Key moments, in order
Simply asking whether a condition is there is the same as asking the patient "do you think you have ADHD?" and ending the interview. Running the structured interview for the disorder the patient asked about almost guarantees that diagnosis.
Whatever time you have, partition it into rough percentages in a set sequence: set the frame, spend the first 20% on free-form narrative, then focused history, and finish with your impressions and treatment plan.
Open every intake with the same introduction of yourself and the purpose of the meeting, even for experienced patients, because doctors vary so much. Establish your role and modality as early as possible, since it changes the whole interview.
There is an argument for disclosing the limits of confidentiality in full at the first visit.
Say what the visit is: an attempt to get as accurate a picture as possible, diagnostically and in formulation, in order to recommend treatments, medications or other.
Warn them: in the first appointment I interrupt, I jump around and I ask a lot of questions; I promise to be better at listening from here on.
You do not need every detail today. Leave a black box for what you will get to when they feel safer; the fact that it cannot be gathered now is itself information.
Dysregulation does not get you good information anyway, and it can be subtle: freezing rather than agitation.
The frame is not only telling the patient what will happen. It also establishes some common ground between you.
Aim for physicianly concern and interest: warm, but not overly so, so the patient feels warmth while understanding this is a professional relationship.
Even if the patient does not pay a cent, they are paying emotionally, and that is a big toll.
The first 20% should never be structured interview questions or eliciting specific criteria. That is a poor way to get to a diagnosis.
About 90% of patients with genuine psychopathology will describe the DSM criteria without you ever mentioning them, because that is their lived experience.
Ask for a free-form narrative and steer away from clinical terms, which mean different things to different people. When someone says depression, anhedonia or increased anxiety, ask what they actually experienced.
DSM criteria are not a diagnostic test. They are a description of a syndrome, and an incomplete one, even though test-builders often treat them as a test.
Draw a graph of the patient's life from birth to the day they see you, going up and down in levels and types of psychopathology and in life functioning.
Ask how many depressive episodes they have had. Some patients look confused and ask what you mean by episodes, and that alone tells you a lot.
While gathering the narrative, use the major DSM chapters as your differential. Do not drill down to a specific diagnosis within each chapter yet; think of the major category first.
Run through the categories: psychosis, bipolar, depression, anxiety, OCD, trauma, dissociative disorders, somatic disorders, eating disorders, sleep problems, sexual problems, impulse control and addiction.
Personality is the most clinically neglected category today. What we mostly see is clinicians ignoring it, or diagnosing off vibes and countertransference rather than digging up the clinical data across the lifespan.
If you, as a seasoned clinician, feel stupid during an intake, that is diagnostic information about what is going on.
Disregard the review of systems. It is an artifact of insurance, not part of the medical interview.
With poor insight and a motive to look well, a structured interview just gets every symptom denied: "No, doctor, I'm perfectly fine."
Ask why each medication was switched and what was going on at the time. You can get almost a full narrative life history that way.
You will run out of time, so go deeper where the narrative points. If you suspect personality, dig into work history, conflicts, firings, arrests, and how many relationships and how long they last.
Do not neglect medical history. It is so important, and many people do not recall theirs accurately.
Ask about criteria without giving the game away: not did you have racing thoughts, but what happened with your thoughts around that time?
In the interview, the way they answer the question is much more important than the answer itself.
With experience you learn to tell someone whose memory is being activated and turned into a story live from someone presenting a prepared story, consciously or not.
Keep the basic history questions in front of you so you can fall back on them whenever you get confused. Take notes, and warn patients you will: you are there to make a diagnosis.
Do not be afraid to pause and think in front of the patient: give me a moment, I need to sit with that.
If an hour is not enough to think the patient through, say so, spend the whole hour collecting information, and finish the interview at the next appointment.
Show your work: what you found and why. Our terms are loaded and lay understandings are often wrong, and patients trust a diagnosis far more when you show your reasoning and what you reviewed.
The crux is figuring out the diagnosis, getting on the same page with the patient, and then giving treatments that fit that diagnosis.
Invite disagreement: I really want to know if you disagree or are not sure. How do you feel about my impressions? Does that make sense?
Read through all of the DSM-5. Each diagnosis has a differential diagnosis section under its criteria that strengthens your core knowledge.
Each moment is a point made in the episode, linked to the second it was said. The topic links open everything the podcast has said on that topic.