Without evidence of benefit, an intervention should not be presumed to be beneficial or safe.

- Rogue Medic

To Restrain or Not To Restrain, But That’s Just the Beginning of the Question

In a JEMS article,[1] Dr. Keith Wesley reviews a study of an education program on the use of restraints in EMS.[2] This is a one hour teaching module inserted into their paramedic curriculum. The lecture part of it is taught by a physician, even though there is no chemical restraint portion to the class. This is just one hour, but appears to include a pre-test, a lecture, 5 video scenarios, a 14 point module that covers all of this, and a post-test. To me, this seems like a lot to cover in an hour. It seems a bit too ambitious.

If you really want people to understand patient restraint, there is no substitute for a real violent patient. Not a teacher, or another student, pretending to be violent. How much time would it take at a psychiatric facility before a student would have exposure to a violent patient?

Oh, my! You would endanger a student by intentionally putting the student in a violent situation?

Of course. What do you think they will be doing once they are working with real patients?

In a psychiatric facility, there should be plenty of people around to assist with restraining patients. The goal is to have some experience with this before being let loose on real patients without adequate backup.


Emergency medical services (EMS) providers may encounter agitated and violent patients,1-7 and these encounters can result in significant injury to the patient and to EMS personnel.2,3,7,8 In one retrospective descriptive study by Mechem et al., EMS workers in a large urban EMS system submitted 1,100 injury reports during a two-year period. Of these, 44 (4.0%) injury reports were the result of an assault. Paramedics were assaulted in 35 (79.5%) of these incidents and firefighters in nine (20.5%). Forty-one assaults (93.2%) occurred during patient care activities.9 In our own Metro EMS system, which responds to approximately 65,000 calls per year, EMS personnel also frequently face violence from agitated patients.10 EMS providers restrain agitated patients to ensure the protection of providers, to protect the patient from injury, or to facilitate delivery of medical care.1-3,6,7 [3]


The study by Dr. Mechem is from Philadelphia. The study looking at the teaching module is from Pittsburgh. Opposite sides of the same state. For the past few years, they have supposedly been working off of the same protocols. Local medical directors can make their protocols more restrictive. Dr. Mechem is likely to do that. From what I have heard of Pittsburgh, that is less likely, but I do not know how either has handled these protocols. Why do I mention the protocols, since we are looking at a teaching module that does not address chemical restraint?

This study is to be part of a multiphase prehospital restraint use study determined to evaluate the effectiveness of various interventions in reducing patient agitation and resulting assaults on EMS personnel. In the future, we are adding chemical restraints to the system protocols and will add this to the educational module. It was a limitation of the study that chemical restraint information was not included in the module.[3]


As they mention in the study, this is just the beginning of what they are doing. chemical restraints were added to the Pennsylvania ALS (Advanced Life Support) Protocols in November 2008.

The addition of the ability to chemically restrain a patient without calling command for orders is also a start. Pennsylvania seems to be trying to take an evidence based approach to EMS treatment. Unfortunately, it is mostly by slow baby steps.

From the Agitated Behavior/Psychiatric Disorder Protocol –

Contact Medical Command, if possible

If continued struggling,2 Administer Sedation
(See box below)

Monitor continuous ECG and Pulse Oximetry, when feasible

Sedation Options:
(Choose one)

Lorazepam 1-2 mg IM/IV/IO 3 (0.1 mg/kg, max 2 mg/dose)
may repeat every 5 minutes until maximum of 4 mg


Diazepam 5-10 mg IM/IV/IO 3 (0.1 mg/kg)
may repeat every 5 minutes until maximum 0.3 mg/kg


Midazolam 1-5 mg IM/IV/IO 3,4 (0.05 mg/kg)
may repeat every 5 minutes until maximum of 0.1 mg/kg


Protocol Footnotes:


2. Do not permit patient to continue to struggle against restraints. This can lead to death due to severe rhabdomyolysis, acidosis, dysrhythmia, or respiratory failure. Medical command should be contacted for possible chemical restraint with sedative medication.

3. If age > 65, reduce doses of sedative benzodiazepines in half.

4. Regional or service policy may permit intranasal midazolam, but this may not be as effective as parenteral medications.[4]


The maximum doses for restraint are not adequate. I have doubled the dose and not caused any decrease in room air oxygen saturation. This is just a start by the state. Perhaps they will improve the protocol as they see it in use. As they see it fail to control patients. As they see EMS and patients hurt, due to confidence that the maximum dose would be enough. If the goal is to protect EMS, police, family, and patients, the dose needs to be capable of actually causing sedation of the agitated patient. These doses may be effective on the sleepy patient, or if the patient has been thoughtful enough to pre-treat himself with alcohol.

According to the medical directors I have asked about restraint, Pennsylvania will never approve of the use of haloperidol (Haldol) or droperidol (Inapsine) for chemical restraint. but these are topics for another post.

Back to the topic at hand. Dr. Wesley states:

Violent patients represent a major risk to you and are a potential of great liability to EMS and law enforcement. This study is, I hope, just the first in what should be a multiphase, multi-center trial. The authors readily recognize its limitations.[5]



The worst thing that can happen is for educators and curriculum writers to read the conclusion and dismiss the value of including such a module into both initial and refresher education merely because it showed no change in behavior in this one small group of students. With the mantra of “Is the scene safe, BSI” forever emblazoned in our minds, I believe it’s the violent patient for whom we are unprepared that is more likely to harm us than any germ, virus or downed power line.[5]


Dr. Wesley has made some great points. In Pennsylvania, I think they are still unprepared, but they are improving. Years ago, when I was on a protocol committee, I was told that we would never have standing orders for opioids. There are now standing orders for morphine and fentanyl, except where the local medical director refuses to allow standing orders, so that the medical director can keep incompetent medics working.

I was told that we would never have a protocol for chemical sedation. There is now a statewide protocol for chemical sedation, except where the local medical director refuses to allow use of this protocol, again so that the medical director can keep incompetent medics working.

We need to put more emphasis on the safety of the patients and the safety of those treating the patients. We need to decrease the emphasis on the ability of the medical director to allow a dangerous paramedic to treat patients. Inappropriately limiting the treatments available to patients, to what the medical director thinks the least common denominator paramedic can use without killing patients, is bad medicine.

I have written about these dangerous medical directors here, here, here, here, and here.


[1] To Restrain or Not To Restrain
Keith Wesley, MD, FACEP
Street Science
2008 Dec 15

[2] Impact of a restraint training module on paramedic students’ likelihood to use restraint techniques.
Campbell M, Weiss S, Froman P, Cheney P, Gadomski D, Alexander-Shook M, Ernst A.
Prehosp Emerg Care.
2008 Jul-Sep;12(3):388-92.
PMID: 18584509 [PubMed – indexed for MEDLINE]

[3] Impact of a restraint training module on paramedic students’ likelihood to use restraint techniques.
Same as footnote 2.

[4] Pennsylvania EMS Statewide ALS Protocols
Effective November 2008
Pages 100/121 to 102/121 in the pdf page counter
Page with link to the full text PDF of the protocols.

[5] To Restrain or Not To Restrain
Same as footnote 1.



  1. […] – comment Mon, 02 Feb 2009 20:00:49 +0000 By Rogue Medic Leave a Comment In the comments to To Restrain or Not To Restrain, But That’s Just the Beginning of the Question, jeg43 wrote, I am astonished that restraint is an issue in this day and […]