Friday, July 12, 2019

2000 Words About the Last Ten Minutes of a Psychiatric Evaluation





I do psychiatric interviews all week long. At this point I’ve been doing it for over three decades. I am always interested in improving the process. I came up with this idea recently and wish I had thought about it many years ago. My standard initial interviews are anywhere from 45 minutes to 60 minutes long and that includes a lot of information transfer. The actual amount of information transferred and whether it can be used come up with formulation and diagnosis is always an unknown variable. There are people who can come in and succinctly produce that information in 30 minutes or less. Other people will take several hours. At the end of every interview, I have to dictate a report that contains fairly standardized sections including a diagnosis and treatment plan.

When physicians are trained there is an implicit understanding that they will be able to adhere to a specific protocol and come up with a diagnosis and treatment plan. In the real-world things are never that linear. In inpatient psychiatry, the patient may not be able to produce a coherent history or any history at all if they are severely ill and the diagnosis needs to be pieced together by collateral information and the medical and behavioral presentation. In outpatient clinics, the format is much different. It is still possible to interview a patient who communicates low information content but it is much more likely that patients describe what is necessary for diagnosis and treatment. One of the main problems is the discussion at the end of the assessment. It is typically truncated by time constraints. Many patients leave the assessment and realize they should have mentioned specific concerns or asked more questions.  I often have people tell me that they are drawing a blank in the interview and want to know if they can call me if they have additional questions. I reassure them that they can.

These brief observations highlight the need for adequate time and structure following the diagnostic interview to facilitate the discussion and make sure that the patient can send and receive the communication that is both necessary and what they want beyond that. I came up with the structure over the past two weeks and I think is a good starting point. I also refer interested readers to the only post on this blog that was not made by me. It was written by Cedric Skillon, MD one of my psychiatric colleagues because I observed his superior treatment planning in all of his notes that I encountered.

This exercise assumes that any psychiatrist reading this as adequate time to do their work. I don’t think psychiatry can be reduced to a checklist and I don’t think it can be done in a short period of time. I realize that there are psychiatrists reading this who may have evaluations scheduled back to back in the morning and 15-minute appointments scheduled back to back all afternoon. We all know how that schedule gets accommodated and the outcomes are not good. What typically happens is there is very little emphasis on getting to know and understand patients and little emphasis on communicating with them about their problem. I also understand that the people who really control all psychiatrist practice these days don’t really care about quality treatment. The emphasis here is always on quality and the obvious fact that quality work is not easy and it takes time.

What follows is my proposed outline for the patient discussion. I think that in even moderately complex cases at least 25% of the allotted time should be for this discussion. I have not done it yet but I hope to hand an outline to the patient in order to facilitate discussion. I’m very interested in whether their psychiatrist have used this technique and also speculation on why it may or may not be a good idea.

1.  A discussion of the diagnosis and/or formulation

Of the people I talk with very few seem interested in the specific diagnosis. It does happen, and when it does, I think it is important to cover most likely diagnosis and complicating factors. For example, when seeing a patient to has been drinking excessively for 20 years and gives a history of anxiety and depression prior to the onset of that alcohol use and current anxiety and depression can there be a valid anxiety or depression diagnosis given the intervening alcohol use. This is a basic level of complexity that psychiatrists encounter. In covering the diagnosis, I will typically talk about a most likely diagnosis as well as the provisional or working diagnosis and my recommendations for clarifying the diagnosis. In some cases, people are interested in whether or not they have a severe psychiatric disorder. I encounter a lot of people at the end of the initial interview who asked me “So am I nuts/crazy Doc?” Some people have a specific diagnosis in mind that they either want or don’t want and I give them my best assessment. Some people have studied diagnostic features online or in the DSM and use that jargon during my interview. They may or may not be right.

I think a formulation is always necessary in addition to a psychiatric diagnosis. I was first impressed with rapid formulations done in brief psychodynamic psychotherapy. I also found several courses given at the annual APA conference in diagnostic formulations to be useful. The patient may have their own formulation or one that they worked on with the therapist that I would typically discuss in the original interview. I don’t think the formulation discussion has to be excessive or detailed and it may complement the diagnostic assessment. As example the case of the diagnosis of major depression, it is reasonable to discuss the interpersonal elements relevant to the depression as well as psychosocial stressors.

2.  A discussion of a no treatment option

A no treatment option should be explicit. It should be discussed as part of the informed consent process. The potential risks and benefits of no treatment should be discussed but in many cases no treatment is a viable option for people. The case that always comes to my mind is the person who consulted with me decades ago after receiving years long psychoanalytic treatment who wanted to know if he should find a new therapist and continue. After reviewing course of therapy and how his anxiety had improved over the years, he was very relieved to hear from me that he probably did not need to continue psychoanalysis. He had formally terminated with a psychoanalyst and no further treatment was recommended but he wanted a second opinion from a psychiatrist. Other people have mild anxiety and depression but are still functional and they had concerns about both medication and psychotherapy. With psychotherapy - the time, financial commitment, and past therapy experience are generally the determining factors. With medication the main concern is side effects and typically concerns about taking any medication. I commonly hear “Doc - I don’t like to take any medication, not even aspirin.” I think it is possible to discuss no treatment in a number of scenarios as well as the contingencies. The common contingency might be returning for reassessment if symptoms worsen or there are other complications.

3.  A discussion of potential medical treatments and medical concerns – especially those that may affect treatment

Outpatient psychiatrists are generally seeing patients after all the smoke has cleared. If they have access to an electronic health record (EHR) there are often laboratory results, ECGs, and imaging results and reports that the patient may or may not be aware of. After taking a complete medical history and review of systems I generally ask the patient if they had any questions about recent medical procedures or results and we have a discussion at that time.

The medical treatment of the identified disorder is reviewed. There are often several approaches to the problem and when I discuss all those approaches patients generally want my recommendation. I provide them with that, the necessary prescribing literature, and both short and long-term plans. I also talk about rare but serious complications of psychiatric pharmacology like antidepressant withdrawal, agranulocytosis, Stevens Johnson syndrome, priapism, serotonin syndrome, akathisia, suicidal ideation, and neuroleptic malignant syndrome. In a previous post, I describe some of the discussion. I generally want to make sure that the patient is aware of any black box warnings, contraindications, and warnings. I will often do a drug interaction search in front of the patient so that I can discuss it with them rather than doing it after the interview.

In some cases, my treatment will depend on my discussion with their primary care physician, cardiologist, or other specialist. It may also depend on me ordering tests that I view as critical and getting those test results.  

4.  A discussion of potential psychotherapy treatments

There are research proven psychotherapies that should be part of any informed consent discussion. I commonly inform patients with posttraumatic stress disorder (PTSD) that exposure therapy and the EMDR variant are very good approaches to the disorder and may be superior to pharmacotherapy. That does not mean that I won’t try to treat that patient for PTSD symptomatology, but I want to give them a clear message that the psychotherapy results may be much better. I have similar discussions regarding the psychotherapy of anxiety and depressive disorders specific to those conditions.  The psychotherapy may include supportive and crisis intervention approaches that occur right in the initial interview.  In psychiatry treatment interventions that involve psychotherapy can occur right at the time of the assessment. 

5.  A discussion of potential environmental/social approaches

One of the commonest problems in outpatient psychiatry is that there are clear environmental factors that are leading to anxiety, depression, insomnia, and substance use morbidity. Practically all of the people in those situations or consulting psychiatrists are looking for a medical treatment to address those environmental and social factors. I generally have detailed discussion about the evidence that medication works in those situations (there is very little) and what can be done.  I typically discuss examples like grief responses and the very common scenario of a stressful work environment or boss. I want to be sure that I emphasize the point that medical treatment of these acute and chronic stressors is almost always not enough to solve the problem.

6.  General treatment parameters

If the patient and I decide to proceed with treatment – a discussion of the general parameters of treatment needs to occur.  That would include indications for calling me and how to address potential medical and psychiatric emergencies.  In this day of external control of medical treatment – prescription refill policy can be discussed and how insurance company denials or prescriptions (prior authorization) will be handled.  In large systems of care, a discussion of who has access to the information in the chart and systems that deal with emergency and crisis calls can be a topic for discussion.  In the current Internet age modes of communication (email, texts, calls) also need to be discussed especially if there is no formal EHR portal.  Limitations on privacy are typically part of the general permission to treat but may require additional explanation like the CFR42 regulations for drug and alcohol treatment.

7.  Definition of the treatment alliance

As part of the general treatment discussion, definition the treatment alliance can be useful for many people.  Physical appearance and the appearance of one’s office can be a projective test for people entering treatment.  When I started out, I would get a lot of comments like: “You seem to be too young to be a doctor!”.  Now that I am an old man, my introduction has stayed the same.  Within a few minutes people know how long I have been in practice.  Most people can get past that and engage in a dialogue, but some people are overly deferential and seem intimidated.  Others decide to challenge the old man to see what he knows or if he can be pushed around. Both of those scenarios can be a topic of discussion but that might be too confrontive for most and lead to more problems in communication.  A more neutral approach is to discuss the typical diagnostic interview and treatment alliance. I find a basic definition that works for most people is: “You and I working on a problem or set of problems that you identify to see if we can find the solutions.”   

There have been entire books written on this topic, but most don’t focus on the ten or 15 minutes at the end that should be for the benefit of the person seeking help. That is my outline so far. Over the years I have had these discussions but not in a structured way. It de-emphasizes the diagnostic template or checklist and puts the emphasis on information that is directly useful for the patient. I plan to take the headings, add bullet points, and see if it adds to the discussion. I think it will because it will at least define the boundaries of the discussion, what I hope to communicate, and provide more solutions to people coming in to see me. And mostly I hope that it will counter the implicit assumption that all psychiatrists do is prescribe medications and that it is an easy thing to do.


George Dawson, MD, DFAPA




Supplementary 1:

The preliminary form described above is contained in the 3 PNG files below. There are 49 bullet points so far. I have it as a Word file and a PDF so if you are interested in it send me an email and I will send you the format you want.  I consider this to be a work in progress. Click on the files below to enlarge for easy reading.




Supplementary 2:

As a preliminary screen - I looked at one day of intakes in my practice and discovered that I am routinely covering 30-36% of the bullet points on this form. That seems like a sufficient number but it would be a good idea to run a trial looking at the baseline for a number of clinicians (rated by independent observers) and then a second rating after the patients were provided the form ahead of time.








1 comment:

  1. I can't really add much to this, but I wanted to say that I like this post.

    ReplyDelete