Difficult Cases, Part 1: Before You Increase the Quetiapine, Start With ABC
Approximately 40% of Eva's Tea readers have asked for more content on difficult or complex psychiatric cases.
A difficult case might be an older adult with dementia who suddenly becomes aggressive during morning care. It might be a patient with schizophrenia who repeatedly calls 911 despite appearing calm minutes later. It could be an adolescent refusing school, an adult with unexplained behavioral escalation, or a nursing home resident described simply as being "out of control."
Different patients. Different diagnoses. Different settings.
But many difficult cases share the same problem: We start searching for the solution before we have gathered the right information.
The medication list comes up.
The diagnosis is reviewed.
Someone asks whether the quetiapine can be increased.
And before long, we are discussing an intervention for a behavior that nobody has clearly defined.
I have seen some version of this conversation throughout my career:
"Can we go up on her quetiapine? She's completely out of control today."
My first thought is almost always the same:
What does "out of control" actually mean?
Did she strike someone?
Was she yelling?
Did she throw something?
Was she frightened?
Did she refuse care?
What changed?
Those questions may sound basic. They are.
They are also where good clinical reasoning often begins.
For the first article in this series on difficult psychiatric cases, I want to start with one of the simplest and most useful behavioral frameworks available to us:
ABC: Antecedent → Behavior → Consequence.
It is not a diagnostic test. It will not replace your psychiatric assessment, medical differential, mental status examination, or clinical judgment. But it can force us to do something that becomes surprisingly difficult when a case feels chaotic:
☕ Slow down the story long enough to see what is actually happening.
And sometimes, that changes everything.
Before ABC Comes the Relationship
There is an important reality we need to acknowledge before discussing the framework itself.
The quality of your clinical reasoning is partly dependent on the quality of your information.
And in many psychiatric settings, particularly long-term care, that information comes from other people.
The patient may have advanced dementia, psychosis, aphasia, intellectual disability, poor insight, or simply limited ability to explain what occurred. Family may not have witnessed the event. The medical record may contain only a brief note written after the fact.
That leaves the people who were there.
The nurse.
The CNA.
The caregiver.
The teacher.
The parent.
These individuals may hold the most important pieces of the clinical puzzle. But having access to an information source does not automatically mean you will get good information.
You need a relationship.
This is especially important when the person giving you information is also asking you to do something.
"Can you increase her medication?" "He needs something stronger." "We can't keep doing this."
These statements often arrive with frustration, fear, exhaustion, or legitimate concern about safety. If your immediate response sounds like a challenge—
"Why would we increase the medication?"
or a rejection
"I'm not changing anything without more information."
You may technically be correct, but you may also have made your assessment more difficult.
The staff member now feels dismissed.
You feel pressured.
The patient is somewhere in the middle.
This is where relationship and clinical reasoning intersect.
☕☕ I have learned that before I can challenge a conclusion, I often need to show that I respect the concern behind it.
That doesn't mean agreeing that the medication should be increased. It doesn't mean endorsing an inaccurate description of the behavior. And it certainly doesn't mean prescribing simply to reduce the discomfort of the moment.
It means creating enough alignment to gather better information.
Something as simple as this can change the conversation:
"I hear you. It sounds like today has been difficult, and I know you're concerned about safety. Before we decide what needs to change, walk me through exactly what happened."
That response does three things.
It acknowledges the concern.
It establishes a shared purpose.
And then it redirects the conversation from conclusion to observation.
That last part is critical.
Because "increase the quetiapine" is a conclusion.
"She became tearful and yelled when staff attempted to assist her into the shower, and she stopped yelling when the shower was postponed" is clinical information.
Now we have something to work with.
The Most Basic Assessment We Keep Forgetting
The ABC model asks three questions:
Antecedent: What happened immediately before the behavior?
Behavior: What did the patient actually do?
Consequence: What happened immediately after the behavior?
That's it.
Simple enough to teach in a few minutes.
Powerful enough to change a treatment plan.
The ABC model is rooted in behavioral theory and functional analysis. One of the seminal contributions to the field came from Iwata and colleagues, whose work demonstrated that behaviors that may look similar on the surface can serve very different functions depending on the conditions surrounding them.
That is an important concept for psychiatric clinicians.
Two patients can display the same behavior for entirely different reasons.
Two patients can yell.
One may be attempting to escape an overwhelming demand.
Another may be frightened by a hallucination.
One may be in pain.
Another may have learned that yelling is the fastest way to bring someone into the room.
The behavior alone does not tell us its function. We need context.
At its core, ABC recognizes that behavior does not occur in isolation. It happens within a context. Events precede it. Responses follow it. And sometimes those responses unintentionally make the behavior more likely to happen again.
The purpose of ABC is not to blame the patient, staff, family, or environment.
The purpose is to identify patterns.
What tends to happen before the behavior?
What exactly does the behavior look like?
What changes after it occurs?
Those questions move us away from vague psychiatric language and toward observable clinical data.
Consider these two reports:
"She's extremely agitated and impossible to manage."
Versus:
"When staff entered her room at 6:30 a.m. and turned on the overhead light to begin morning care, she yelled, pushed the aide's hand away, and attempted to get out of bed. Staff left the room for 20 minutes, returned with a familiar CNA, and she accepted care."
The first report creates urgency.
The second creates possibilities.
Was she awakened abruptly?
Was she frightened?
Was the lighting disorienting?
Was she in pain?
Was there something different about the first caregiver?
Was 6:30 a.m. too early for this particular resident?
Did leaving the room reduce overstimulation?
Did the familiar CNA provide reassurance?
Could the behavior have functioned, at least in part, to escape or delay an unwanted demand?
Now we are thinking.
A Is for Antecedent: What Happened Right Before?
The antecedent is what occurred immediately before the target behavior.
Sometimes it is obvious:
A staff member attempts a shower.
A parent takes away a phone.
A teacher assigns a group project.
A patient is told discharge will be delayed.
A resident is redirected away from an exit.
Other times, the antecedent is more subtle.
A shift change occurred.
The environment became noisy.
A preferred caregiver went home.
The patient had not slept.
The patient was constipated.
The room was cold.
The person misinterpreted someone's approach.
The demand was made too quickly.
The patient had been sitting alone for three hours.
One of the most useful questions in a difficult case is also one of the simplest:
"What was happening in the five minutes before this started?"
I prefer that question to, "What triggered the behavior?"
Why?
Because trigger encourages interpretation.
"What was happening in the five minutes before?" encourages observation.
There is a difference.
A nurse might say:
"She was triggered because she's psychotic."
That may or may not be true.
But ask what happened in the preceding five minutes, and you may learn:
"Her daughter called and told her she wasn't coming to visit. About two minutes later, she began pacing and yelling that people were abandoning her."
That is extremely useful information.
The antecedent does not necessarily explain the entire behavior. But it gives us a place to begin.
B Is for Behavior: What Did the Patient Actually Do?
This is where psychiatric descriptions often become dangerously vague. Agitated.
Even worse, someone calls me and says the patient is "behavioral."
Or my personal favorite:
"Very, very behavioral."
That one gets me every time. I have learned that I need to pause and slow myself down before responding.
Other favorite vague descriptor’s include:
Aggressive.
Combative.
Manipulative.
Attention-seeking.
Sundowning.
These words may communicate that something is wrong, but they often tell us remarkably little about what actually happened.
If I tell you a patient was "aggressive," what do you know?
Did he raise his voice?
Clench his fist?
Throw a cup?
Strike a staff member?
Threaten someone?
Push a wheelchair?
Swing his arms while being dressed?
These are not interchangeable behaviors.
They have different levels of risk, different possible causes, and potentially very different interventions.
The B in ABC forces us to operationalize the behavior.
Instead of:
"He became aggressive."
Try:
"When the CNA attempted to change his shirt, he grabbed her wrist and pushed her hand away."
Instead of:
"She was out of control."
Try:
"At 3 p.m., she yelled continuously for approximately 15 minutes, struck the wall twice with an open hand, and attempted to leave the unit."
Instead of:
"He's attention-seeking."
Try:
"He called the nurses' station eight times between 7 p.m. and 9 p.m., primarily when alone in his room."
Now we have behaviors we can assess.
And importantly, we have behaviors we can measure.
If an intervention is started tomorrow, how will we know whether it worked if the original target was simply "agitation"?
But if the target is:
"Calling the nurses' station eight times in two hours when alone in his room"
we can track frequency, timing, context, and response to intervention.
Good behavioral description makes treatment more precise.
It also makes documentation better.
C Is for Consequence: What Happened Next?
This is the part of ABC that clinicians sometimes overlook.
We naturally ask what happened before a behavior. We are less likely to examine carefully what happened afterward.
But the consequence can be enormously important because it tells us what changed as a result of the behavior.
The resident yelled, and the shower was stopped.
The patient threatened to leave, and three staff members immediately surrounded him.
The adolescent refused school, and the parent allowed him to stay home.
The patient called 911, and suddenly received prolonged attention from staff, EMS, and the emergency department.
None of this means the behavior was consciously planned.
That distinction matters.
A patient does not need to think:
If I yell, they will cancel my shower. The behavioral pattern can still develop.
If yelling repeatedly results in an unwanted demand disappearing, the behavior may become more likely the next time that demand occurs.
Likewise, if a patient who feels intensely lonely receives his greatest amount of human contact only after a crisis, crisis behavior may inadvertently become associated with connection.
This is why the question is not:
"Is the patient doing this on purpose?"
The better question is:
"What changes for the patient after this behavior occurs?"
That shift in thinking can open an entirely different treatment pathway.
Let's Return to the Quetiapine
Imagine a nurse tells you:
"Mrs. J is completely out of control during morning care. We need to increase her quetiapine."
You could review the medication list, assess the current dose, and consider an increase.
Or you could start with ABC.
Antecedent: Staff enter at approximately 6:30 a.m., turn on the overhead light, remove the blankets, and begin preparing Mrs. J for a shower.
Behavior: Mrs. J yells, swears, pushes staff away, and attempts to pull the blankets back over herself.
Consequence: Staff stop the shower attempt and leave the room. They return later in the morning, often with a different CNA, and Mrs. J is usually more cooperative.
Suddenly, the case looks different.
Perhaps the quetiapine is inadequate.
But perhaps Mrs. J is being awakened before she is ready.
Perhaps she is frightened by the abrupt approach.
Perhaps she has pain with movement.
Perhaps the overhead lighting is disorienting.
Perhaps she does better with a familiar caregiver.
Perhaps she is resisting the shower specifically rather than experiencing generalized psychiatric deterioration.
Or perhaps several of these factors are operating simultaneously.
ABC doesn't give us the diagnosis.
It gives us better questions.
And better questions are often the beginning of better treatment.
This matters particularly in dementia care, where the reflex to escalate an antipsychotic deserves careful scrutiny. The 2023 AGS Beers Criteria recommend avoiding antipsychotics for behavioral problems of dementia or delirium unless non pharmacological options have failed or are not possible and the older adult is threatening substantial harm to themselves or others.
The evidence also suggests that this is more than a philosophical preference for nonpharmacologic care. A large network meta-analysis of 163 randomized clinical trials involving more than 23,000 people with dementia found that nonpharmacological interventions appeared more efficacious than pharmacological interventions for reducing aggression and agitation.
That does not mean antipsychotics are never appropriate.
They can be.
It means that "increase the quetiapine" should not be the beginning of the assessment.
ABC Is Not Just for Dementia
Although the ABC framework is frequently associated with dementia care and applied behavior analysis, the underlying clinical reasoning can be useful across psychiatric populations and across the lifespan.
Consider an adolescent refusing school.
Antecedent: A group presentation is scheduled at school.
Behavior: The adolescent reports severe stomach pain, cries, and refuses to get dressed.
Consequence: The parent calls the school and allows the adolescent to remain home.
The function might involve escape from a socially threatening situation. That does not mean the adolescent is "faking" anxiety. The anxiety may be very real. ABC simply helps us see the cycle that could maintain school avoidance.
Or consider an adult psychiatric inpatient.
Antecedent: The patient spends several hours alone with minimal staff interaction.
Behavior: He begins yelling bizarre statements loudly in the hallway.
Consequence: Multiple staff members immediately approach, engage him, and remain nearby.
Again, ABC does not prove that attention is maintaining the behavior. But it gives us a hypothesis worth testing.
Could scheduled positive contact reduce the need for crisis-driven contact?
That is a much more interesting clinical question than simply asking whether we should increase the antipsychotic.
Where ABC Fits Into a Bigger Psychiatric Formulation
For clinicians familiar with the 4 Ps—predisposing, precipitating, perpetuating, and protective factors—the ABC framework fits particularly well within the perpetuating part of the formulation.
The 4 Ps ask broader questions:
Why is this patient vulnerable?
Why did this problem emerge now?
What is maintaining it?
What strengths or protections might help?
ABC zooms in much closer.
It asks:
What happened immediately before this specific episode?
What exactly did the patient do?
What happened immediately afterward?
You can think of the 4 Ps as the wide-angle lens and ABC as the zoom lens.
The patient's trauma history may be a predisposing factor.
The death of a spouse may be a precipitating factor.
But if every episode of yelling occurs when the patient is left alone after dinner and immediately results in prolonged one-to-one staff attention, ABC may reveal something important about what is maintaining the current pattern.
This is what makes ABC so actionable.
History helps explain how the patient arrived here.
ABC may help us understand what is happening now.
And sometimes your initial ABC hypothesis will simply be wrong. That's okay.
The goal is not to force every patient into a three-column chart.
The goal is to develop the reflex of asking:
What happened before?
What actually happened?
What happened next?
Those questions do not replace sophisticated psychiatric reasoning. They improve it.
The Information Source Matters
There is one final piece that brings us back to where we started.
You can know the ABC framework perfectly and still get poor data.
Imagine asking a busy travel nurse:
"What was the antecedent to the behavior?"
You may get a blank stare.
Or:
"There was no antecedent. She's just like that."
Fair enough.
So ask differently.
"Walk me through exactly what happened."
Then:
"What was she doing right before that?"
Then:
"And what did everyone do after she started yelling?"
Now you are gathering ABC data without turning the conversation into a lecture on behavioral analysis.
This is where your relationship with the information source matters.
Staff members are much more likely to give you useful details when they believe you respect their observations and understand the realities of their work.
And if you plan to challenge the requested intervention, that relationship becomes even more important.
Imagine the difference between:
"I'm not increasing the quetiapine. This sounds medical."
And:
"I can see why you're concerned. So, based on what you described, I want to first look at why this is happening specifically during morning care, because that pattern may give us a better target than simply increasing medication. Let's try changing the approach and timing, track what happens over the next several days, and then reassess together."
Same medication decision.
Very different collaboration.
That is how clinical reasoning becomes clinical leadership.
From ABC to a Shared Plan
The best assessment in the world accomplishes very little if nobody buys into the plan.
This is especially true when your intervention depends on other people.
You may decide that a resident should be approached later in the morning, by a familiar caregiver, with less environmental stimulation and more time to process one-step instructions.
But you are probably not the person providing morning care tomorrow.
Someone else has to carry out that plan.
That means collaboration is not an optional soft skill.
It is part of the intervention.
A useful approach is to make the reasoning visible:
"Here's what I'm hearing. The behavior seems to happen most often when morning care starts early and quickly. The priority is figuring out whether changing that approach reduces the episodes before we expose her to more medication. Here's what I'd like us to try, and here's what I need you to watch for."
That closes the loop.
The Clinical Habit I Want You to Take From This
If you remember nothing else from this article, remember this:
Before you intervene, define what actually happened.
When someone tells you a patient is agitated, ask what that looked like.
When someone tells you the behavior came "out of nowhere," ask what was happening five minutes earlier.
When someone tells you a medication isn't working, ask what specific target symptom is persisting.
When someone asks you to increase the quetiapine, resist the temptation to start with the dose.
Start with the story.
Then break the story into three parts:
Antecedent. Behavior. Consequence.
What happened before?
What did the patient actually do?
What happened next?
The answers will not solve every difficult case.
But they will often reveal that the case you thought you were treating is not actually the case in front of you.
And that is exactly where this series will go next.
Gather better information.
Define the behavior.
Look for patterns.
Separate assumptions from observations.
Consider medical, psychiatric, cognitive, environmental, and interpersonal contributors.
Then build a plan that people can actually follow.
Because difficult cases do not always require a more complicated answer.
Sometimes they require a better question.
And before you increase the quetiapine, ABC is a very good place to start.
Selected Readings
American Geriatrics Society Beers Criteria® Update Expert Panel. (2023). American Geriatrics Society 2023 updated AGS Beers Criteria® for potentially inappropriate medication use in older adults. Journal of the American Geriatrics Society, 71(7), 2052–2081.
Contreras, B. P., Tate, S. A., Morris, S. L., & Kahng, S. (2023). A systematic review of the correspondence between descriptive assessment and functional analysis. Journal of Applied Behavior Analysis, 56(1), 125–140.
Hurl, K., Wightman, J., Haynes, S. N., & Virués-Ortega, J. (2016). Does a pre-intervention functional assessment increase intervention effectiveness? A meta-analysis of within-subject interrupted time-series studies. Clinical Psychology Review, 47, 71–84.
Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & Richman, G. S. (1994). Toward a functional analysis of self-injury. Journal of Applied Behavior Analysis, 27(2), 197–209. (Reprinted from Analysis and Intervention in Developmental Disabilities, 2, 3–20, 1982).
Lanovaz, M. J., Argumedes, M., Roy, D., Duquette, J. R., & Watkins, N. (2013). Using ABC narrative recording to identify the function of problem behavior: A pilot study. Research in Developmental Disabilities, 34(9), 2734–2742.
Moniz Cook, E. D., Swift, K., James, I., Malouf, R., De Vugt, M., & Verhey, F. (2012). Functional analysis-based interventions for challenging behaviour in dementia. Cochrane Database of Systematic Reviews, (2), CD006929.
Siegel, M., McGuire, K., Veenstra-VanderWeele, J., Stratigos, K., King, B., Bellonci, C., Hayek, M., Keable, H., Rockhill, C., Bukstein, O., Walter, H. J., & McClellan, J. (2020). Practice parameter for the assessment and treatment of psychiatric disorders in children and adolescents with intellectual disability (intellectual developmental disorder). Journal of the American Academy of Child & Adolescent Psychiatry, 59(4), 468–496.
Tamayo-Morales, O., Patiño-Alonso, M. C., Losada, A., Gómez-Marcos, M. A., & García-Ortiz, L. (2021). Behavioural intervention to reduce disruptive behaviours in adult day care centre users: A randomized clinical trial (PROCENDIAS study). Journal of Advanced Nursing, 77(4), 1919–1930.
Watt, J. A., Goodarzi, Z., Veroniki, A. A., Nincic, V., Khan, P. A., Ghassemi, M., Thompson, Y., Lai, Y., Treister, V., Tricco, A. C., & Straus, S. E. (2019). Comparative efficacy of interventions for aggressive and agitated behaviors in dementia: A systematic review and network meta-analysis. Annals of Internal Medicine, 171(9), 633–642.