When the topic of agitated patients in the ED is brought up, most people conjure up an image of a psychotic manic patient or tell the story of a patient brought in after ingesting a combination of methamphetamine and bath salts. Acutely agitated patients like this are a danger to themselves and anyone near them. Assessing and treating these undifferentiated patients in the ED can be a formidable task even if it is possible to get a collateral history and perform more than a cursory exam. Making these presentations even more challenging, they are often agitated for clinically important reasons that can result in death.
While it is important to discuss patients with extreme agitation, we see a very broad spectrum of agitated patients in the ED. They range from docile elderly patients with psychosis to the patients described above. To anticipate and prevent the escalation of agitation we need an approach that will allow us to assess a patient’s current level of agitation and address it in a way that recognizes the potential for further escalation and avoids it.
Agitation scales have been created for ICU patients, delirious cancer patients, psychiatry inpatients, etc, but none have been developed or validated for use in the emergency department. While the American Association for Emergency Psychiatry (AAEP) do “not consider one agitation rating scale to be better than another” in their Consensus Statement on the Evaluation and Treatment of the Agitated Patient, they note that they “find the BARS is easy to use reliably, even for one not trained in psychiatry or emergency medicine” which seems like a decent endorsement to me. BARS is the “Behavioral Activity Rating Scale,” a scale developed by pharmaceutical companies to assess agitation in drug trials. It divides patients into 7 levels of agitation:
As far as I know, this scale has never been studied in the context of undifferentiated emergency department patients. It could certainly be criticized for defining levels of agitation by their required treatment (ie – need for restraint) rather than symptoms. However, I think its high interrater reliability (as demonstrated in this study) confirms that we can classify a patient’s level of agitation without a complicated scale that takes 15-20 minutes of observation to use (OASS I’m looking at you). We know that an intoxicated patient who is sleeping has the potential to get upset when they wakes up, we can tell if a patient is agitated very quickly by speaking with them, we appreciate when a patient is agitated enough that they may need something to calm them down, and we recognize when a patient is beyond reason and needs to be restrained quickly.
I’m not going to do a study on it, but I think this scale could be simplified to unagitated (4), mildly agitated (5), moderately agitated (6) and extremely agitated (7) to define the levels of agitation that can be addressed with various treatment options. The treatment of unagitated and mildly agitated patients will be discussed in this post. A post on moderately and severely agitated patients in the ED will follow early next week.
Most of the patients in the ED are not agitated, however, some of them are at higher risk of becoming agitated than others. While no treatment is required for an unagitated patient, assessing their potential for agitation in the short-term may allow efforts to be made to minimize agitating stimuli and prepare for potential aggression.
This 2008 American study of 43093 adult American civilians that completed face-to-face surveys ~3 years apart (70.3% cumulative response rate) assessed the characteristics of people that made them most likely to commit violent acts. While this information is not completely applicable to an ED presentation of agitation, it make sense to me that a patient at risk to commit violent acts in general would also be at risk to commit violent acts in the ED where they may have additional reasons for agitation. The study found that historical (past violence, juvenile detention, physical abuse, parental arrest record), clinical (substance abuse, perceived threats), dispositional (age, sex, income) and contextual (recent divorce, unemployment, victimization) factors played a significant role in predicting violence. While severe mental health conditions (schizophrenia, major depression, bipolar) were a significant predictor in the univariate analysis (likelihood of violence in people with vs without a severe mental health condition), they were not an independent predictor when these other factors were taken into account in the multivariate analysis. Notably, the items that correlated most strongly with an increased risk of violence were young age (<43 years old), male sex and a history of violence.
This can be interpreted to mean that patients with severe mental health conditions are more likely to have other risk factors making them likely to commit violent acts, but in isolation a severe mental health condition does not make someone more likely to commit a violent act. Unfortunately, while this study characterized the epidemiological likelihood of an individual to commit a violent act, many of these variables will be unknown upon presentation to the ED and the study did not specifically assess characteristics that make an ED patient likely to become violent in the ED.
While researching this post I requested FOAM resources related to agitation on twitter and Minh Le Cong sent me several useful links. One of them was a talk he gave on Psychiatric Aeromedical Retrieval that touched on the tool that is used in Queensland to assess a mentally ill patient’s level of risk and advise appropriate restraint strategies. While this tool has not been validated, it weighed several of the risk factors discussed in the study above while paying particular attention to expressions of agitation/anger, history of violent behavior, and state of intoxication/withdrawal. If you’re involved in aeromedical retrieval of psychiatric patients I highly recommend you checking out Minh‘s entire lecture on this topic for yourself here.
So what’s the bottom line on the assessment of agitated patients in the ED? Unfortunately, there is not a lot of good literature. However, there is evidence that a combination of youth, a history of violence, and altered mental status (secondary to acute psychosis, intoxication, withdrawl or other) increase the likelihood that a patient will become agitated.
The mildly agitated patient can be calmed down with verbal deescalation and by fixing the easy stuff. The AAEP have an entire Consensus Statement devoted to the former and refer to the latter as “environmental modification.”
In their consensus statement on this topic the psychiatrists state that:
“in an emergency department, both the clinician and patient can slip into irrational thinking or expediency at the price of engaging each other. A clinician who has many patients to see and too little time may prematurely use medication to avoid verbal engagement.”
Are we too quick to jump to the drugs? Perhaps sometimes. I suspect that emergency physicians’ skill and patience vary markedly in this area. However, I don’t see psychiatrists lining up around the block to complete initial assessments on these patients prior to them being calmed and medically cleared of the many things that could result in this presentation that may also kill them. Graham Walker went on a rant about medical clearance for psychiatric patients on ERCast Rant-off 2013 recently that fires a few shots in the other direction.
The AAEP Consensus Statement noted that there was limited data to guide verbal deescalation strategies but did provide recommendations based on expert opinion. Much of it was common sense, for example:
However, I did find a couple of descriptions of Jedi psychiatrist powers that I will attempt to adopt into my own practice:
Agree with the truth (Patient complains about temperature, you say “Yes, it is quite cold in here, could we get you a blanket?”)
Agree in principle (Patient believes they have been treated unfairly, you say “I believe that everyone should be treated fairly”)
Agree with the odds (Patient upset about waiting to see the psychiatrist, you say “Odds are other patients would probably be upset as well”)
The AAEP Consensus Statement aptly suggests offering the patient choices to help give them a semblance of control over their circumstances and help to decrease their agitation. They note that “Food and something to drink may be a choice the patient is willing to accept that will stall aggressive behaviors.” These items have the added benefit of being cheap, readily accessible and abundant in most ED’s. We should use them liberally.
Finally, some actual evidence from the ED! The PICO question for this double-blind RCT was: In a group of agitated, nicotine-dependent schizophrenic patients, does nicotine replacement therapy decrease agitation relative to placebo?
They found that nicotine replacement reduced agitation on the Agitated Behavior Scale significantly in schizophrenic patients with both low and high nicotine dependence at both 4 and 24 hours. This reduction was quantified both in the relative (~20-30%) and absolute (9-10%) difference in the agitation score. While this was only a single small study (40 patients), the significance of its findings, physiologic plausibility, lack of harm and ease of implementation have me convinced that nicotine replacement therapy should be offered earlier to agitated, nicotine-dependent patients.
Waiting to be seen in the ED is an almost universal problem. We’ve all seen some normally level-headed people nearly lose it while waiting for themselves or their loved ones to be seen. It’s no surprise that patients who are already agitated can become even more agitated while waiting. In 2012 this phenomena was studied and, surprise, surprise: seclusion, chemical restraint and physical restraint all correlated with the ED census. More than just another way to demonstrate that overcrowded ED’s harm patients, this has implications for their management. I would argue that for those patients at high risk of increasing aggression (recall: young, male, history of violence, altered mental status), it may make sense to invest time in seeing them early to prevent their agitation from escalating to a point that will require more resources down the line.
While we often focus on the most out of control patients, patients with a broad spectrum of agitation levels present to the ED. In order to prevent escalation we need to match our interventions with their level of agitation. This post reviewed the BARS scale for quantifying levels of agitation, discussed the risk factors for a patient becoming increasingly agitated and offered some pearls for how to safely calm the mildly agitated patient. My next post will discuss the treatment of moderately and severely agitated patients.
Thanks for reading! If you found this post helpful please tweet/retweet it on twitter, follow me on twitter, share it on facebook, or sign up for e-mail notifications about new posts. Links to all of these options are at the top of the sidebar.