Instead they established a new standard of reasonable foreseeable action. In other words, a malpractice claim could be initiated if harm suffered was a “reasonably foreseeable action” of a physician’s actions.

Amici curiae briefs were filed by the Minnesota Medical Association partnering with the American Medical Association and separately by the Minnesota Hospital Association. The main arguments were that this standard would stifle collaboration and informal consultation and that might possibly lead to patient harm. The most shocking result of this court decision is that a physician can be sued without any formal physician-patient relationship.  On the surface it all seems to be a standard malpractice issue complicated to some degree by new treatment relationships. But are there larger issues here. The fact that the Minnesota Hospital Association filed an amicus brief suggests that there are.

Minnesota has had the highest penetration of managed care organizations of any state for a long time. Practically all of the care the state is managed by three major healthcare companies who dominate the market. They have their own standards organization that sets guidelines and standards for certain practices in the state. A good example would be the Diamond Project that led to the state government adopting a standard that they would collect depression rating scale scores as a proxy for the quality of depression care being delivered in primary care settings in the state.  This was an early form of “collaborative care” that is currently being pushed by several organizations. In collaborative care for psychiatric purposes, a psychiatrist reviews charts, rating scales, case manager notes, and possibly primary care physician notes and makes a recommendation for a specific treatment. The original Diamond Project focused only on depression and antidepressant prescriptions. I attended the APA sponsored course on collaborative care and learned that according to some standards all psychiatric disorders can now be treated remotely without seeing the patient.

Managed-care organizations and every healthcare plan has a vested interest in providing this type of care to their patient-subscribers. For the past 20 years they have been putting “physician extenders” in hospital and clinic settings with varying degrees of supervision. Many of these personnel like nurse practitioners can now practice independently without physician supervision. The only way physician supervision occurs in many cases is because of local rules within the health plan that employs the physician and the physician extender.

I have seen tremendous variation in how physicians work with nurse practitioners and physician assistants. In one case, I saw a physician working with four nurse practitioners in an outpatient clinic and in half a day - they would see 40 to 60 patients. The incentive for that arrangement was the physician would get the “productivity” numbers.  The organizations call this “population-based medicine” for the emphasis on treating large numbers of people rather than individual patients.  Of course for physicians, there is really no such thing. Seeing large numbers of patients, even with nurses, nurse practitioners, or physician assistants - should not reduce the amount of information needed in the time needed to make decisions and recommendations in that person’s best interest on that day. Some organizations have triage systems in place even before patients see anyone. I wrote about one of those systems four years ago that led to the inappropriate care of an 80-year-old man after a fall.

This court decision is really a managed-care friendly decision in that it affirms a system where physicians have no direct contact with patients but are responsible for the outcomes. Like managed care health systems, the court is basically saying that physicians can direct healthcare provided by nonphysicians remotely. The implication here is that the quality will be the same. A further implication is that the healthcare company does not end up appearing to be a two-tier system, staffed at one level by physicians and at another level by nonphysicians. That would not allow for the indiscriminate draping of quality banners over all the buildings in a certain organization.

Some people see this court decision as a “guild issue” or scope of practice issue.  I have already seen arguments about physician extenders replacing physicians and healthcare organizations and the implications of those maneuvers. Since nurse practitioners can practice independently - this is about physician supervision or collaboration available in health care settings that assume physicians can operate that way.

From a strategic standpoint, this decision points out that you may not be able to ration physician expertise on the one hand and avoid negative clinical or legal outcomes on the other. The average patient does not appreciate this effort until they realize that significant medical care has been denied and the physicians treating them are powerless to advocate for them. Although this case has not been litigated in a civil hearing - the ground rules have been suggested. In this case we have a nurse practitioner who according to Minnesota state law is licensed to practice independently without physician supervision. That nurse practitioner contacts a physician to discuss the case and the physician is implicated in the outcome. Managed-care organizations have successfully used this dynamic for years. The clearest example is the imposition of case managers in hospital settings telling physicians when they must discharge patients based on business rules rather than medical indications.  Another common example is the “physician reviewer” who reviews medical care remotely without any direct knowledge of the patient and makes a decision on whether they should be discharged from the hospital or not. In that situation physician reviewers are immune from liability according to federal law. Warren v. Dinter shows that there may be no such exemption for physicians working in healthcare organizations who provide consultation about medical care and hospital utilization.

Once physicians were sufficiently undermined so that their opinion carried no more weight than nonmedical business personnel, healthcare businesses were able to dictate medical care they way they wanted to. That direction is generally to make a profit and in the case of publicly held companies - to make as large profit as possible to satisfy their shareholders.  

This decision suggests that there may be no protection from civil litigation similar to the federal immunity for business physician reviewers for any working physician making recommendations on the basis of the same limited data.

George Dawson, MD, DFAPA 

Saturday, January 25, 2020

Medical Journals Continue To Support The Business Intrusion Into The Profession

Ran across an article that was posted to Twitter today entitled “The medical profession is breaking its psychological contract with medical students and trainees.”  It was not posted on Medline yet so I am concluding that this is a blog post on the British Medical Journal website. As far as I can tell none of the authors are physicians. The qualifications listed suggest they are all business school professors. Rather than accept my brief summation of the article, I encourage anyone reading this post to read the article in full at this link.

The authors develop their argument from a business concept called the psychological contract. They link to it in their post.  It is from a 1995 book written by Denise Rousseau called Psychological Contracts in Organizations - Understanding Written and Unwritten Agreements.  Searching the author shows that she is a University Professor at Carnegie Mellon University in Organizational Behavior and Public Policy.  The closest definition I could find in this section of pages occurred on pages 9 and 10:

“When two people work interdependently, such as a worker and a supervisor, agree on terms of a contract, performance should be satisfactory from both parties’ perspectives. As individuals work through their understandings of each other’s commitments over time, a degree of mutual predictability becomes possible: ‘I know what I want from you and you know what you want from me’. Commitments understood on both sides may be understood based on communications, customs, and past practices.”  

By the second paragraph the authors have concluded that a violation of the subjective psychological contract has led to the well-known morbidity and mortality within the medical profession although they are focused primarily on trainees for the purpose of their argument. They provide links to burnout, stress, and depression. They suggest that this provides direct evidence that violation of the psychological contract has occurred.  They go on to point out how training environments or “cultures” have a neagtive effect as a number of vaguely defined and poorly quantitated negative outcomes. They never really comment on how widespread the abusive culture is or the total number of people affected. One of their stunning conclusory statements is:

“A cursory examination of the first interactions that physician trainees have with medical schools and residency or specialty training programmes suggests that from day one, the relationships begin somewhat adversarially, suspiciously, and with potentially lower levels of trust between the parties.” 

When I looked at that sentence - as I hope any physician reading this will do - I asked myself if these were the kinds of relationships I had with attending physicians on day one of my residency training. I also asked myself if I had these kinds of relationships with my medical school professors and the residents and medical students that I was teaching. The answer was a resounding “No!”. Medical training is of necessity intense and prolonged but it is not focused on “book smarts” and "high-stakes”.  The authors lack an understanding of why medical training requires this approach and that has to do with pattern matching. Medicine is not learned by “book smarts”. Medicine is learned by seeing as many possible patterns of illness as you can during residency training. In the case of surgical training, that involves as many supervised surgical procedures as possible. Only when this pattern matching has occurred will a physician be safe to see patients and practice medicine independently. If there is any expectation at all on the part of trainees it should be that their training program provides them with these experiences and adequate time with attending physicians so that they might also benefit from the experience of those physicians. 

Every good training program provides that experience as well as the necessary relationships with attending physicians. All through medical school and residency training, a training physician has direct contact with senior residents, attending physicians, and various consultants. They all have varied skills and motivations for teaching, but it is hard to imagine that the training in the United States one cannot find several outstanding teachers and role models in any residency program. I have role models and residents that I trained who I am in contact with to this day. We are still all focused on patient care and united by the goal of quality care and being able to take care of patients with complex problems.

I also have first-hand experience with what directly interferes with the teaching experience. Without a doubt it is the intrusion of business practices into academic settings. Pharmaceutical sales and detailing has been the usual focus but that is completely benign compared with managed care. I have highlighted a few major problems with managed-care and academic medicine in the table below and will elaborate on some of those points.

Business Practices Adversely Affecting Medical Education

1. Lack of quality - before the intrusion of business practices there were medical standards of quality. Those of been replaced by business standards of “value” that have essentially no meaning in the practice of medicine.  The role of physician as a "steward of resources" is a business idea and not a medical one.

2. Unrealistic productivity standards - any academic practice that requires RVU production and awards no credit for teaching productivity necessarily detracts from medical education.

3. Unscientific metrics - medical students and residents can observe attending physicians being ordered around by nonphysicians based on business metrics such as length of stay that have nothing to do with patient care.

4. An unscientific environment - is there adequate time in a managed-care teaching setting to discuss something other than rationing techniques? Is there time on rounding to discuss the latest scientific research? In most cases it is seriously eroded.

5. Documentation burden - it is currently immense relative to before businesses took over the field and that necessarily leads to less direct contact with teachers and mentors and less academic discussion.  This is an artifact of a very low quality information technology environment both in terms of records and security that is the direct result of business based standards in medicine.

6. Less disagreement and controversy - one of the key concepts that every medical trainee must learn is that medical science is an active dialogue and critical papers and concepts change over time. The business influence on teaching environments suggest otherwise and make it seem as though completely unscientific ideas like utilization review and prior authorization represent some sort of immutable standard - the criteria for which never seem to be completely available.  Science is secondary to the proprietary business environment.  Physicians on the faculty who disagree with that are frequently scapegoated and fall into disfavor. Managed-care companies cull the ranks of trainees looking for “managed-care friendly physicians” to maintain the business-based practice.

7. Revolving door policies - there is probably nothing more demoralizing for an intern than to have to readmit a patient who has been discharged because they were in the hospital too long and who returns because they were not stable at the time of discharge. Those discharges are generally based on business metrics.  These policies also eliminate the possibility of residents seeing their patients recover and verifying that their diagnosis and treatment plan was correct.

8.  Unnecessary bureaucratic burden - I was fortunate enough to be an intern at the time managed care was just starting to take off. At that time I had a critically ill patient in an intensive care unit and I was contacted and told that they needed to be transferred to another hospital because of their insurance contract.  Today practically all physicians routinely encounter the managed-care intrusion into their patient care on a daily basis. With physicians in training it is no different. They are still subjected to the review processes and spending far too much time getting medications approved that are clearly indicated. All of this detracts from teaching and learning time.

I have directly observed all the above items taking its toll on teaching and learning medicine but I have a couple of anecdotes that bring together many of these areas. The first was my experience on a neurosurgery rotation as a medical student. I did two neurosurgery rotations with the same residents and attendings and at one point hoped to be a neurosurgeon. There is no more rigorous course of training. All of the senior residents were essentially on-call 24/7 all year long. In those days we were rounding on about 35 patients 10 of which were in the neurosurgery intensive care unit. We would typically be done with rounds in two hours and the documentation would be done at about the same time. Our documentation would include the postoperative day number, patient’s subjective status, what the surgical wound looked like, and review of their vital signs, labs, and physical exam. A typical note was no longer than five lines and we could complete it as we were moving from patient to patient. Over the years the federal government developed documentation guidelines that were turned over to the managed-care industry so that every medical encounter these days takes an excessive amount of documentation. If we were doing the same rounding procedure today it would take us additional 2 to 3 hours just to complete the documentation. That 2-3 hours would detract from time in the operating room where residents were learning how to perform neurosurgery and medical students were learning by observing those procedures. That 2-3 hours would detract from time where the senior neurosurgeons would teach imaging rounds and review all of the brain and spine imaging from all of our patients that week.  In short, business practices would have essentially cancelled out most of the teaching on neurosurgery.

My other anecdote has to do with materials available for teaching. At my last teaching position I enjoyed presenting an annual review for psychiatric residents taking the annual “in training” exam. This examination includes questions about neurology, neuropsychiatry, and brain imaging. As an attending physician focused on neuropsychiatry I always had plenty of brain images that were relevant to the practice of psychiatry. With the electronic health record implementation it was relatively easy to download and de-identify those images for teaching purposes. When I sought permission to do that from the medical director at our clinic she stated: “Why would we let you use our images?”  I was stunned because prior to the takeover by businesses, reviewing films even if they were not de-identified was standard teaching practice.

The final anecdote is probably the best. Back in the early days of the business takeover of medicine, the FBI was actually engaged in investigating medical billing and making sure physicians completed the correct documentation template. If they didn’t they could be charged with a crime up to and including a RICO violation. Of course these templates were completely subjective but that is not the way the FBI was treating them. There were several well-known prosecutions of large medical clinics based on the fact that attending physicians were not documenting enough when they supervised residents.  There were no guidelines at the time about what might be involved and so my business people were telling me that I had to document the standard note whether I was working with a resident or not. You can imagine the demoralizing effect that has on a resident when they notice their attending is putting in a separate note every day and their note seems to be irrelevant. When I noticed that happening I suspended all of my teaching of residents because I did not want to insult them just because business and government bureaucrats were telling me what to do. Eventually that guideline was relaxed so that I could go back to documenting that I had discussed the case with the resident but not until considerable damage had been done.

Based on these experiences and more, the opinion piece in BMJ strikes me as another effort to exert top-down control by business interests on the field of medicine.  It is an extension of three decades of failed business initiatives that nonetheless still dominate the practice of medicine in the United States.  Businesses and governments alike are still using the failed strategies. As I pointed out, the same failed strategies have already taken a toll on medical education. And yet these authors suggest that another vague business concept should be applied to medical education.

When I think about my mentors, my colleagues, the residents I have mentored, and what we have all accomplished - we need to keep business concepts out of medical education. We also need to look at the overall strategy and why business authors keep appearing in the pages of our journals. It all seems to be based on the premise that is business managers are experts at everything. 

That is clearly not true in medicine.  They have introduced chaos and stress into the clinical field.  They have already seriously stressed medical education and this opinion piece provides another non-solution that can only be suggested in the context of having wrested control of the clinical practice of medicine away from physicians. 

George Dawson, MD, DFAPA


I have seen recent psychiatric treatment that may illustrate what happens when business managed settings limit patient contact.  In my current employment, I see a lot of people who are treated with antipsychotic medication, antidepressants, and mood stabilizers for a presumed psychiatric diagnosis.  The medications are started and titrated rapidly.  By the time they see me they are experiencing clear side effects, taking too much medicine, and the diagnosis is not clear.  Hospitalizations today are so short and so focused on doing something in a short period of time that physicians in training have limited exposure to the concept of substance induced psychiatric disorders and how they can be best treated.  That also includes the appropriate detoxification of these patients - many of whom are sent out to social detoxification units unless they worsen and are sent back.  All of this decision making should be part of the knowledge base of psychiatrists and primary care physicians.

Thursday, January 9, 2020

The Era of Blockbuster Drugs In Psychiatry Is Almost Over

Four years ago, I commented on a prediction that the era of “blockbuster drugs” in psychiatry would soon be over. That designation is used for medications that generate $1 billion in sales per year. Even at the time of my original comment, most of the drugs prescribed by psychiatrists were generic drugs that were often very inexpensive. The example that I frequently refer to is the antidepressant citalopram that was available for four dollars per month.

Last week PharmaCompass, a drug industry analytic and service company came out with a list of the top 100 drugs by sales in that industry. They listed the top 30 drugs by sales and their website lists the top 500 drugs with additional metrics such as whether or not sales are increasing or decreasing and by the percentage. There were no psychiatric drugs in the top 30. Lyrica or pregabalin, a drug used primarily for neuropathic pain and fibromyalgia came in at number 16 and generated $4.6 billion in sales. Some consider it to have a tertiary indication for anxiety but I would not consider it to be a drug primarily for psychiatric applications.

Company / Companies
Product Name
Active Ingredient
Main Therapeutic Indication
2018 Revenue in US$ billion
AbbVie Inc., Eisai
Autoimmune Disorder
Bristol Myers Squibb, Pfizer
Cardiovascular Diseases
Bristol Myers Squibb, Ono Pharmaceuticals
Amgen, Pfizer, Takeda
Merck & Co.
Mabthera/ Rituxan
Oncology, Immunology
Bayer, Johnson & Johnson
Cardiovascular Diseases
Bayer, Regeneron
Johnson & Johnson, Merck & Co., Mitsubishi Tanabe
Autoimmune Disorders
AbbVie, Johnson & Johnson
Prevnar 13/ Prevenar 13
Pneumococcal 7-Valent Conjugate
Johnson & Johnson, Mitsubishi Tanabe
Immunology, Dermatology

Medications for neurological and mental disorders had a total market share of $55.8 billion or 10.2% of the entire market. There were only three medications that would qualify as blockbuster drugs and being used by significant numbers of psychiatrists and they included Latuda (lurasidone), Chantix (varenicline), and Abilify Maintena (aripiprazole).  The blockbuster drugs listed are all concentrated in a few specialties. Several are from the same class.  For example both Eliquis and Xarelto are novel oral anticoagulants (NOACs).  The total market share of the top 15 drugs greatly surpasses any amounts every attained by psychiatric medications. That makes sense because many are treating conditions that have a high prevalence and these medications generally represent an advance in pharmacotherapy.

I posted this mostly to note the trends over time. Journalists and other commentators tend to get carried away with market snapshots. I have not really noticed any negative commentary about the fact that the current blockbuster drugs that have nothing to do with psychiatric disorders are generating much more pharmaceutical company income. I also have not heard any commentary on how there is a natural fluctuation in this market based on drug discovery and the current rules on patent exclusivity. Many of the previous psychiatric “blockbusters” are obviously rapidly losing market share now that there are generic versions. I would argue that the current data also may indicate that restrictions on current medications may be limiting their market share.

The issue of direct to consumer advertising has typically not been discussed in these opinion pieces.  I have seen Rexulti frequently in television ads.  The application is generally antidepressant augmentation - accompanied by a very fast and diluted compilation of symptoms of both tardive dyskinesia and neuroleptic malignant syndrome. How much do those ads contribute to the market growth of Rexulti? My speculation is quite a lot. It would not take much for a person taking an antidepressant and not experiencing adequate relief to call their primary care physician for the addition of Rexulti.  I have seen similar ads for Vraylar.

Looking only at market share and pharmaceutical revenue only provides a look at one cost of this market. There is a huge pharmaceutical benefit manager component that generates revenue primarily by requiring their authorization for specific medications and also packaging deals for managed-care organizations. There is definite cost what they do but they currently are allowed to force physicians to work for them for free.

I hope that this information allows people take a more skeptical look at political and journalistic pieces that use pharmaceutical sales numbers and specialty sales for one rhetorical purpose for another. I am specifically referring to the recent good old days of the psychiatric drug “blockbusters” where much of what was written was an indictment against psychiatry and those psychiatrists who either worked directly for pharmaceutical companies or received some minor perk. There was the conspiracy theory that pharmaceutical companies had influenced the DSM-5.  That is both unrealistic and it greatly exaggerates the importance of that book. The data illustrates that no amount of physician manipulation can maintain a pharmaceutical product as a “blockbuster” independent of market forces.  

Those market forces also have a significant political and regulatory component where the real conflict of interest lies.  Allow me to translate that to Congress and the FDA.

George Dawson, MD, DFAPA

Graphics Credit

The top table from PharmaCompass was posted with their permission. 


I had to identify and hand count the psychiatric drugs on the PharmaCompass web site from their graphic interface.  It is possible that I missed some or the count is off (they were not numbered).  Let me know if you detect any errors.