Skip to main content

Wandering and Elopement in Nursing Homes


Lester PE, Garite A, Kohen I. Wandering and elopement in nursing homes. Annals of Long-Term Care: Clinical Care and Aging. 2012;20(3):32-36.


Paula E. Lester, MD, FACP; Adrianna Garite, DO; Izchak Kohen, MD

Dr. Lester is an assistant professor of clinical medicine, Stony Brook University Medical Center, and associate fellowship program director, Division of Geriatric Medicine, Winthrop University Hospital, Mineola, NY. Dr. Garite is a hospice medical director and faculty member in geriatric and palliative medicine, Division of Geriatric Medicine, Winthrop University Hospital, Mineola, NY. Dr. Kohen is an attending faculty geriatric psychiatrist, Zucker Hillside Hospital/NS-LIJ Health System, Glen Oaks, NY, and an assistant professor of psychiatry, Hofstra Medical School, Uniondale, NY.


It is estimated that up to 31% of nursing home residents and between 25% and 70% of community-dwelling older adults with dementia wander at least once.1 A 2006 study reported that one in five people with dementia wander.2 Estimates on the prevalence of wandering vary widely between studies due to a lack of a consistent definition of wandering and diverse clinical sampling.
Related Content
New Wearable Device Enables Instant Location of Wandering Residents
Engaging Nursing Home Residents in Meaningful Activities

Cognitively impaired individuals who wander are at high risk of injury.3 Elopement, in which the person wanders away from home or from a skilled nursing facility, is the most dangerous type of wandering and has resulted in death. We discuss the case of an elderly wheelchair-bound adult with Alzheimer’s dementia who was found outside of the nursing home where he resided. This case supports the need for facilities to regularly assess residents’ propensity to wander, even in individuals who may not appear capable of wandering or elopement, and taking proactive measures to protect these individuals. We also review the literature and discuss triggers for wandering and elopement, assessment tools that can be used to determine risk, and individualized and facility-wide measures that nursing homes can implement to reduce the incidence and risk of wandering and elopement.

Case Report

An 85-year-old man with moderate Alzheimer’s dementia, osteoarthritis, and hypertension had resided in a long-term care (LTC) facility for several years. His medications at the nursing home included metoprolol, aspirin, and acetaminophen. The resident was nonambulatory and used a motorized wheelchair, which had been set to a very slow speed for safety reasons.

One day, the resident was found by staff in his wheelchair on the sidewalk in front of the nursing home, unsupervised and without permission to leave the facility. The resident was unharmed and agreed to return to his unit. Although he was on nursing home property when found, he was outdoors, unconfined, and could have wandered across the street or beyond within minutes.

Through interviews with the staff and close observation of the resident, it was determined that he had maneuvered to the bank of elevators and waited in front of one, entering as soon as it could accommodate him. On several prior occasions, he had been found waiting in front of the elevators, but was always returned to his main unit; these were most likely averted elopement attempts. It is unclear how he managed to make it past the front desk, but because the facility often receives visitors in motorized wheelchairs, it is possible he was mistaken for a guest.

The resident had no prior known episodes of wandering out of the facility, had not had a recent medical event, and was not in a delirious state based on chart review. His most recent cognitive and functional assessment had taken place approximately 6 weeks before the incident, but revealed no significant changes that would have merited reassessing his use of a motorized wheelchair. It was concluded that the wandering behaviors may have been triggered by a recent change in roommates. Although the resident sustained no injuries, his case raises the particular concern for assessing patients who do not superficially appear capable of elopement.


Wandering is a common reason for placing community-dwelling individuals with dementia in an LTC facility.4 Wandering, as it is applied to individuals who are disoriented or cognitively impaired, is generally characterized by excessive ambulation that has a tendency to lead to safety concerns or nuisance issues.1 It is a purposeful behavior triggered by a desire to fulfill a need.3 Although the nature of the need varies, it may relate to physical discomfort, such as the urge to urinate, or to emotional discomfort, such as the need for more or less stimulation.

Regardless of its etiology, wandering may represent the greatest safety risk to older adults with dementia. It increases their risk of injury or death and is especially dangerous for those who succeed in leaving a safe environment. Wick and Zanni2 reported that nursing home residents who wander have double the risk of fracture compared with residents who do not wander. Lesser adverse outcomes of wandering include undocumented falls, weight loss, abuse by other residents, and social isolation,5 all of which are serious concerns. Despite its dangers, patients who wander sometimes benefit when the activity serves as exercise, promoting circulation and oxygenation and decreasing contractures.1

Risk Factors for Wandering 

In addition to cognitive impairment, patient risk factors for wandering include older age, male sex, poor sleep patterns, agitation, aggression, and a more socially active and outgoing premorbid lifestyle.1 As shown by our case report, even nonambulatory patients who are confined to a wheelchair can wander, demonstrating that a resident’s propensity to wander should not be underestimated.

Unmet needs and environmental factors can also contribute to wandering risk (Table). It is well known that people with dementia have a lower threshold for stress, and their ability to cope with internal and environmental stress continues to erode as the disease progresses. Behavioral symptoms like wandering can emerge when internal discomfort, especially when coupled with external demands (eg, a noisy environment), exceed the individual’s threshold. Dementia patients with unmet physical or psychosocial needs, such as the need for toileting assistance or the need to find a place of safety or someone familiar, may be more prone to wandering.

table 1

Other theories behind wandering include the environmental docility hypothesis, which proposes that as competency decreases, the influence a person’s external environment has on his or her behavior increases.6 Thus, when people with physical or cognitive impairment are unable to meet their needs by adapting to their external environment, they become more sensitive to their environment. A biomedical hypothesis involves right parietal dysfunction, positing that a functionally impaired neural circuit leads to wandering.1 Wandering might also result from dysfunction in spatial perception and memory1; for example, a patient may wander because he or she cannot recall where the bathroom is.

Wandering Classifications

There are numerous classifications of wandering, including (but not limited to) elopement, environmentally cued wandering, reminiscent/fantasy wandering, tactile wandering, recreational wandering, and agitated purposeful wandering. Elopement, in which the confused person leaves an area and does not return, is considered the most dangerous type of wandering. People who elope are typically distinguished from those who merely wander by their purposeful, overt, and often repeated attempts to leave the premises. According to Wick and Zanni,2 residents are least likely to elope between 12 am and 7 am, and the majority of those who elope are repeat offenders, with approximately 72% of successful elopers attempting to do so again. A review of elopement claims against nursing homes found that 80% involved residents described as “chronic” wanderers; of note, 45% of incidents for which claims were filed occurred within 48 hours of admission.7                 

Patients who engage in environmentally cued wandering respond to signals received from their milieu. For example, seeing a hallway cues the patient to walk, whereas a chair cues the patient to sit. Patients who engage in tactile wandering use their hands to explore the environment as they travel. When someone attempts to reach a familiar place from the past, it is called “reminiscent wandering” or “fantasy wandering.” Recreational wandering is linked to a need for exercise and activity. Agitated purposeful wandering occurs in confused, frightened people who cannot be reasoned with and may become combative.8

Assessing Residents for Wandering Risk

Because wandering has been defined in numerous ways, it can be challenging to assess a patient’s risk; however, several tools are available to determine an individual’s propensity to wander. One such useful tool is the Cohen-Mansfield Agitation Inventory (CMAI), a seven-point rating scale used to assess the frequency with which the patient displays any of 29 behaviors associated with agitation, a risk factor for wandering. Another is the Rating Scale for Aggressive Behavior in the Elderly (RAGE), which has the observer rate the frequency with which the patient exhibits any of 21 aggressive behaviors during a 3-day period. There is also the Neuropsychiatric Inventory (NPI), a questionnaire typically administered by a clinician and used to score the frequency, severity, and distress associated with 10 to 12 behavioral domains common in dementia. Finally, the Revised Algase Wandering Scale (RAWS) can be used. RAWS was designed specifically to assess an individual’s risk of wandering and comes in two forms: the community version (RAWS-CV) and the LTC version (RAWS-LTC).9 RAWS-LTC has caregivers use a 1 to 4 scale (1=not a wanderer; 4=problem wanderer) to score patients on 19 domains allocated into three subscales: persistent walking, spatial disorientation, and eloping. The RAWS-CV includes 37 domains that fall within five subscales, which comprise the three used in RAWS-LTC plus routinized walking and negative outcomes. The rating scale is 1 to 5, with 1 indicating “never or unable” and 5 indicating “always.”

Minimizing Wandering and Elopement Risk

LTC facilities can decrease the risk of wandering and ensure the safety of residents who wander by instituting policies that require assessing residents on admission and reevaluating their behaviors frequently to identify potential wanderers. Federal regulations require that nursing homes that participate in Medicare or Medicaid conduct a comprehensive, accurate assessment of each resident’s needs “no later than 14 days after the admission”10 and “at least every 3 months thereafter,”11 unless there is “a significant change in the resident’s physical or mental condition,”12 in which case reassessment is needed immediately. Performing an assessment as soon as possible after admission is ideal because studies have shown most elopement occurs within 48 hours of admission.7 The assessment of elopement risk should be recorded in a readily identifiable section of residents’ charts, and the facility should institute a reliable mechanism for notifying all staff of high-risk residents.

It is also important to provide staff with proper training and support, enabling them to intervene appropriately to minimize wandering risk. Staff members not directly involved in resident care, new employees, per diem staff, and all visitors should be informed of the facility’s policies and procedures for preventing wandering. For example, uninformed visitors could inadvertently compromise the safety measures in place by allowing dementia patients to pass through routinely locked doors.

When developing policies and procedures to prevent wandering, a challenge for facilities is balancing resident safety while maximizing residents’ personal freedom per federal regulations, which state “services are to be furnished to attain or maintain the resident’s highest practicable physical, mental, and psychosocial well-being.”13 However, there are many individualized and facility-wide interventions that can be adopted to reduce wandering risk that do not severely compromise residents’ autonomy.

Individualized Approach. Taking an individualized approach to preventing wandering behaviors requires having staff closely attend to the at-risk resident’s specific physical, social, and emotional needs. They should be taught to be especially responsive to residents with dementia who exhibit behaviors such as excessive walking, show signs of anxiety or agitation, or verbalize a “need to get home.” Because individuals with dementia are less able to regulate their environment, it is important to attune the immediate environment to their needs as much as possible.14 Staff members should learn to recognize and promptly resolve the resident’s hunger, thirst, and toileting urges and seek to optimize the surrounding climate and noise levels for residents likely to wander.

Mealtime interventions, such as encouraging the at-risk resident to sit at the table with other residents rather than eat alone in his or her room and providing greater staff supervision, can make it harder for the resident to wander. These changes might also result in better nutritional intake, reducing subsequent hunger, which is an established risk factor for wandering behavior.

Facilitating social contact between extroverted residents with dementia and concerned, attentive staff members, friends, family, or even fellow residents can help keep these patients engaged and less inclined to wander. For residents capable of movement and exercise, physical activities help them stay busy, relaxed, oriented to time and place, and interested in their surroundings; it is important to give these residents opportunities to be active, regardless of their cognitive limitations. Increasing at-risk residents’ level of participation in interactive activities designed to enrich the social environment has also been shown to prevent boredom and lessen wandering behavior.15

Antipsychotic medications and physical restraints are not appropriate interventions for wandering behaviors. They increase the risk of pressure ulcers, infections, falls, and sedation, and can promote anxiety, agitation, or violence.16-18 

Facility-Wide Interventions. Carefully designed environmental interventions can enhance the safety of residents with dementia and improve their functionality. One obvious option for preventing the elopement of cognitively impaired residents who are ambulatory is housing them in locked dementia units. Although this measure might keep residents safer, it deprives them of freedom and is not a guarantee against elopement. Other facility-wide interventions might include establishing controlled indoor and outdoor areas where residents can wander freely, such as outdoor wander gardens; however, to promote better supervision, access to any outdoor area should be located in a central area of the living space. Having fewer residents per living area has also been found to improve residents’ orientation, decreasing wandering behavior by promoting their ability to navigate around the facility. Signs should be posted to remind visitors not to assist residents in leaving the facility.

Measures that steer wanderers toward safe environments include providing walking companions, maintaining clutter-free interior pathways, and creating enclosed exterior walkways.14,15,19 Clearly visible markings at the end of corridors can serve as meaningful cues on orientation for confused residents. A German study found that having corridors with clearly visible endings facilitated good orientation and suggested that spacious places such as the dining room be located in a distinctly memorable area.14 Many facilities have sought to reduce wandering behaviors by adopting visual barriers, such as stop signs, mirrors, camouflaged doors, and grids of tape, but a recent Cochrane review found no evidence that subjective barriers decrease wandering.19,20

Some LTC facilities have experimented with instituting physical barriers to wandering. Door locks, for example, can be used to restrict patients to a safe area. Elopement warning systems, such as alarms that sound when residents leave the bed, chair, or room, or when a wheelchair is set in motion, are widely used in nursing homes. The sound from these electronic-alert devices can cue the resident to his or her behavior and interrupt an attempt to wander, but in some residents the noise from the device might cause stress, thereby increasing the likelihood of wandering. Therefore, a better option may be use of transmitting devices that are designed to notify personnel when chair, bed, and door alarms are activated, allowing staff to intervene and avert wandering behavior without subjecting residents to unpleasant alarms.

Residents can also be equipped with transmitting wristband devices that alert staff members when the resident nears an alarm point or that use electromagnetism to lock doors automatically as the resident approaches; however, any automatic door locking system should be configured to deactivate during a fire emergency. Fitting cognitively impaired residents with electronic alert bands does invite concerns about privacy and dignity. Although cognitively impaired residents lack the capacity to consent to the devices, they are generally perceived as a necessary safety precaution. By routinely screening all residents for wandering risk, facilities ensure that they only use electronic alert systems to restrict at-risk individuals, enabling those residents allowed to leave the premises to do so without unnecessary supervision.

Many aspects of behavioral disturbance have visual or auditory components, and a team of Pennsylvania investigators developed an algorithm for detecting elopements.21 The investigators set up a network of video surveillance cameras in the public areas of a nursing home dementia unit and collected footage of residents’ daily activities. They programmed a computer to scan the footage and differentiate between ordinary behaviors and behaviors associated with elopement. The intelligent video monitoring system demonstrated nearly 100% accuracy at detecting escape attempts, leading to a significant reduction in elopements; however, it did produce a high rate of false alarms. The researchers concluded more information was needed to determine whether the rate could be reduced and to assess the logistical and financial costs of using intelligent video monitoring in LTC facilities. Information from a system like this could guide geriatricians in implementing interventions to prevent wandering and elopement.

Facilities’ Response to Elopement

Elopement is a serious concern for nursing homes,22 and facilities should establish a plan to manage the crisis of resident elopement. This plan should include a formal search procedure that clearly defines staff roles and responsibilities. Photographs and other identifying information should be on file for all residents so that this information is readily available for searchers and police in the event of an elopement.

Urgency should be heightened for residents who wander outdoors due to the significantly increased dangers associated with this. In a retrospective analysis of elopement incidents in Virginia involving community-dwelling adults and LTC residents with dementia, Koester23 said, “There is a 25% fatality rate if the subject is not found within the first 24 hours.” The likelihood of survival decreases as the amount of time the patient is missing increases. Only 60% of patients are found alive 72 or more hours after last being seen, which drops to 46% for those missing 96 hours or longer.24 Most deaths are attributable to hyperthermia, dehydration, and drowning; very few involve trauma.24

Koester24 reported that 89% of missing persons with dementia were found within 1 mile of their residence, with the distance traveled averaging 0.5 miles. They tended to travel along a path of least resistance (ie, downhill). Searchers should not expect missing patients to assist in their own rescue. Koester24 found that only 1% of patients responded to shouts or calls from rescuers and fewer than 1% of them called out to rescuers.                 

Increasingly, residents and their families are suing LTC facilities for elopement incidents in which a resident allegedly suffered injury or death due to the facility’s negligent care. Nearly 70% of elopement claims filed against nursing homes involve a resident’s death,7 underscoring the urgency of finding missing residents quickly. Residents who are injured while wandering have a fair degree of success when claims are filed that they were injured because the nursing home failed to provide the proper standard of care.25


Wandering and elopement frequently occur in nursing homes and pose serious risks for residents with cognitive impairment and for the facility, which can be held liable if a resident leaves without permission and is injured. In implementing measures to deter wandering and elopement, nursing homes must balance safety with autonomy. Although instruments are available to help assess wandering risk, it is important that facilities never underestimate their residents’ propensity to wander or elope. Physical handicaps and a lack of prior wandering or elopement attempts are no guarantee against elopement. As our case illustrates, even nonambulatory residents with cognitive impairment may be at high risk of wandering, especially if they are able to self-propel in wheelchairs.

The authors report no relevant financial relationships.



1. Lai CK, Arthur DG. Wandering behaviour in people with dementia. J Adv Nurs. 2003;44(2):173-182.

2. Wick JY, Zanni GR. Aimless excursions: wandering in the elderly. Consult Pharm. 2006;21(8):608-612, 615-618.

3. Aud MA. Dangerous wandering: elopements of older adults with dementia from long-term care facilities. Am J Alzheimers Dis Other Demen. 2004;19(6):361-368.

4. Logsdon RG, Teri L, McCurry SM, Gibbons LE, Kukull WA, Larson EB. Wandering: a significant problem among community-residing individuals with Alzheimer’s disease. J Gerontol B Psychol Sci Soc Sci. 1998;53(5):P294-P299.

5. Beattie ER, Song J, LaGore S. A comparison of wandering behavior in nursing homes and assisted living facilities. Res Theory Nurs Pract. 2005;19(2):181-196.

6. Lawton M. Environment and Aging. 2nd ed. Albany, NY: Center for the Study of Aging; 1986.

7. Rodriguez J. Resident falls and elopement: costs and controls. Nurs Homes. 1993;
42(4):16-17. Accessed February 7, 2012.

8. Hall GR, Buckwalter KC, Stolley JM, et al. Standardized care plan. Managing
Alzheimer’s patients at home. J Gerontol Nurs. 1995;21(1):37-47.

9. Nelson A, Algase DL, eds. Evidence-Based Practice for Managing Wandering Behaviors. New York, NY: Springer Publishing Company; 2007.

10. Code of Federal Regulations. Title 42. 483.20(b)(4)(i).

11. Code of Federal Regulations. Title 42. 483.20(b)(5).

12. Code of Federal Regulations. Title 42. 483.20(b)(4)(iv).

13. Code of Federal Regulations. Title 42. 483.20(k)(i).

14. Marquardt G, Schmieg P. Dementia-friendly architecture: environments that facilitate wayfinding in nursing homes [in German]. Z Gerontol Geriatr. 2009;24(4):333-340.

15. Moore DH, Algase DL, Powell-Cope G, et al. A framework for managing wandering and preventing elopement. Am J Alzheimers Dis Other Demen. 2009;24(3):208-219.

16. Castle NG, Engberg J. The health consequences of using physical restraints in nursing homes. Med Care. 2009;47(11):1164-1173.

17. Ray WA, Griffin MR, Schaffner W, Baugh DK, Melton LJ 3rd. Psychotropic drug use and the risk of hip fracture. N Engl J Med. 1987;316(7):363-369.

18. Devanand DP, Schultz SK. Consequences of antipsychotic medications for the
dementia patient. Am J Psychiatry. 2011;168(8):767-769.

19. Hermans DG, Htay UH, McShane R. Non-pharmacological interventions for wandering of people with dementia in the domestic setting. Cochrane Database Syst Rev. 2007;(1):CD005994.

20. Price JD, Hermans DG, Grimley Evans J. Subjective barriers to prevent wandering of cognitively impaired people. Cochrane Database Syst Rev. 2000;(4):CD001932.

21. Chen D, Bharucha AJ, Wactlar HD. Intelligent video monitoring to improve safety of older persons. Conf Proc IEEE Eng Med Biol Soc. 2007:3814-3817.

22. Chung JC, Lai CK. Elopement among community-dwelling older adults with dementia. Int Psychogeriatr. 2011;23(1):65-72.

23. Koester RJ, Stooksbury DE. The lost Alzheimer’s and related disorders search subject: new research and perspectives. In: Response 98 NASAR Proceedings. Chantilly, VA: National Association of Search and Rescue; 1998:165-181. Accessed February 7, 2012.

24. Koester RJ. Lost Person Behavior: A Search and Rescue Guide on Where to Look for Land, Air, and Water. Charlottesville, VA: dbS Productions LLC; 2008:165-169. Accessed February 7, 2012.

25. Frolik LA. Nursing home liability because of resident wandering and elopement. Health Care Law Mon. 2000:15-18.

Back to Top