Effect of Dementia and Treatment of Dementia on Time to Discharge from Assisted Living Facilities: The Maryland Assisted Living Study Constantine G. Lyketsos, MD, MHS, à wz Quincy M. Samus, MS, w Alva Baker, MD, wz Mathew McNabney, MD, z Chiadi U. Onyike, MD, MHS, z w Lawrence S. Mayer, MD, PhD, w Jason Brandt, PhD, $z Peter Rabins, MD, MPH, wz and Adam Rosenblatt, MD wz OBJECTIVES: To estimate the association between dementia and time to discharge from individual assisted living (AL) facilities and examine, in residents with demen- tia, factors associated with shorter duration of residence in individual AL facilities. DESIGN: Prospective cohort study. SETTING: Twenty-two AL facilities in central Maryland. PARTICIPANTS: Stratified random sample of 198 AL res- idents followed for a median of 18 months. MEASUREMENTS: Detailed assessments to diagnose de- mentia; assess treatment of dementia; and rate clinical; cognitive, functional, and quality-of-life measures. RESULTS: Residents with dementia remained in a facility 209 fewer days at the median (P 5.001) than residents without dementia. After adjustment for other variables, lack of treatment for dementia (P 5.01) and more-serious medical comorbidity (P 5.02) were associated with earlier discharge in participants with dementia. Impaired mobility and limited activity participation had weaker associations with earlier time to discharge. CONCLUSION: Dementia may accelerate time to dis- charge, and its treatment may attenuate this effect. The hy- pothesis that the detection and treatment of dementia might delay discharge from AL should be tested in randomized trials. J Am Geriatr Soc 55:1031–1037, 2007. Key words: dementia; assisted living; length of residence; risk factors A ssisted living (AL) is a rapidly growing long-term res- idence option designed to facilitate ‘‘aging in place.’’ 1 Aging in place is a central, highly valued principle of the AL movement. 1 Residents in AL facilities receive more services than traditional community-based or retirement communi- ty–based homes but fewer services than are typically avail- able in a nursing home. 2 There may be as many as 20,000 AL residences in the United States currently housing more than a million Americans. 2 This number is expected to double or perhaps triple over the next few decades, making AL the most rapidly growing housing option for aging adults in the United States who are unable to live alone. Recent estimates 3 suggest that as many as two-thirds of adults residing in AL suffer from dementia, or three-quar- ters when one considers all forms of cognitive disorder (e.g., including mild cognitive impairment). In AL, dementia and its associated behavioral symptoms have been predictive of greater impairments in disability 4 and worse quality of life, 5,6 yet dementia goes undetected in as many as one quarter of residents living in AL, 7 and only half receive proper treatment for this condition. 3 Because transition from AL to a nursing home can be demoralizing and disruptive for an older person, much effort is placed on keeping residents in their facilities as long as possible. This commitment to help residents ‘‘age in place,’’ or live in the same facility for as long as possible, is a major tenet of the AL movement. Despite this, few empir- ical studies have investigated factors that influence the abil- ity of AL residents to age in place, as reflected in their length of residence in a given AL facility. Extrapolations from nursing homes may be useful in some situations, but are probably not applicable, because AL residents have less medical morbidity and are younger. 8 Address correspondence to Constantine G. Lyketsos, MD, MHS, The Elizabeth Plank Althouse Professor and Chair, Department of Psychiatry, Johns Hopkins Bayview, 4940 Eastern Avenue, A4 Center, Room 458, Baltimore, MD 21224. E-mail: kostas@jhmi.edu DOI: 10.1111/j.1532-5415.2007.01225.x From the à Department of Psychiatry Johns Hopkins Bayview, Johns Hopkins Medicine; w Division of Geriatric Psychiatry and Neuropsychiatry, Depart- ment of Psychiatry and Behavioral Sciences, School of Medicine, Johns Hopkins University; z Copper Ridge Institute, Sykesville, Maryland; $ Division of Medical Psychology, Department of Psychiatry and Behavioral Sciences, School of Medicine, Johns Hopkins University; z Division of Geriatric Medicine and Gerontology, Department of Medicine, School of Medicine, Johns Hopkins University. JAGS 55:1031–1037, 2007 r 2007, Copyright the Authors Journal compilation r 2007, The American Geriatrics Society 0002-8614/07/$15.00