Patient Safety Tip of the Week

April 2, 2013

Absconding from Behavioral Health Services



We’ve participated in several root cause analyses (RCA’s) pertaining to patient absconds or elopements from behavioral health units. One question we always ask is “Is there a validated abscond risk assessment tool?” While behavioral health has risk assessment tools for suicide, falls, substance abuse and others, we have yet to find a risk assessment tool for absconding.


While most patients who abscond from behavioral health units return without any adverse consequences, some may commit suicide, others may commit violent acts, and others may be the victims of violent acts. In addition, absconding may lead to delays in treatment and even have economic consequences (Muir-Cochrane 2008). Patients may be vulnerable if they are drowsy from medications or may miss medication doses and there are also emotional responses felt by staff when patients abscond (Muir-Cochrane 2012a). Patient absconds may also reflect poorly on the hospital’s reputation in the community. So it is important to understand who is at high risk for absconding (note that we will use the terms abscond and elopement interchangeably in this column) and what circumstances may lead to absconding so that we might takes steps to minimize the risks.


A new review (Brumbles 2013) examines the reasons for patient elopement, how they do it, and what might be done to prevent it. The authors rely heavily on previous work done by Dr. Len Bowers and colleagues a decade ago (Bowers 1999a, Bowers 1999b, Bowers 1999c) and a previous excellent review done by Muir-Cochrane and Mosel (Muir-Cochrane 2008). They note that patient level factors are important but that environmental and situational factors are equally important. The review provides a table of reasons patients give for eloping.


Early studies indicated that those who abscond are often young, male, single and from a disadvantaged group. But the environment is important, too. Some patients are simply bored in the hospital and don’t have enough activities. Others feel confined. But a big factor is how the patient perceives safety of the unit. Patients may feel intimidated by other patients or even by staff. One study (Muir-Cochrane 2012b) found that the perception of the psychiatric unit as an unsafe place by patients increased the probability of absconding. The perception of safety was a complex interaction of physical, social, individual and symbolic aspects of the unit. Familiarity with the unit, comfortable environment, formed therapeutic relationship with staff, and positive experiences with other patients helped develop a perception that the unit was safe, reducing risk of absconding.


Other patients feel they need to look after their belongings at home or have responsibilities to family or others at home and abscond to attend to these.


Patients with dual diagnoses may have substance abuse desires, physiological (withdrawal) or psychological and may abscond because they feel these needs are not being attended to.


One potential risk factor we found to be glaringly absent in the literature is smoking. Due to Joint Commission and other regulatory reasons (as well as health reasons) hospitals today are tobacco-free. That includes the behavioral health units in most hospitals. In our December 2012 What’s New in the Patient Safety World column “Just Went to Have a Smoke” we noted a study (Regan 2012) showing that 18.4% of general hospital patients who smoke will smoke at some time during their inpatient hospitalization. Rates of smoking are 70% higher among people with mental illness, with 36 percent of adults with a mental illness being cigarette smokers, compared with only 21 percent of adults who do not have a mental illness (CDC 2013). So it shouldn’t be surprising to see behavioral health patients leave the unit or the hospital to smoke. Hospitals usually do a good job of identifying smokers on admission and offering them nicotine replacement products and counseling on smoking cessation. But it often stops there. Just assessing tobacco cessation issues and offering nicotine replacement therapy and counseling on admission is not enough. It is really incumbent upon hospitals to incorporate continued assessment, perhaps even daily, of tobacco issues into their care plans.


Another review of absconding from a psychiatric hospital in Australia (Mosel 2010) noted an absconding rate of 13.3%. Though males were more likely to abscond the gender difference between absconders did not reach statistical significance. Most absconders were in the age range 20-29 and schizophrenia was the predominant diagnosis. Absconding around the time bad news is to be delivered, such as extension of an involuntary admission, was also noted frequently. About half the absconders had also absconded previously. Of particular interest was that the time of day of absconding had two peaks, between 1900 and 2059 and between 1500 and 1559. These time periods corresponded to nursing handovers or nursing breaks at this hospital. The earlier study by Bowers et al. (Bowers 1999b) had also shown that nursing shift handovers were a favorite time for patients to abscond.


Though some patients abscond when on leave with permission or during planned (supervised) outings, most simply leave via doors on the units. Some leave through doors unintentionally left unlocked. Some use unattended or stolen keys to open doors. Others leave via windows. In an interesting article outlining the pros and cons of locking doors on psychiatric units, Muir-Cochrane and colleagues (Muir-Cochrane 2012a) many patients and visitors noted they felt that if a patient really wanted to get out of a locked ward they could.


An intervention to prevent elopements resulted in a 25% decrease in elopements (Bowers 2003, Bowers 2005). Key components were a host of staff educational materials, posters, and laminated cards with risk factors. These stressed 6 key elements: rule clarity with a sign out/sign in book, identification of those at high risk for absconding, targeted nursing time for those at high risk, careful breaking of bad news, post-incident debriefing, and multi-disciplinary review after 2 elopements. The 25% reduction in absconds was achieved both in the original 5-unit pilot and the larger 15-unit multiple hospital study.


What do you do after an abscond or elopement takes place? As noted in the intervention above, debriefing and review of each case is important. The Brumbles review provides a table with common questions to ask post-event. They are obviously aimed at identifying factors that may have led to the abscond and identify unfulfilled patient needs or interventions that might be important in preventing another abscond. To that we’d add that the debriefing should also address concerns and anxieties that the staff (i.e. the “second victim”) may have.


The literature has wide variation in reporting adverse outcomes of absconds or elopements. The Brumbles review notes rates of suicide as high as 20-30%. However, Bowers et al. (Bowers 1999c) note that many facilities do not report all absconds so that statistics may be biased. In their own study they found 2.4% of patients harmed themselves during absconds, 1.6% harmed others, and the other 96% had benign outcomes. The vast majority actually went home and did typical day-to-day activities. Nevertheless, the potential for suicide or other harm or harm to others is there during absconds and it is important to prevent them. The Bowers group also points out that even when patients return unharmed a considerable amount of staff anxiety has occurred plus a large amount of work on the part of staff and police in many cases. And confidence in the hospital may decline amongst patients, families, and the community.


We hope that some of the learnings from these studies may prove useful to those hospitals having behavioral health units. All would agree that simply locking the doors or using other physical constraints is not enough. It is important to understand why patients abscond and make the behavioral health environment feel safe to the patient. The human interactions are far more important than the physical ones.







Muir-Cochrane E, Mosel KA. Absconding: A review of the literature 1996-2008. Int J Ment Health Nurs 2008; 17(5): 370-378



Muir-Cochrane E, van der Merwe M, Nijman H, Haglund K, Simpson A, Bowers L. Investigation into the acceptability of door locking to staff, patients, and visitors on acute psychiatric wards. Int J Ment Health Nurs 2012; 21(1): 41–49



Brumbles D, Meister A. Psychiatric Elopement: Using Evidence to Examine Causative Factors and Preventive Measures. Archives of Psychiatric Nursing 2013; 27(1) 3-9



Bowers L, Jarrett M, Clark N, et al. Absconding: why patients leave. J Psychiatr Ment Health Nurs 1999; 6(3): 199-205



Bowers L, Jarrett M, Clark N, et al. Absconding: how and when patients leave the ward. J Psychiatr Ment Health Nurs 1999; 6(3): 207-211



Bowers L, Jarrett M, Clark N, et al. Absconding: outcome and risk. J Psychiatr Ment Health Nurs 1999; 6(3): 213-218



Muir-Cochrane E, van der Merwe M, Nijman H, Haglund K, Simpson A, Bowers L. Investigation into the acceptability of door locking to staff, patients, and visitors on acute psychiatric wards. Int J Ment Health Nurs 2012; 21(1): 41–49



Muir-Cochrane E, Oster C, Grotto J, Gerace A, Jones J. The inpatient psychiatric unit as both a safe and unsafe place: Implications for absconding. Int J Ment Health Nurs 2012; Article first published online : 25 SEP 2012



Regan S, Viana JC, Reyen M, Rigotti NA. Prevalence and Predictors of Smoking by Inpatients During a Hospital Stay. Arch Intern Med 2012; ():1-5, Published online ahead of print November 5, 2012



CDC. Vital Signs: Current Cigarette Smoking Among Adults Aged ≥18 Years with Mental Illness — United States, 2009–2011. MMWR 2013; 61: 1-7 early release February 5, 2013



Mosel KA, Gerace A, Muir-Cochrane E. Retrospective analysis of absconding behaviour by acute care consumers in one psychiatric hospital campus in Australia

Int J Ment Health Nurs 2010; 19(3): 177–185



Bowers L, Alexander J, Gaskell C. A trail of anti-absconding intervention in acute psychiatry wards. J Psych Ment Health Nurs 2003; 10: 410-416



Bowers L, Simpson A, Alexander J. Real world application of an intervention to reduce absconding. J Psych Ment Health Nurs 2005; 12(5): 598-602







Print PDF version












Tip of the Week Archive


What’s New in the Patient Safety World Archive