For my first post, I will attempt to succinctly summarize the basic structure of the initial psychiatric interview on an inpatient psychiatric unit. My target audience are medical students who are just starting out and anxious to start their new adventure on the ward.  However, I will attempt (and fail) to make this post as complete as possible and relevant to more seasoned psychiatry residents without turning it into a book or make it overwhelming to newcomers.

Please keep in mind that there is no “one right way” to conduct a psychiatric interview, but there exists a “less right way”.  Every rule has exceptions, especially given that our patients have profoundly different pathologies, cultures, and beliefs.  Many of these exceptions are “common-sense” but I will try to point out some of the more common ones.  For example, here is one exception that irks me when a student misses: if at any time the patient complains of pain, stop and ask about the pain.  I have seen a patient complain of chest pain, only to be ignored by the student who, in the interest of staying on topic, was insisting that they answer questions about their childhood.  Please do not forget that we are medical doctors first.  Do not forget everything you learned in medical school because you are on a psychiatric floor.

I encourage you to use the comment section to open discussions based on your experiences.  I also welcome all questions and criticisms.

The first encounter you will likely have with a new patient is when you read the emergency room note in the chart.  Pay close attention to who brought them to the hospital (police, ambulance, family, self, etc.) and under what circumstances.  Also review labs, vital signs, and any toxicology reports.  With a bit of experience, you will already be able to formulate differential diagnoses just based on the ER report.  Reviewing the ER note will not only help guide your questions during the interview, but more importantly it will also warn you of potential violence.  Remember that safety for yourself, other staff, and patients receive top priority.

There are three possible locations to interview patients: a private interview room, the patient’s room, and near the nurses’ station.  In the most extreme cases, I will refuse to interview a violent patient without sufficient security escort (I can only recall one such case in my years as an attending).  Choosing the appropriate venue requires an assessment for safety.  For patients who are calm and have no history of violence, the appropriate location would likely be in a private setting.  A quick review of the ER note and a glance at the patient is sometimes enough to estimate a probability of violence.  If a patient was hostile in the ER but who now appears calm may still be appropriate for a private interview setting.  Many times, asking the patient “How are you doing today?” can give you clues as to the current level of agitation.  A calm response does not insure safety because patients can have very labile mood.  Listen to your instinct or “gut feeling” since it should kick in if a patient poses a threat.  Remember that you are the descendant of a long line of successful generations who owe their success to accurate instincts.  When a patient has a history of violence but appears very calm, I may ask the patient “Am I safe taking you to a private room?”  Surprisingly, from my experiences the patients usually answer honestly.  For example, they may answer, “Only if you discharge me home.” *insert red flag*

Regardless of choice of interview setting, things can go wrong very quickly.  Always plan an escape route beforehand so when things go sour you can safely exit.  Never let a patient stand between you and the door unless your intention is to test your ninja skills against a threatening patient who is blocking your only path out.  Keep in mind where the safety call buttons are and sit next to them.  Sitting arrangement is important and should be set up prior to bringing the patient into the room.

Before we start the actual interviewing phase of the assessment, I want to mention a few things that may be useful for you to keep in mind.  One of the ways that psychiatric interviews are different from other medical interviews is that you have to constantly monitor your own emotional state.  Our mirror neurons allow us to empathize with others.  If you interview a patient and they make you feel sad for reasons unbeknownst to you, then that may be a clue that the patient may be depressed but is hiding their emotional state.  On the other hand, if they make you feel very happy and giggly, then they may be manic.  Euphoria is contagious!  If they make you feel frustrated and angry, consider a personality disorder.  Of course, a strong counter-transference may also say something about you just as much as about the patient, so consider why they make you feel the way you feel carefully before labeling a patient.  If the patient claims that they are very depressed but you feel nothing, again ask yourself why.  Perhaps the patient is malingering, or has a personality disorder, or is psychotic and is unrelatable.

Since we are on the topic of counter-transference, remember that patient may make you feel angry. Very angry. Patients may know exactly which buttons to press to get a response out of you.  Be meta.  Analyze the situation in the room as objectively as possible and determine what the patient is doing to rile you and why it is succeeding in getting to you. Remember to always set your frontal lobe to overdrive before you react.  Here is a video of Master Yoda handling the situation:

You will eventually learn how to deal with various hostile situations, but as a rule of thumb: everything you say during the interview should have a thought-out reason.

“Be meta” applies to all situations in psychiatry and not just adversarial instances.  For example, if a patient is depressed and is speaking very softly, some students tend to match their own tone and volume to that of the patient’s.  My recommendation is to start “one level” above the patient’s and gradually increase your own volume and tone to see if the patient is able to follow you.  A patient with depression that is less severe may gradually become more animated using this approach, and thereby you can assess the patient’s range of affect. Conversely, a patient with mania will be very loud and speak rapidly. Again, students tend to match the patient’s volume and rate. Instead, slow yourself down and speak in a calm and soothing manner to assess the patient’s ability to do the same.

The first step in assessing the patient is to take a moment to just observe them.  Are they disheveled or well groomed?  Do they look like they are in physical or emotional distress?  Do they appear sad, euphoric, or flat? Do they appear paranoid with shifting eyes and constantly looking over their shoulder? Are they responding to internal stimuli?  Do they make eye contact?  Do they have cuts or burn marks on their forearms or legs?  Any unusual tattoos or piercings?  All of these observations will be necessary when formulating differential diagnoses.

Next, start by introducing yourself and the rest of the people in the room.  Proper introductions are essential because you are asking a patient to open up their innermost thoughts and feelings to a group of strangers.  Also, paranoid patients may feel safer knowing the identity and roles of all the people in the room.  I generally keep the initial introductions verbal and do not shake hands unless the patient insists on it.  Keep in mind that some patients may have germ phobias, autism, paranoia, or other reasons for feeling uncomfortable with hand-shakes. Another reason: lice and fecal matter.

Following proper introductions, we want to get the Chief Complaint in the patient’s own words.  There are several methods to approach this, but I will generally say something like: “We read your chart and therefore know a bit about why you came in. However, I would like to get the information directly from you in order to make sure everything that was written is accurate.  I also wanted to get more information so that we know how best to help you.  So, what brought you to the hospital?”

There are several reasons that I start by explaining that we already read about the patient.  First, the patient has likely told their story numerous times by now.  Some patients will therefore become impatient and yell “I already told them! Why do I need to keep repeating myself?!”  Also, the patient feels more at ease knowing that you are listening to them and not what others (like the police) wrote about them.  Sometimes the patient’s first remark is “Can I go home today?”  To this, I sometimes counter with something like: “I understand that being discharged is important for you, but we need to finish this interview first before we can know the discharge plan.  We will get to that at the end.”

The next thing that happens can be anything.  The patient may be honest, lie, or deny knowing anything.  This is where knowing a priori about the circumstances that brought the patient into the ER can help.  My approach is that of a private investigator.  Start with open-ended questions, but close gaps with more specific closed-ended questions.  This can sometimes feel like putting a puzzle together.  However, this puzzle is not only missing pieces, but the pieces that you have may be fabricated or just too incoherent to fit into the whole.  The important thing is to listen carefully for anything that may be unusual with what the patient tells you and latch on to it.  For some patients, psychosis is very obvious, but for others it may be more subtle.  If a patient tells you something that you do not understand, consider psychosis as a possibility and explore further.  Remember that when you test a patient’s thought process, you are using your own as a metric.  Therefore, you assume that your thought process is linear and logical, and therefore if something makes no sense to you, do not blame yourself and move on (a very common medical student mistake).  Ask more details to clarify.  For example, if the patient mentions briefly that their aunt is WCA, they should be interrupted and asked to explain.  Perhaps it will turn out that WCA stands for “WitchCraft Association” and that the patient believes that she has cast a spell on his cat that made the cat run away from the patient (yes, this came from an actual patient).  My rule of thumb here is: Listen very carefully and follow the delusions down the rabbit hole to see how far they go.  You may be surprised at how linear someone may seem until you ask them to elaborate on a delusional belief.  Sometimes it can be like pulling a thread; even if the thread is minimally sticking out of a garment, pulling on it can unravel the entire thing.

During your investigation into the chief complaint, you are formulating differential diagnoses.  For example, you may decide that the patient likely has schizophrenia and possibly also has PTSD.  When you feel that you understand the circumstances that brought the patient to the ER, you should ask questions about specific DSM criteria to rule in or rule out your differentials.  If on the top of your differential is borderline personality disorder, start asking about the DSM criteria that would confirm it or exclude it.  If second on your list is major depression, ask the criteria for that (SIG-E-CAPS) next.  The reason that prioritizing the questions based on the differential in this way is important is because some patients are very impatient or impulsive and may terminate the interview at any moment.  At the very least, before this happens I try to maximize the probability that I have enough information to make a diagnosis and be able to decide on an appropriate plan for treatment.  This requires accurately sorting the differential diagnoses based on likelihood and then picking the lowest hanging fruit first.

Continue specific questioning of criteria until you have checked every item on your differential.  You will likely develop your own style for asking specific questions, but avoid medical jargon.  For example phrases like “racing thoughts” or words like “euphoria” may not mean much to a patient without a strong educational background.  More appropriately, you can rephrase these question to something like “thoughts going so fast in your head that you cannot keep up” and “feeling like you are on top of the world.” Instead of “pressured speech,” you can ask about “people telling you that you are speaking too fast.”  Regarding auditory hallucinations, surprisingly patients usually know exactly what you are talking about when you ask “Do you hear voices.” However, you will occasionally get the response “Yes I hear voices. I hear your voice right now!”  Students sometimes then reply, “I mean voices that nobody else can hear.”  However, the patient may not know that other people cannot also hear the voices.  I usually just ask, “Do you sometimes hear voices when you are by yourself?”  I am not going to list every single DSM criteria and give examples of how I may ask or not ask about it, but my recommendation is that while you are studying and memorizing the DSM criteria, try to come up with simple ways to inquire about them.

One quick tip: don’t say to the patient “I’m going to ask you some questions that may seem weird.”  Imagine that you are in a hospital and the doctor says “this may mean you’re weird, but do you feel pain right here?”  Personally, I may deny feeling pain no matter what (especially if my goal is to leave the hospital).  Just go ahead and ask the questions.  There is usually no reason to preface the questions.

For any symptoms that the patient admits to having, you will likely need to inquire further.  For example, if the patient admits to hearing voices, the next questions should include some variation of the following: Do you recognize the voices? Male or female? How many voices are there? What do they tell you? Do they ever give you commands? If you had to guess, where would you say the voices are coming from? Do they ever tell you to hurt yourself or others? Do they ever speak to each other about you? Do you talk back to them? What makes them louder? Softer? (Remember: private detective!)

Always pay close attention to body language and facial expressions, and pause to inquire about any unusual behaviors.  I had a case where the patient would at times raise either his left or right shoulder throughout the interview.  I finally asked him why, and the patient calmly explained “I don’t want to interrupt our conversation, so I answer God’s voice by raising my left shoulder for ‘yes’ and right shoulder for ‘no’.”

Regardless of what is on your differential, always ask questions for psychosis, depression, and mania, and always include questions about thoughts to hurt self or others.  Of course, you will need to either memorize the DSM criteria for common pathologies or have them written down.

Before moving on to the next phase of the interview, I want to give another quick pointer.  I believe that there is nothing wrong with having a sheet of paper in front of you during the interview in order to stay organized and avoid skipping critical questions. Jumping around from topic to topic can be very disruptive to the flow of the interview, and this is especially true when the patient is disorganized.  Sometimes medical students blank out and cannot think of what to ask next, at which point they all of a sudden jump to a random topic and then either try to come back to the current topic a few minutes later or they forget where they were.  I would argue that having a few seconds of silence is better than jumping back and forth and getting everyone confused.

Some patients may require frequent redirection.  Medical students tend to have difficulty knowing when to listen to the patient and when to interrupt them and bring them back on track.  Here is my rule of thumb: If the patient is off track and you are not getting any useful information that may help you diagnose or treat the patient, go ahead and bring them back on topic.  There is no need to let them talk on and on about unrelated topics.  For example, if a patient answers the question “What brought you to the hospital today?” with “Well… It all started when I was five years old…” followed by a long-winded explanation of their childhood, I would politely say something like, “That sounds very important, and we will come back to that.  However, first I want to specifically hear about what happened that brought you to the hospital now.”  if you can’t think of a question, don’t just jump to a random topic.

Congratulations!  At this point you are done with the “history of present illness” which is usually the most difficult part of the interview.  The remainder is more straight-forward.  The order of the following is flexible (and based on the specific patient), but you want to make sure to ask the following:

Past psychiatric history: Ask about the number of psychiatric hospitalizations. When was the first hospitalization? When was the last? What was the diagnosis? Which psychiatric medications has the patient tried? Which medications worked and which did not? Were there any side-effects?

Substance abuse history:  Patients will frequently minimize what they reveal about their substances use.  If the urine toxicology or collateral information contradicts their report, gently confronting them with this fact is usually appropriate.  You should try to quantify the amount of each drug used, when it was last used, how often it is used, and when the patient started using the drug.  Note that many patients do not consider alcohol or marijuana to be drugs, so you need to specifically ask about them. Also, you will need to ask specifically about cigarette use, encourage them to stop smoking, and offer a nicotine patch/gum for withdrawal symptoms.  I usually also specifically name some drugs that are commonly abused in the community such as cocaine, crack, methamphetamine, opiates, benzodiazepines, etc.

Family psychiatric history: Ask about any psychiatric illness in the family.  If such illness exists, ask the patient for information about the diagnosis, psychiatric hospitalizations, and which medications helped.  Also ask about history of suicides. In addition, you should inquire about family members with substance dependence.

Medical History: Ask about any history of medical problems, surgeries, or pain.  I ask specifically about diabetes, hypertension, and any heart problems.

Legal history: Has patient ever been arrested? Any time spent in prison? What was the crime? Any history of violent crimes?

Social history: The social history is very important for understanding the patient’s baseline functioning and for disposition planning.  Crucially, we should also inquire about children to make sure that all kids are safe and taken care of.  Ask about where the patient lives and with whom, marital status, history of domestic violence, educational history (“How far did you get in school? Why did you drop out?”) , employment history (“When was the last time you worked? How long did you work there? Why did you stop working?”), and current income (“Are you on SSI/SSD/GR/food stamps? Who is your current payee?”).  If undomiciled, when was the last time they had a place to live and why they became homeless? Who does the patient have for social support? How many siblings do they have?  Is the patient in touch with their family?  If so, how is their relationship?  If not, why not?  Does the patient have friends? Did they ever have friends? What happened to them?  Finally, we can ask who raised the patient growing up and what their childhood was like.  Were they sexually, physically, or emotionally abused?  Note that the order of these questions is flexible and is based on the individual patient.  I try to ask what I believe to be the most important questions, for that specific patient, first.

To conclude the interview, ask the rest of the team if they have any questions for the patient.  Then ask the patient how we can help them while they are in the hospital, and if they have any questions for us.  After answering their questions, usually the attending will discuss the treatment recommendations and have them sign medication consent and the treatment plan paperwork.

That’s pretty much it.  Simple but requires a bit of practice.  Remember, as far as I know, nobody is born knowing how to conduct a good psychiatric interview.  The psychiatric interview involves a very different way to interact with people than your normal, every-day interactions.  Like everything else, it requires plenty of practice and determination to master.

As a final note, at times you will encounter a patient who minimizes all symptoms and appears calm and collected during the interview. However, you may get a sense that underneath the facade there may lie a psychiatric illness.  For example, a student believing that a patient may get angry if asked the wrong question may “walk on eggshells” to avoid any conflict.  They may then conclude in their note that the patient is calm, implying that the patient may be soon ready for discharge. However, better to test the patient’s impulse control in the hospital than in the community.  Our job is to make sure that the patients and the community is safe, and therefore the patient must be able to control their impulses when people in the community challenge them. Towards the end of the interview, I may sometime nudge the patient a bit to assess their reaction.  I do not mean that I continue to shove until they throw a chair at me, but I do want to test their range of emotions and impulse control.

If you have any additional tips or beef with anything I said, please let me know in the comments!  I appreciate all feedback.

The Psychiatric Interview – Starter Pack

Omer Liran


Omer Liran is an attending inpatient psychiatrist at Olive View-UCLA Medical Center and an Assistant Professor at UCLA School of Medicine.


Post navigation


5 thoughts on “The Psychiatric Interview – Starter Pack

  1. Thank you so much for taking the time to write this introduction. Thank you especially for including how to word questions in layman’s terms.

  2. This is extremely useful for medical students and will be very helpful in my first psychiatric interview this week. It is greatly appreciated. Thank you very much!

  3. Thank you very much for this introduction! I found really helpful to think that we are playing the role as a detective and we need to ask logical questions to dig deep into the situation. It’s also wonderful to see some examples of layman’s term questions. Great read!

Leave a Reply

Your email address will not be published. Required fields are marked *