Dr. Nordstrom is emergency psychiatrist at Denver Health Medical Center and outpatient psychiatrist at Colorado West Regional Mental Health Center. She is also an instructor with the University of Colorado, Denver. Dr. Allen is Director of Research at the University of Colorado, Denver Depression Center.

Disclosure: Dr. Nordstrom is consultant to Alexza Pharmaceuticals. Dr. Allen is consultant to Alexza; on the speaker’s bureau of Pfizer; receives research support from Janssen, the National Alliance for Research on Schizophrenia and Depression, and the National Institute of Mental Health (NIMH); and receives honoraria from AstraZeneca and the NIMH.

Please direct all correspondence to: Kimberly Nordstrom, MD, JD, PO Box 770433, Steamboat Springs, CO 80477; Tel: 970-819-1736; Fax: 970-879-8825; E-mail: nordstrom_kimberly@yahoo.com.



Focus Points

• All clinical settings for psychiatric patients have the potential of being dangerous.

• Some psychiatric patient characteristics have been correlated with a higher risk for violence.

• Psychiatrists and clinicians need to be aware of potential dangerousness and have a plan in place.

• Offices should provide education on de-escalation techniques.



It is well understood that nursing staff and physicians face potentially dangerous situations in giving care to patients. Studies have shown that patient assaults cause short- and long-term consequences for the staff, including anxiety disorders and burn-out. This topic has been well discussed in the literature but focuses on inpatient settings. The authors of this article ferret out some of the differences between inpatient versus outpatient—including the differences in staffing and physical setting. Thought is given to situations that are uniquely different with the hope that clinicians might not take security for granted and possibly implement effective measures that help prevent an event from occurring. As this is not always possible, creating a protocol for likely emergent situations can be helpful “in the moment” when thoughts are not always clear. After an incident has occurred, debriefing is critical to help prevent a related anxiety disorder and possibly reduce burn out. Administrative review can be useful to help reduce further incidents.



Being in the healthcare field can pose physical risks to clinicians providing direct patient care. Patients can become verbally or physically violent secondary to a wide assortment of diagnoses and in a range of environments from emergency departments to hospital wards and nursing homes. Studies have focused on aggressor and victim characteristics; weapons screening; education; and after-effects that consist of decreased job satisfaction, burn out, and, for some, posttraumatic stress disorder (PTSD).1 It is unclear in the literature how all of this relates to the outpatient setting. There is very little that the emergency department and inpatient environment have in common with the outpatient environment. The situations, obstacles, and even responses can prove challenging. Except for highlighted tragedies in media reports, there is relatively little written about this topic. With this in mind, the authors think this is a topic that needs to be more thoroughly discussed.



The actual patient emergencies in the outpatient setting are not uniquely different from those in the hospital, but the setting can create a more challenging situation. Patients can present acutely suicidal, homicidal, intoxicated, or drug seeking. They can have the same diagnoses and level of severity in the inpatient or outpatient settings. Some outpatient populations may be higher risk than others. The risk may increase if the psychiatrist has a forensic or substance focus. Some of this increased risk may be related to the primary disorder but there also may be a correlation to those having compulsory care. Those patients ordered to be in treatment by the court system may be less likely to engage fully or establish a more positive rapport. Studies4 suggest some patients with schizophrenia are also more prone to violent, impulsive acts. There are predictive factors in positive versus negative symptoms (Table).


Patients seeking abusable drugs can become aggressive. Behavioral escalation by a patient may occur when physicians deny such requests. Some thought should be given to how that is done and it may be necessary for the provider to excuse him or herself and return with help in order to do so safely. This is certainly preferable to filling a prescription under duress. In some cases the situation can be prevented through appropriate screening prior to the patient being seen in the clinic. The front desk should be given a screening questionnaire that asks directly why the patient needs to be seen and possibly asks for a current medicine list. This can aid in the selection of patients that would be suitable for the practice and might prevent the scenario of the drug-seeking patient.

A subset of patients may be prone to stalking behaviors. Psychiatrists and other mental health professionals are well-known targets for stalking. A survey5 of psychiatrists in a large mental health organization found that a majority of stalkers were male and had a diagnosis of personality disorder or other major mental illness. Retrospective studies with perpetrator populations help identify characteristics of those more likely to have stalked. One study6 from a high-security hospital in the United Kingdom found those classified as stalkers were more likely young, single, males with minimal education; most had diagnoses of psychosis and/or personality disorder. A study7 of individuals who had been convicted of stalking-related offenses found a link of “psychopathic traits” with stalking behaviors. Some have worked to understand which stalkers are more likely to be persistent in their behavior. A study8 of 200 stalkers found that those who tended to be persistent were >30 years of age, having a resentful or intimacy-seeking motivation, or were psychotic. It is essential to make sure appropriate boundaries are in place and are kept throughout treatment. Since a large group of stalkers tend to be of the intimacy-seeking type, reflecting on transference issues can help in prevention.

Stalking is not the only unwanted behavior that can occur secondary to transference. This is easily understood in the first few sessions with a patient, but issues of transference and counter-transference can occur at any point in the relationship. The unwanted behavior by the patient may even be a reaction to counter-transference by the physician. The physician-patient dynamic needs to be considered during the entire length of treatment.

THE CHALLENGES: staff and setting

Challenges can be divided into two categories: those of the practitioner and those directly related to the setting. Regarding the practitioner, victims of violence have tended to be those with the most physical contact with patients, primarily nurses.9,10 In fact, there are so many acts toward inpatient public nurses that assaults are considered one of their most frequent workplace-related injuries.11 Rate of violence in outpatient settings is not followed so it is unclear which staff members would be most at risk. In typical private clinics, a psychiatrist may be completely alone or have only minimal staff. Larger clinics may more closely resemble hospitals in terms of staffing, with multiple doctors, nurses, technicians, and office staff. The risk of violence is thought to be proportional to the amount of direct patient care.

An important consideration is the balance of a patient’s privacy and clinician safety. If seeing a potentially dangerous patient alone, the provider may alert a colleague to the situation. Some patients should be scheduled only when other staff members are present in the clinic. Even a few patients should not be seen alone in any setting and may require another person be present or nearby. When the potential for violence is clear, it may be best for the patient to be directly referred to a hospital setting with appropriate transport.

The literature9,12,13 supports training in de-escalation and violence prevention. Many practitioners who have experienced verbal or physical assaults have felt that they were not adequately trained for the situation.14 The belief is that with adequate training, clinicians can recognize situations before they escalate in order to respond earlier, prior to a violent event. The goal is prevention. In outpatient settings, training itself can be an obstacle. Training may be less accessible for outpatient doctors in a small clinic than those associated with a larger organization or university.

The clinic setting can lead to its own challenges. Simple decisions such as placement of furniture can have a profound effect on safety. Open floor plans where both clinician and patient have the ability to exit quickly can be reassuring for the clinician as well as create what a patient might perceive as a safer, less closed-in environment. Sitting behind a desk or having furniture in the room such as coffee tables might cause an obstruction, making it more difficult to leave the room. When a clinician has foresight regarding a particular patient being at higher risk for violence, furniture can be adjusted prior to the patient presentation. Unfortunately, one tends not to have this information prior to the violent act. When there are hints of potential violence, it may be useful to discuss the situation with a colleague or supervisor. Fear tends to constrain the thought process while informal or formal consultation may lead to more creative solutions.

Unlike the hospital or emergency room, it would not be cost effective (cost of equipment and manning compared to yield) and possibly likely to undermine the therapeutic relationship to have metal detectors outside of clinic doors. A number of reports15,16,17 based on emergency department data suggest that it is helpful and does not affect the doctor-patient relationship, but it could be reasoned that patient expectations are different in the two settings. What makes sense for an emergency department (metal detectors) might be considered very unreasonable in outpatient settings, signifying a lack of trust by the physician. Another major issue to having metal detectors is who mans the metal detector and takes away the contraband. If a member of the staff, such as a receptionist, is given this responsibility, the staff member would obviously be placed in harm’s way. Alternatively, hiring security may be cost prohibitive. At the other extreme, a survey18 regarding assaults in emergency medicine showed that 38% of emergency department physicians, randomly selected from the Michigan College of Emergency Medicine, carry handguns or knives for protection. This, of course, creates other hazards. Some settings should have a weapons policy and patients should be asked if they carry a weapon. Also, most hospitals have patrolling security guards but this would not be feasible for small clinics or even large clinics in stand-alone buildings.

Many clinics have alarm systems and “panic” buttons. Having this available may reassure the clinician, but this reassurance might be to his or her detriment. Alarm systems need to be tested regularly with drills to ensure the alarms are functioning and respondents are prepared. An alarm with no one on the other end may inspire false confidence.

Another challenge might be related to lack of staff when support is needed. An example of this is when dealing with a suicidal patient. The situation has to be contained while emergency medical service/police need to be contacted. This is not usually a problem when the patient is willing to go to the hospital, but may prove difficult when the patient is adamant about not seeking further help. The patient may escalate and try to complete the attempt in the office, elope, or even become dangerous to the physician. Since each of these situations can create a highly charged environment (and increased emotion for the clinician), it is important to have a protocol in place. For the highly impulsive or suicidal patient, access to a medicine cabinet could be dangerous. Locking the cabinet is not enough; the key needs to be in a secure place.


The response to an emergency should be considered and a formal plan made prior to an actual emergency. It is important for the entire staff to be educated to clinic policies and taught individual roles. These policies should be reviewed regularly. Key factors in any plan should address both patient and clinician/staff safety. If the patient’s emotional state is escalating, it may be best to let the person go rather than create an intolerable holding environment. It may make the situation worse if the patient feels “cornered.” The police can be called if a patient who needs emergency treatment leaves. There should be a low threshold for calling 911 or the police. They can be helpful for safety checks if the patient has eloped. With this in mind, having up-to-date patient information, eg, phone numbers and address, is essential. Calling 911 is also the mechanism for aiding a patient who may need to go directly to the emergency department. In the unfortunate event of an assault or a credible threat, the police will need to be involved to press charges or generally for public safety. If the patient gives information to the clinician that involves a plan to injure or kill a third party, it is necessary to get as much information as possible about the other person in order to warn of the threat. This threat would also need to be reported to police. The Tarasoff “duty to warn” is not followed by every state, which brings up an ethical versus legal discussion that will not be addressed in this paper.

THE AFTERMATH: Debriefing and administrative review

Studies1,19,20 have shown that hospital nurses victimized by patient assault are vulnerable to acute stress reactions and even PTSD. It is believed that debriefing immediately after an incident can reduce the psychological effects.1,21 A survey of mental health staff working in the five West Midlands Trusts in the UK showed that a large proportion of nurses and psychiatrists reported no support after an incident, with most wanting support.10 Debriefing can be informal or structured.19 In a one-physician private practice, debriefing might be through a colleague. The reason that an administrative review should be done after an incident would be to help find where the breakdown in a protocol occurred and to help with modification of policies. Administrative review, while it may seem tedious, could potentially prevent another occurrence.


Patient presentations to outpatient and inpatient settings may be similar, but the settings are actually so unique that one must give thought to the challenges of safety in the outpatient environment and not rely solely on the literature around inpatient safety. Emergency events, including assaults, occur in both settings. Prevention is the key. Thus, thought should be given to creating a safety plan for the outpatient office, with staffing and the particular office setting in mind. In the event of an occurrence, debriefing can aid in recovery of the clinician and administrative review can be helpful for the prevention of further occurrences.   PP


1. Alexy EM, Hutchins JA. Workplace violence: a primer for critical care nurses. Crit Care Nurs Clint North Am. 2006;18(3):305-312.
2. Baker A. Queens Man is Arrested in Killing of Therapist. New York Times. Feb 17, 2008. Available at: wwwnytimes.com/2008/02/17/nyregion/17murder.html. Accessed August 4, 2009.
3. “Cops: Massachusetts Teen Killed His Therapist.” FOXNews.com Feb 07, 2008. Available at: www.foxnews.com/story/0,2933,329285,00.html. Accessed on August 4, 2009.
4. Swanson JW, Swartz MS, Van Dorn RA, et al. A national study of violent behavior in persons with schizophrenia. Arch Gen Psych. 2006;63(5):490-499.
5. McIvor RJ, Potter L, Davies L. Stalking behaviour by patients towards psychiatrists in a large mental health organization. Int J Soc Psychiatry. 2008;54(4):350-357.
6. Whyte S, Petch E, Penny C, Reiss D. Who stalks? A description of patients at a high security hospital with a history of stalking behavior. Crim Behav Ment Health. 2008;18(1):27-38.
7. Storey JE, Hart SD, Melroy JR, Reavis JA. Psychopathy and stalking. Law Hum Behav. 2009;33(3):237-246.
8. McEwan TE, Mullen PE, Mackenzie R. A study of predictors of persistence in stalking situations. Law Hum Behav. 2009; 33(2):149-158. 2008 Jul 15 (E-publication).
9. Whittington R, Shuttleworth S, Hill L. Violence to staff in a general hospital setting. J Adv Nurs. 1996;24(2):326-333.
10. Nolan P, Dallender J, Soares J, Thomsen S, Arnetz B. Violence in mental health care: the experiences of mental health nurses and psychiatrists. J Adv Nurs. 1999;30(4):934-941.
11. Love CC, Hunter ME. Violence in public sector psychiatric hospitals. Benchmarking nursing staff injury rates. J Psychosoc Nurs Ment Health Serv. 1996;34(5):30-34.
12. Fernandes CM, Bouthillette F, Raboud JM, et al. Violence in the emergency department: a survey of health care workers. CMAJ. 1999;161(10):1245-1248.
13. Brasic J Fogelman D. Clinician Safety. Psychiatr Clin North Am. 1999;22(4):923-940.
14. Schwartz, T, Park T. Assaults by patients on psychiatric residents: a survey and training recommendations. Psychiatr Serv. 1999;50(3):381-383.
15. Anderson AA, Ghali AY, Bansil RK. Weapon carrying among patients in a psychiatric emergency room. Hosp Community Psychiatry. 1989;40(8):845-847.
16. McKillop W. Weapons screening. Hospital Community Psychiatry. 1987;38(2):203.
17. McCulloch LE,  McNieal D, Binder RL, Hatcher C. Effects of a weapon screening procedure in a psychiatric emergency room. Hosp Community Psychiatry. 1986;37(8):837-838.
18. Kowalenko T, Walters BL, Khare RK, et al. Workplace violence: a survey of emergency physicians in the state of Michigan. Ann Emerg Med. 2005;46(2):142-147.
19. Whittington R, Wykes T. Staff strain and social support in a psychiatric hospital following assault by a patient. J Adv Nurs. 1992;17(4):480-486.
20. Richter D, Berger K. Post-traumatic stress disorder following patient assaults among staff members of mental health hospitals: a prospective longitudinal study. BMC Psychiatry. 2006;6:15.
21. Kaplan Z, Iancu I, Bodner E. A review of psychological debriefing after extreme stress. Psychiatr Serv. 2001;52(6):824-827.