Scholarly article on topic 'Elevated Hospitalization Risk of Assisted Living Residents With Dementia in Alberta, Canada'

Elevated Hospitalization Risk of Assisted Living Residents With Dementia in Alberta, Canada Academic research paper on "Clinical medicine"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{Dementia / "assisted living" / "long-term care" / hospitalization / predictors}

Abstract of research paper on Clinical medicine, author of scientific article — Colleen J. Maxwell, Joseph E. Amuah, David B. Hogan, Monica Cepoiu-Martin, Andrea Gruneir, et al.

Abstract Objectives Assisted living (AL) is an increasingly used residential option for older adults with dementia; however, lower staffing rates and service availability raise concerns that such residents may be at increased risk for adverse outcomes. Our objectives were to determine the incidence of hospitalization over 1 year for dementia residents of designated AL (DAL) facilities, compared with long-term care (LTC) facilities, and identify resident- and facility-level predictors of hospitalization among DAL residents. Methods Participants were 609 DAL (mean age 85.7 ± 6.6 years) and 691 LTC (86.4 ± 6.9 years) residents with dementia enrolled in the Alberta Continuing Care Epidemiological Studies. Research nurses completed a standardized comprehensive assessment of residents and interviewed family caregivers at baseline (2006–2008) and 1 year later. Standardized administrator interviews provided facility level data. Hospitalization was determined via linkage with the provincial Inpatient Discharge Abstract Database. Multivariable Cox proportional hazards models were used to identify predictors of hospitalization. Results The cumulative annual incidence of hospitalization was 38.6% (34.5%–42.7%) for DAL and 10.3% (8.0%–12.6%) for LTC residents with dementia. A significantly increased risk for hospitalization was observed for DAL residents aged 90+ years, with poor social relationships, less severe cognitive impairment, greater health instability, fatigue, high medication use (11+ medications), and 2+ hospitalizations in the preceding year. Residents from DAL facilities with a smaller number of spaces, no chain affiliation, and from specific health regions showed a higher risk of hospitalization. Conclusions DAL residents with dementia had a hospitalization rate almost 4-fold higher than LTC residents with dementia. Our findings raise questions about the ability of some AL facilities to adequately address the needs of cognitively impaired residents and highlight potential clinical, social, and policy areas for targeted interventions to reduce hospitalization risk.

Academic research paper on topic "Elevated Hospitalization Risk of Assisted Living Residents With Dementia in Alberta, Canada"

JAMDA

journal homepage: www.jamda.com

Original Study

Elevated Hospitalization Risk of Assisted Living Residents With Dementia in Alberta, Canada

Colleen J. Maxwell PhDa'b'c'*, Joseph E. Amuah PhDd, David B. Hogan MDb'e, Monica Cepoiu-Martin MD, MScb, Andrea Gruneir PhDf, Scott B. Patten PhD, MDb, Andrea Soo MScb, Kenneth Le Clair MDg, Kimberley Wilson MSWh, Brad Hagen PhDi, Laurel A. Strain PhDj

a Schools of Pharmacy and Public Health and Health Systems, University of Waterloo, Waterloo, Ontario, Canada b Department of Community Health Sciences, University of Calgary, Calgary, Alberta, Canada c Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada

d Health System Performance Branch, Canadian Institute for Health Information, Ottawa, Ontario, Canada e Division of Geriatric Medicine, Department of Medicine, University of Calgary, Calgary, Alberta, Canada f Department of Family Medicine, University of Alberta, Edmonton, Alberta, Canada

g Division of Geriatric Psychiatry, Queen's University and Center for Studies in Aging and Health, Providence Care, Kingston, Ontario, Canada h Department of Family Relations and Applied Nutrition, University of Guelph, Macdonald Institute, Guelph, Ontario, Canada i Faculty of Health Sciences, University of Lethbridge, Lethbridge, Alberta, Canada j Department of Sociology, University of Alberta, Edmonton, Alberta, Canada

CrossMark

ABSTRACT

Keywords: Dementia assisted living long-term care hospitalization predictors

Objectives: Assisted living (AL) is an increasingly used residential option for older adults with dementia; however, lower staffing rates and service availability raise concerns that such residents may be at increased risk for adverse outcomes. Our objectives were to determine the incidence of hospitalization over 1 year for dementia residents of designated AL (DAL) facilities, compared with long-term care (LTC) facilities, and identify resident- and facility-level predictors of hospitalization among DAL residents. Methods: Participants were 609 DAL (mean age 85.7 ± 6.6 years) and 691 LTC (86.4 ± 6.9 years) residents with dementia enrolled in the Alberta Continuing Care Epidemiological Studies. Research nurses completed a standardized comprehensive assessment of residents and interviewed family caregivers at baseline (2006—2008) and 1 year later. Standardized administrator interviews provided facility level data. Hospitalization was determined via linkage with the provincial Inpatient Discharge Abstract Database. Multivariable Cox proportional hazards models were used to identify predictors of hospitalization.

Results: The cumulative annual incidence of hospitalization was 38.6% (34.5%—42.7%) for DAL and 10.3% (8.0%—12.6%) for LTC residents with dementia. A significantly increased risk for hospitalization was observed for DAL residents aged 90+ years, with poor social relationships, less severe cognitive impairment, greater health instability, fatigue, high medication use (11+ medications), and 2+ hospitalizations in the preceding year. Residents from DAL facilities with a smaller number of spaces, no chain affiliation, and from specific health regions showed a higher risk of hospitalization. Conclusions: DAL residents with dementia had a hospitalization rate almost 4-fold higher than LTC residents with dementia. Our findings raise questions about the ability of some AL facilities to adequately address the needs of cognitively impaired residents and highlight potential clinical, social, and policy areas for targeted interventions to reduce hospitalization risk.

© 2015 AMDA — The Society for Post-Acute and Long-Term Care Medicine. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

This work was supported by the Alberta Heritage Foundation for Medical Research (grant number 200400893); Alberta Health Services and Alberta Human Services as part of the Collaborative Research Grant Initiative: Mental Wellness in Seniors and Persons with Disabilities; the Canadian Institutes of Health Research (CIHR) (grant number MOP81216); and, the CIHR-Institute of Aging Northern and Rural Health Research Initiative (grant number HAS-63179). The authors declare no conflicts of interest.

* Address correspondence to Colleen J. Maxwell, PhD, Schools of Pharmacy and Public Health and Health Systems, University of Waterloo, 200 University Avenue West, Waterloo, Ontario, Canada N2L 3G1.

E-mail address: colleen.maxwell@uwaterloo.ca (C.J. Maxwell).

http://dx.doi.org/10.1016/jjamda.2015.01.079

1525-8610/© 2015 AMDA — The Society for Post-Acute and Long-Term Care Medicine. This is an open access article under the CC BY-NC-ND license (http://creativecommons. org/licenses/by-nc-nd/4.0/).

The increasing number of older adults with dementing disorders1-4 presents significant challenges to providers and policy makers striving to ensure cost-effective, high quality care for aging populations. Similar to trends observed in the US,5-11 older adults with dementia, often with significant comorbidity, are increasingly being cared for in assisted living (AL) settings within Canada.12,13 Current estimates suggest that between 40% and 60% of AL residents across North America have a diagnosis of dementia.5,9,12 The rapid expansion of AL over recent years reflects both the lower costs of this residential option compared with long-term (ie, nursing home) care (LTC) as well as individual preferences for more homelike settings.14,15 Yet, the AL sector remains poorly defined and understood.12,16 In particular, uncertainty persists regarding how best to structure and implement integrated care to support personal preferences while limiting costly and potentially inappropriate transitions in care.17-19

AL facilities aim to provide secure housing, personal support and limited health care while promoting choice, autonomy, privacy, and independence.5,20 Although the specific role of AL varies considerably across regions, in many instances it is viewed as a substitute for long-term care.12,15 Yet, the care philosophy and approach adopted by the AL sector (emphasizing a social care model with support services provided in a home-like residential setting) differ substantially from the traditional LTC setting.12,15 Although it is acknowledged that the AL philosophy may promote functional independence and satisfaction among residents, many clinical and quality of care issues remain unanswered.5,17 The lower staffing levels with predominantly nonprofessional care providers and limited availability of ancillary services within AL (relative to LTC) has raised concerns that more vulnerable residents may be at an increased risk for poor quality of care and adverse health outcomes.9,17,21-23 Delayed detection of emerging health issues, suboptimal medication management, and limited ability to augment care in the AL setting could ultimately result in higher use of hospital care and costs.16,23-27 These concerns are heightened for older residents with dementia who often exhibit considerable comorbidity, atypical disease presentation, and difficulties in communicating their symptoms and needs.18,19,28

Relative to LTC,29-36 empirical data on the health and social needs and outcomes of older residents with dementia in AL remain scarce.7-9,37 It is unclear whether observed differences in resident, facility, and system characteristics contribute to elevated risks for potentially adverse outcomes.9,38-40 In the US Collaborative Studies for Long-Term Care,9 adjusted analyses revealed a significantly higher 1-year hospitalization rate for residents with dementia in AL compared with LTC. Interestingly, no other outcome differences were observed across these 2 settings.8,9,17 These findings illustrate the challenges faced by AL facilities in caring for cognitively vulnerable residents with substantial and often unstable medical and nursing needs.9,21,26,37

Hospitalization is an important outcome not only as a potential quality of care indicator for the AL sector but also in light of the health risks posed for older persons with dementia who experience an acute care admission. Community- and claims-based studies have consistently shown that relative to matched controls, those with dementia are significantly more likely to be hospitalized (often for potentially preventable reasons),18,28,41 to have longer lengths of stay,42,43 and higher health care costs.41,44 These vulnerable patients are also at greater risk for experiencing poorer health and functional outcomes during and posthospitalization.28,45-47 Given the increasing use of AL facilities and the relative absence of data on the extent and risk factors for hospitalization among older AL residents with dementia, further large-scale investigations are needed for more informed clinical and policy decisions.17

We sought to estimate the incidence of hospitalization among designated (publicly funded) assisted living (DAL) residents with

dementia in Alberta over 1 year, to compare this rate to that observed among LTC residents with dementia from the same catchment areas and time period, and to identify DAL resident and facility characteristics associated with an increased risk for hospitalization.

Methods

Study Design

Data were derived from the Alberta Continuing Care Epidemiological Studies (ACCES), a longitudinal investigation of AL and LTC residents in the Province of Alberta, Canada.20,48 The AL cohort included older residents of designated (publicly funded) AL and supportive housing facilities (DAL) in 5 health regions (2 major urban and 3 largely rural regions). At the time of the study, these regions accounted for over 80% of provincial continuing care beds. The DAL settings included in ACCES fell under the supportive living stream adopted by the province at this time and, thus, encompassed a philosophy and approach for providing support services within a secure housing and "home-like" environment. Although the settings shared similar features (including a focus on promoting resident independence, autonomy, privacy, and aging in place), there were some differences across the health regions in target populations, and in the type and availability of care staff and services.12,20 Further details regarding the residents and facilities included in ACCES have been

published elsewhere.12,20,48

To be eligible for inclusion, DAL facilities had to be in operation for 6+ months; not primarily serving residents with mental illness (other than dementia) or developmental disabilities; and, housing a minimum number of DAL residents 65+ years old (>4 for small and >10 for large facilities). Fifty-nine of the 60 DAL facilities meeting these criteria agreed to participate. Residents in participating DAL sites were excluded if they were aged less than 65 years, recently admitted (<21 days), receiving palliative care (expected survival <6 months), and/or their participation was otherwise deemed inappropriate by staff or family. A total of 1089 participants of the 1510 eligible residents (72.1% response rate) were enrolled and assessed [339 (22.5%) refused and for 82 (5.4%) their legally designated surrogate could not be contacted]. Age and sex were available for 364/421 (86.5%) of the nonparticipants and showed a similar distribution (mean age 84.4 ± 7.1, 74% women) to those enrolled. Of the 1066 residents with outcome measures (excluded were 3 residents with unknown outcomes and 20 who refused consent for administrative data linkage), there were 609 (57.1%) with a recorded diagnosis of dementia.

A random sample of small and large LTC facilities (based on median bed number) was selected in each region employing the same facility eligibility criteria. Within Alberta, LTC facilities provide support and 24-hour registered nursing care (including on an unscheduled basis) for persons with complex and chronic health needs. At the time of the study, these facilities did not provide short-stay or post-acute care or intravenous therapy. As such, residents are comparable to the long-stay nursing home population in the United States. Fifty-four of the 59 facilities approached agreed to participate. A random sample of 1731 eligible residents (same resident criteria) in the participating facilities was approached, and 1000 were enrolled and assessed (57.8% response rate). Age and sex were available for 665/731 (91%) of nonparticipants and showed a similar distribution (mean age 84.7 ± 7.5, 67% women) to participants. Of the 976 LTC participants with outcome measures (3 could not be linked with administrative data and 21 did not consent to data linkage), 691 (70.8%) had a recorded diagnosis of dementia.

Ethics approval was obtained from the University of Calgary Conjoint Health Research Ethics Board, the University of Alberta

Health Research Ethics Board and the University of Lethbridge Human Subject Research Committee.

Resident Level Characteristics

At baseline (2006-2008), trained research nurses administered the Resident Assessment Instrument for Assisted Living or LTC Facility (interRAI-AL or interRAI-LTC Facility) among DAL and LTC residents, respectively. These validated instruments provide a comprehensive, standardized assessment of residents' sociodemographic characteristics, physical and cognitive status, health conditions, behavioral problems, and use of medications and services.49,50

Resident characteristics examined included age, sex, marital status, length of stay, social engagement, cognitive and functional status, depressive symptoms, health stability, fatigue (defined as inability to complete normal daily activities in past 3 days), aggressive behaviors, number of chronic diseases and medications [including hyperpoly-pharmacy27 (11 + medications) and use of a cholinesterase inhibitor and/or memantine], falls, previous hospitalizations (past year), bladder and/or bowel incontinence, and presence of advanced directives. Residents' functional and health characteristics were captured by validated scales derived from assessment items on the interRAI tools, including the: Cognitive Performance Scale (CPS)51; Activities of Daily Living Self-Performance Hierarchy Scale52; Depression Rating Scale53; Changes in Health, End-Stage Disease and Symptoms and Signs Scale (for health instability)54; and, Aggressive Behavior Scale.55 Higher scores on all scales indicate more severe impairment. Comorbidity was measured by the sum of recorded diagnoses on the interRAI instruments. A total of 48 possible diagnoses (in addition to dementia) were considered in this comorbidity score. All diagnoses (including that of dementia, which was used to define our study cohorts), recorded on the interRAI assessment were considered present if indicated (in the chart) as the main reason for the current stay or if assessed by care providers as relevant to the resident's current functional and cognitive status, treatment and monitoring needs, or risk of decline in health status. Previous research has shown relatively high sensitivity estimates (of 0.80 or greater) for several recorded diagnoses on the RAI tools (including dementia) when compared with acute care discharge abstracts.56 Social engagement was assessed by 2 measures calculated from items on the instruments: (1) strength of social relationships; and, (2) average time involved in activities when awake and not receiving treatments or activities of daily living assistance.

Facility Level Characteristics

Facility administrators, managers or directors of care (ie, someone familiar with the facility with direct knowledge about the residents) were surveyed approximately midway during follow-up. Facility characteristics examined included: location (health region; community size), ownership (for-profit vs not; whether part of a multifacility system or chain of facilities), year spaces opened, availability of other levels of care on site including LTC and acute care, presence of dementia beds (may or may not include specialized care), type and size of facility (number of spaces and total facility spaces), and staffing levels and oversight [24 hour/7 day availability of licensed practical and/or registered nurses (LPNs/RNs) on site; physician involvement/ affiliation with site].

Outcomes

The primary outcome was time to first acute care hospitalization within a year of baseline assessment. This was determined via linkage with the Alberta Inpatient Discharge Abstract Database. The date of

admission, most responsible diagnosis (based on International Classification of Diseases, 10th Revision, Canadian codes),57 length of stay (LOS), and alternate level of care (ALC) bed-days (ie, occupying a hospital bed when not requiring the intensity of resources/services provided in this care setting) were examined. An ALC designation means that a medical decision has been made that the patient does not require further acute care services and should be discharged. A patient designated ALC and remaining in hospital is an individual that cannot be discharged because adequate care is not available else-where.43 We assessed the first discharge event associated with an admission to acute care rather than total hospitalizations as the latter may include hospitalization occurring after a move from the original setting and may reflect characteristics of the new location. This approach captured nearly all residents hospitalized [90.2% (220/244) of DAL and 94.9% (74/78) of LTC residents] during follow-up. Detailed information on other transitions was obtained from facility discharge tracking forms (provided at the time of transfer or death), family caregiver discharge/decedent interviews (performed around the time of transfer or death), and family caregiver interviews at 1-year follow-up (assessing all moves from baseline).

Analysis

Descriptive analyses examined the distribution of DAL resident and facility characteristics overall and by outcome status. Incidence of hospitalization was derived for DAL and LTC cohorts accounting for the occurrence of death as a competing risk using Cumulative Incidence Competing Risk curves.58

Multivariable Cox proportional hazards models,59,60 adjusted for clustering of residents within facilities, were used to examine the relative importance of resident and facility characteristics as predictors of time to first acute care hospitalization for the DAL cohort. Residents were classified into discrete outcome groups according to the date of their first event (ie, inpatient hospitalization, LTC admission or death without prior hospitalization, other transitions without prior hospitalization, no event and remained in DAL throughout the year). Residents were censored on the date of occurrence of LTC admission (DAL cohort), death, or discharge to some another setting. Those experiencing none of these events and remaining in DAL throughout the year were censored at their 1-year follow-up assessment date.

Baseline resident and facility characteristics examined as potential predictors of hospitalization were selected based on previous litera-ture.9,31-40 Resident-level variables significant (P < .05) in age-adjusted analyses were entered one at a time and retained if they remained significant predictors (P < .10) in the full model. We then incorporated health region (fixed effect) and tested the significance of each of the facility-level variables entered separately. Because of relatively high correlations among facility characteristics, we examined separate models testing the effect of each facility variable adjusting for resident characteristics.16,27

Analyses were conducted using SAS software (Version 9.2) [SAS Institute Inc., Cary, NC, USA] and R (Version 2.13-1, R Development Core Team).

Results

DAL residents with dementia were typically older widowed women (mean age 85.7 ± 6.6 years, 73.2% widowed and 78% female). Mean number of diagnoses was 4.8 ± 2.0 (range 1-14) with hypertension (56.8%), arthritis (51.7%), depression (36.0%), osteoporosis (29.7%), and coronary heart disease (28.1%) the most common. On average, residents were on 7.7 (standard deviation 3.6) medications (range 0-23) with 42.7% receiving dementia pharmacotherapy

Table 1

Baseline Sociodemographic, Health, and Functional Characteristics of Residents by Outcome Event During 1-Year Follow-Up, ACCES-DAL Dementia Cohort (n = 609)

Total Number (% of Total) Outcome Number (% of Row Total)*-* Hospitalization LTC/Death Still in DAL P value

Overall 609 220 (36.1) 90 (14.8) 298 (48.9)

Mean ± SD 85.7 ± 6.6 86.2 ± 6.0 86.5 ± 6.4 85.1 ± 7.0 .0836

65-79 123 (20.2) 39 (32.0) 15 (12.3) 68 (55.7) .5150

80-85 177 (29.1) 66 (37.3) 24 (13.6) 87 (49.2)

86-89 144 (23.7) 49 (34.0) 23 (16.0) 72 (50.0)

90+ 165 (27.1) 66 (40.0) 28 (17.0) 71 (43.0)

Sex .1957

Female 475 (78.0) 165 (34.7) 68 (14.3) 242 (51.0)

Male 134 (22.0) 55 (41.4) 22 (16.5) 56 (42.1)

Marital status .5664

Widowed 446 (73.2) 160 (35.9) 66 (14.8) 220 (49.3)

Married/partner 95 (15.6) 37 (39.0) 17 (17.9) 41 (43.2)

Never married/separated/divorced 68 (11.2) 23 (34.3) 7 (10.5) 37 (55.2)

Strength of social relationships* .0033

Moderate/high (3-5) 484 (79.5) 166 (34.4) 64 (13.3) 253 (52.4)

Low/none (0-2) 125 (20.5) 54 (43.2) 26 (20.8) 45 (36.0)

Average time involved in activities5 .0076

Most (>2/3 time) 76(12.5) 23 (30.7) 7 (9.3) 45 (60.0)

Some (1/3 to 2/3 time) 229 (37.6) 81 (35.4) 25 (10.9) 123 (53.7)

Little-none (<1/3 time) 304 (49.9) 116(38.2) 58 (19.1) 130 (42.8)

Cognition (CPS score)

Intact(0) 34 (5.6) 16 (48.5) 4(12.1) 13 (39.4) .0016

Borderline intact (1) 63 (10.3) 16 (25.4) 5 (7.9) 42 (66.7)

Mild impairment (2) 227 (37.3) 89 (39.2) 23 (10.1) 115 (50.7)

Moderate-severe-very severe Impairment (3+) 285 (46.8) 99 (34.7) 58 (20.4) 128 (44.9)

Activities of daily living (ADL score)

Independent(0) 198 (32.5) 68 (34.5) 10(5.1) 119(60.4) <.0001

Supervision required (1) 143 (23.5) 49 (34.3) 21 (14.7) 73 (51.1)

Limited impairment (2) 84(13.8) 26 (31.0) 14(16.7) 44 (52.4)

Extensive assistance required/dependent (3+) 184 (30.2) 77 (41.9) 45 (24.5) 62 (33.7)

Health instability (CHESS score)jj

Stable (0) 289 (47.5) 88 (30.6) 34(11.8) 166 (57.6) <.0001

Mild (1) 162 (26.6) 69 (42.6) 20 (12.4) 73 (45.1)

Mild-moderate (2) 108 (17.7) 41 (38.0) 19(17.6) 48 (44.4)

Moderate-high (3+) 50 (8.2) 22 (44.0) 17 (34.0) 11 (22.0)

Fatigue, <3 days <.0001

None 278 (45.7) 86 (31.1) 32 (11.6) 159 (57.4)

Minimal 236 (38.8) 92 (39.0) 32 (13.6) 112 (47.5)

Moderate-severe 95 (15.6) 42 (44.2) 26 (27.4) 27 (28.4)

Primary mode locomotion

Walks independently 182 (29.9) 51 (28.2) 15 (8.3) 115 (63.5) <.0001

Walks with assistive device 337 (55.3) 131 (38.9) 53 (15.7) 153 (45.4)

Wheelchair/scooter1 90 (14.8) 38 (42.2) 22 (24.4) 30 (33.3)

1+ falls/90 days 177 (29.1) 78 (44.1) 31 (17.5) 68 (38.4) .0036

None 432 (70.9) 142 (33.0) 59(13.7) 230 (53.4)

Depressive symptoms (DRS score)

Yes (3+) 140 (23.0) 51 (36.4) 29 (20.7) 60 (42.9) .0577

No (<3) 469 (77.0) 169 (36.1) 61 (13.0) 238 (50.9)

Aggressive behavior (ABS score)**

None (0) 370 (60.8) 140 (37.8) 48 (13.0) 182 (49.2) .5792

Moderate (1 -2) 121 (19.9) 41 (34.2) 20(16.7) 59 (49.2)

Severe (3-5) 90 (14.8) 29 (32.2) 15 (16.7) 46 (51.1)

Very severe (6+) 28 (4.6) 10(35.7) 7 (25.0) 11 (39.3)

Number of chronic conditions

Mean ± SD 4.8 ± 2.0 5.1 ± 2.0 5.2 ± 2.2 4.5 ± 1.9 .0009

0-3 173 (28.4) 50 (29.1) 20(11.6) 102 (59.3) .0072

4-5 223 (36.6) 78 (35.0) 35 (15.7) 110(49.3)

6+ 213 (35.0) 92 (43.2) 35 (16.4) 86 (40.4)

Number of medications

Mean ± SD 7.7 ± 3.6 8.4 ± 3.7 8.0 ± 3.5 7.1 ± 3.4 .0002

0-6 241 (39.6) 69 (28.8) 33 (13.8) 138 (57.5) .0111

7-8 139 (22.8) 50 (36.0) 23 (16.6) 66 (47.5)

9-10 115 (18.9) 46 (40.0) 17 (14.8) 52 (45.2)

11+ 114(18.7) 55 (48.3) 17 (14.9) 42 (36.8)

Dementia medications .8701

Yes 260 (42.7) 91 (35.0) 39 (15.0) 130 (50.0)

No 349 (57.3) 129 (37.1) 51 (14.7) 168 (48.3)

Advanced directive -Do not hospitalize .6976

Yes 63 (10.3) 21 (33.3) 8 (12.7) 34 (54.0)

No 546 (89.7) 199 (36.5) 82 (15.1) 264 (48.4)

(continued on next page)

Table 1 (continued )

Total Number (% of Total)

Outcome Number (% of Row Total)*'

P value

Hospitalization

LTC/Death

Still in DAL

Previous inpatient hospitalizations (past year) 0 1 2+

Bladder incontinence Continent

Some control, infrequent episodes Occasional incontinence Frequent episodes, no control Bowel incontinence Continent

Some control, infrequent episodes Occasional incontinence Frequent episodes, no control

413 (67.8) 131 (31.8) 64 (15.5) 217 (52.7)

138 (22.7) 54 (39.1) 16(11.6) 68 (49.3)

58 (9.5) 35 (60.3) 10(17.2) 13 (22.4)

231 (37.9) 78 (33.9) 19(8.3) 133 (57.8)

83 (13.6) 35 (42.2) 8 (9.6) 40 (48.2)

60 (9.9) 20 (33.3) 8 (13.3) 32 (53.3)

235 (38.6) 87 (37.0) 55 (23.4) 93 (39.6)

412 (67.7) 136 (33.1) 48 (11.7) 227 (55.2)

99 (16.3) 46 (46.5) 15 (15.2) 38 (38.4)

56 (9.2) 21 (37.5) 14 (25.0) 21 (37.5)

42 (6.9) 17 (40.5) 13 (31.0) 12 (28.6)

<.0001

<.0001

<.0001

ABS, Aggressive Behavior Scale; ADL, Activities of Daily Living Self-Performance Hierarchy Scale; CHESS, Changes in Health, End-stage disease and Symptoms and Signs Scale; DAL, designated assisted living; DRS, Depression Rating Scale; SD, standard deviation.

*Except where indicated otherwise.

yOne resident with other outcome (censored at date of first discharge from DAL) omitted from comparisons.

'Social relationships variable was based on a summary score of items assessing whether resident was close to someone in the facility, had a strong/supportive relationship with family, participated in social activities of longstanding interest and visited or had other interactions with longstanding social relation/family member (in past week). xActivity involvement reflected time when awake and not receiving treatments or ADL care.

"Two items (insufficient fluid, noticeable decline in food/fluid) that are usually used to calculate CHESS were not included on interRAI-Assisted Living tool. 'Includes 1 resident who was bedbound.

**The ABS is a summary scale of 4 behaviors (verbal abuse, physical abuse, socially inappropriate, or disruptive, resists care) with higher scores indicating a greater number and frequency of behavioral issues.

[primarily donepezil, 162/260 (62.3%)]. Approximately one-tenth were reported to have a "Do Not Hospitalize" advance directive and one-third had been hospitalized at least once in the previous year. Almost two-thirds (n = 388) resided in a facility with an LPN and/or RN on site 24/7 and 42.2% (n = 257) in a facility with dementia beds. Please see Tables 1 and 2 for additional baseline information.

Relative to DAL, LTC residents with dementia were more likely to be male and married (Supplementary Appendix 1). They were also more likely to have weak social relationships, be less active, and have greater health issues, cognitive and functional impairments, mood and behavioral challenges, and comorbidity. The 2 cohorts were similar in the proportion with recent falls and high levels of medication use, although residents with dementia in LTC were significantly less likely (21.6%) than those in DAL (42.7%) to be receiving dementia pharmacotherapy. LTC residents were significantly less likely (21.8%) than DAL (32.2%) residents to have been admitted to hospital during the year prior to baseline. Nearly one-third of LTC residents were reported to have a "Do Not Hospitalize" directive. LTC facilities were more likely to have not-for-profit ownership, less likely to be part of a chain with AL and LTC sites, generally older with more beds, and universally had 24/7 LPN/RN coverage on site.

During the 1-year follow-up, 220 (36.1%) DAL residents with dementia experienced an acute care hospitalization as their first event. The proportion hospitalized did not differ between those with [21/63 (33.3%)] and without [199/546 (36.4%)] a "Do Not Hospitalize" advance directive. The hospitalization rate was 51.7 per 100 person-years. The cumulative incidence was 23.3% (95% CI 19.9-26.8) at 6 months and 38.6% (95% CI 34.5-42.7) at 12 months (Figure 1A). The median LOS for hospitalizations was 14 days (IQR 5-43 days). Total bed days were 6485. Sixty-four (29.1%) had 1 or more ALC bed-days (total ALC bed-days = 1391; median ALC length of stay 12.5, range 1 -93, IQR 6.5-32.5). The total number of hospitalizations experienced by DAL residents with dementia during the 1-year follow-up was 319 (total bed days = 8553).

During the 1-year follow-up, 74 (10.7%) LTC residents with dementia experienced an acute care hospitalization as their first event with a rate of 13.1 per 100 person-years. LTC dementia residents with a "Do Not Hospitalize" directive were significantly less likely [14/225 (6.2%)]) than those without [60/466 (12.9%)] to be admitted

to hospital (P = .008). The cumulative incidence of hospitalization was 5.9% (95% CI 4.2-7.7) at 6 months and 10.3% (95% CI 8.0-12.6) at 12 months (Figure 1B), significantly lower than that observed for DAL residents (P < .001). Median LOS for hospitalizations was 7 days (IQR 3-13 days) and total bed days were 717. Only 1 admission (1.4%) had any ALC bed-days (16 total ALC bed-days). The total number of hospitalizations experienced by LTC residents with dementia during the 1-year follow-up was 100 (total bed days = 1193).

For both sites, the common causes for hospitalization are described in Supplementary Appendix 2 and the corresponding Cumulative Incidence Competing Risk curves stratified by number of previous hospitalizations are presented in Supplementary Appendix 3.

In adjusted analyses, a significantly increased risk for hospitali-zation was observed for DAL dementia residents aged 90+ years, with poor social relationships, less severe cognitive impairment, moderate to severe fatigue, high medication use (11 + medications), and 2+ hospitalizations during the preceding year (Table 3). Residents with some health instability and those with bowel incontinence had a modestly increased risk. DAL residents from specific (primarily rural) health regions showed a higher risk of hospitalization. As community size was highly correlated with region it was not retained in the models.

In models adjusted for resident characteristics and health region, a significantly higher likelihood of hospitalization was observed for residents from DAL facilities that were smaller (<20 DAL spaces). Residents from facilities with no chain affiliation also showed an increased risk of hospitalization (Table 4). Given the increased hos-pitalization risk for residents within rural regions as well as findings showing these regions were significantly more likely to include DAL facilities that were small, not part of a chain and had neither a LPN nor RN on site, we examined separate models for each of our facility factors excluding health region. The results showed stronger estimates for the above facility factors [eg, adjusted hazard ratios of 2.30 (1.55-3.39) for residents of small facilities and 2.08 (1.18-3.64) for residents of facilities with no chain affiliation] as well as a significant hospitalization risk for resident of facilities with no LPN and/or RN on site [adjusted hazard ratio of 1.61 (1.16-2.24)].

Baseline Care System and Facility Characteristics of Residents by Outcome Event During 1-Year Follow-Up, ACCES-DAL Dementia Cohort (n = 609)

Total Number (% of Total) Outcome Number (% of Row Total)* P value

Hospitalization LTC/Death Still in DAL

Overall 609 220 (36.1) 90 (14.8) 298 (48.9)

Region

1 (urban) 166 (27.3) 49 (29.7) 23 (13.9) 93 (56.4) .0182

2 (mixed urban/rural) 135 (22.2) 48 (35.6) 22 (16.3) 65 (48.2)

3 (rural) 77 (12.6) 41 (53.3) 11 (14.3) 25 (32.5)

4 (urban) 174 (28.6) 57 (32.8) 24(13.8) 93 (53.5)

5 (rural) 57 (9.4) 25 (43.9) 10(17.5) 22 (38.6)

Ownership .0075

For-profit 233 (38.3) 67 (28.9) 34 (14.7) 131 (56.5)

Not-for-profit/RHA 376 (61.7) 153 (40.7) 56 (14.9) 167 (44.4)

Part of chain

No/RHA operated 70(11.5) 33 (47.1) 9(12.9) 28 (40.0) .0139

Yes - AL chain 181 (29.7) 67 (37.0) 16(8.8) 98 (54.1)

Yes - AL/LTC chain 358 (58.8) 120 (33.6) 65 (18.2) 172 (48.2)

Year DAL spaces opened

Before 2002 124 (20.4) 42 (33.9) 20(16.1) 62 (50.0) .9747

2002 or 2003 203 (33.3) 74 (36.6) 30 (14.9) 98 (48.5)

2004 or later 282 (46.3) 104 (36.9) 40 (14.2) 138 (48.9)

Number of DAL spaces

<20 58 (9.5) 29 (50.0) 10(17.2) 19 (32.8) .0388

20-29 103 (16.9) 40 (38.8) 21 (20.4) 42 (40.8)

30-39 147 (24.1) 48 (32.9) 19(13.0) 79 (54.1)

40+ 301 (49.4) 103 (34.2) 40(13.3) 158 (52.5)

Number of total spaces

<55 90 (14.8) 39 (43.3) 15 (16.7) 36 (40.0) .3548

55-89 147 (24.1) 59 (40.1) 21 (14.3) 67 (45.6)

90-147 138 (22.7) 42 (30.4) 21 (15.2) 75 (54.4)

148+ 234 (38.4) 80 (34.3) 33 (14.2) 120 (51.5)

Levels of care on sitey .4266

DAL only/DAL+ Equivalent/lower 507 (83.3) 178 (35.2) 78 (15.4) 250 (49.4)

DAL + higher level 102 (16.8) 42 (41.2) 12 (11.8) 48 (47.1)

Dementia beds on site .0071

Yes 257 (42.2) 82 (31.9) 42 (16.3) 133 (51.8)

No 352 (57.8) 138 (39.2) 48 (13.6) 165 (46.9)

LPN/RN coverage on site .0187

Neither on site 158 (25.9) 71 (44.9) 27 (17.1) 60 (38.0)

LPN and/or RN <24/7 63 (10.3) 23 (36.5) 6 (9.5) 34 (54.0)

LPN and/or RN 24/7 388 (63.7) 126 (32.6) 57 (14.7) 204 (52.7)

Physician (GP) affiliated with site

No 408 (67.0) 144 (35.3) 67 (16.4) 197 (48.3) .5031

Yes, office on site 95 (15.6) 34 (35.8) 10(10.5) 51 (53.7)

Yes, no office on site 106(17.4) 42 (40.0) 13 (12.4) 50 (47.6)

Community size .0005

<10,000 122 (20.0) 58 (47.5) 12 (9.8) 52 (42.6)

10,000-99,999 166 (27.3) 68 (41.0) 31 (18.7) 67 (40.4)

1 million+ 321 (52.7) 94 (29.4) 47 (14.7) 179 (55.9)

DAL, designated assisted living; GP, general practitioner; RHA, regional health authority; SD, standard deviation; 24/7 = 24 hours/day, 7 days/week. *One resident with other outcomes (censored at date of first discharge from DAL) omitted from comparisons.

Equivalent level of care = private AL, residential, respite (not in LTC), community support and transition beds; lower level of care = independent living, lodge, condo; higher level of care = LTC (including respite), acute care.

Fig. 1. Cumulative incidence of hospitalization during 1-year follow-up, DAL (A) (n = 609) and LTC (B) (n = 691) dementia cohorts.

HRs for Hospitalization During 1-Year Follow-Up, ACCES-DAL Dementia Cohort (n = 609)

Characteristic HR (95% CIs)

Age-Adjusted Fully Adjusted - 1* Fully Adjusted - 2y

65-79, ref gp

80-85 1.23 (0.81-1.88) 1.22 (0.80-1.86)

86-89 1.07 (0.68-1.66) 1.05 (0.67-1.64)

90+ 1.39 (1.02-1.90) 1.39 (1.02-1.89)

Female 0.75 (0.56-1.00) 0.83 (0.59-1.16) 0.84 (0.60-1.19)

Strength of social relationships

Moderate/high, ref gp

Low/none 1.43 (1.13-1.79) 1.38 (1.06-1.81) 1.38 (1.06-1.81)

Cognition (CPS score)

Intact (0), ref gp

Borderline Intact (1) 0.44 (0.23-0.87) 0.35 (0.18-0.67) 0.35 (0.18-0.67)

Mild Impairment (2) 0.76 (0.46-1.27) 0.57 (0.35-0.93) 0.56 (0.34-0.93)

Mod-severe-very severe Impairment (3+) 0.72 (0.43-1.19) 0.54 (0.33-0.89) 0.53 (0.32-0.88)

Activities of daily living (ADL score)

Independent (0), ref gp

Supervision required (1) 1.05 (0.72-1.52) - -

Limited impairment (2) 1.00 (0.62-1.62) - -

Extensive assistance Req'd/dependent (3+) 1.54 (1.12-2.10) - -

Health instability (CHESS score)

Stable (0), ref gp

Mild (1) 1.51 (1.13-2.01) 1.28 (0.97-1.70)* 1.30 (0.99-1.71)*

Mild-moderate (2) 1.41 (0.97-2.05)z 1.11 (0.77-1.61) 1.13 (0.79-1.63)

Moderate-high (3+) 2.45 (1.38-4.35) 1.34 (0.67-2.71) 1.32 (0.67-2.59)

Fatigue, <3 days

None (0), ref gp

Minimal (1) 1.40 (1.02-1.90) 1.28 (0.91-1.82) 1.35 (0.98-1.86)*

Moderate-severe (2+) 1.99 (1.34-2.96) 1.65 (1.03-2.63) 1.74 (1.12-2.69)

Primary mode locomotion

Walks independently, ref gp

Walks with device 1.55 (1.09-2.20) - -

Wheelchair/scooter 2.01 (1.25-3.23) - -

1+ falls/90 days 1.58 (1.24-2.01) - -

Number of chronic conditions

0-3, ref gp

4-5 1.30 (0.92-1.82) 1.07 (0.71-1.62) -

6+ 1.89 (1.40-2.56) 1.30 (0.84-2.00) -

Number of medications

0-6, ref gp

7-8 1.41 (1.01-1.97) 1.28 (0.92-1.78) 1.35 (0.98-1.86)*

9-10 1.52 (1.03-2.24) 1.27 (0.81 - 2.00) 1.38 (0.94-2.03)*

11+ 2.14 (1.51-3.02) 1.57 (1.05-2.34) 1.72 (1.14-2.59)

Previous inpatient hospitalizations <1 year

0, ref gp

1 1.30 (0.95-1.79)* 1.09 (0.78-1.53) 1.10(0.80-1.53)

2+ 2.74 (1.95-3.86) 2.28 (1.52-3.42) 2.31 (1.54-3.47)

Bowel incontinence

Any incontinence 1.55 (1.18-2.02) 1.35 (1.00-1.82)* 1.34 (0.99-1.81)*

System/facility factors

Region

1 (urban), ref gp

2 (mixed urban/rural) 1.32 (0.86-2.02) 0.84 (0.55-1.26) 0.88 (0.59-1.31)

3 (rural) 2.51 (1.55-4.08) 1.90 (1.10-3.30) 2.01 (1.20-3.38)

4 (urban) 1.15 (0.73-1.81) 1.02 (0.67-1.55) 1.03 (0.69-1.53)

5 (rural) 1.83 (1.02-3.28) 1.61 (0.93-2.80)z 1.62 (0.95-2.76)*

Community size

<10,000, ref gp

10,000-99,999 0.86 (0.58-1.26) - -

1 million+ 0.51 (0.35-0.75) - -

ADL, Activities of Daily Living Self-Performance Hierarchy Scale; CHESS, Changes in Health, End-Stage Disease and Symptoms and Signs Scale; DAL, designated assisted living;

HR, hazard ratio; ref gp, reference group.

Bold values are statistically significant (P < .05).

*Model 1 derived from Cox proportional hazards regression models (first event analysis), with adjustment for clustering by facility, includes number of chronic conditions. yModel 2 derived from Cox proportional hazards regression models (first event analysis), with adjustment for clustering by facility, excludes number of chronic conditions. *P < .10.

Discussion

The extent and adverse consequences of acute care admissions for older adults with dementia have been well documented.18,28,41-47 Yet,

research is scarce regarding the impact of different models of care (including organizational, staffing, and service characteristics) on risk of hospitalization among vulnerable populations with dementia.9,17 In our investigation of diverse care settings in Alberta, residents with

Adjusted HRs for Hospitalization During 1-Year Follow-Up Associated With Selected Facility Factors, ACCES-DAL Dementia Cohort (n = 609)

Adjusted HR (95% CI)*

Model Ay Number of DAL spaces <20 20-29 30-39

40+, ref group Model By Part of chain No/RHA-operated Yes-AL chain

Yes-AL/LTC chain, ref group

1.62 (1.02-2.56)

0.94 (o.58-1.55) 1.02 (o.69-1.49)

1.63 (0.93-2.83)z 1.24 (0.94-1.64)

CI, confidence interval; DAL, designated assisted living; HR, hazard ratio; RHA,

regional health authority.

Bold values are statistically significant (P < .05).

*Derived from Cox proportional hazards regression models (first event analysis), with adjustment for clustering by facility.

yModels were adjusted for age, sex, strength of social relationships, cognition, health instability, fatigue, comorbidity, number of medications, previous hospitali-zations, bowel incontinence, and region. zP < .10.

dementia receiving care in DAL facilities showed a cumulative incidence of hospitalization almost 4 times higher than those in LTC (38.6% vs 10.3% over 12 months). The length of hospital stay and use of ALC bed days was also significantly higher among DAL residents compared with those in LTC.

Our findings are consistent with American studies,38,40 including the US Collaborative Studies for Long-Term Care,9 which also showed higher hospitalization rates for residents with dementia in AL relative to LTC. Although direct comparisons are difficult across methodologically diverse studies, previously reported hospitalization rates for community- and clinic-based samples with dementia28,61 are generally lower than rates among residents with dementia in AL but higher than rates reported among those in LTC facilities.

The lower rate of hospitalization evident among LTC residents with dementia, despite their overall poorer baseline health and functional status (Supplementary Appendix 1), raises a number of important clinical and policy questions. A lower probability of hospitalization for older adults with dementia in nursing homes (eg, compared with those in the community) has been reported else-where.18,62,63 This may reflect changing goals of care (eg, in response to progression of the dementia) as well as service substitution by LTC. Our findings27 and those from the US7,9,37 suggest that the AL model may have a limited capacity to care for medically complex residents with dementia.

Advance care planning was also more likely to be in place and adhered to within LTC compared with DAL. The relevance of advance directives in reducing potentially avoidable hospitalizations was recently demonstrated in the CASCADE prospective cohort study of nursing home residents with advanced dementia,30 which showed that the absence of a "Do Not Hospitalize" order was the most important (and only modifiable) factor associated with hospitaliza-tion for acute illness.

We previously showed that DAL residents with a diagnosis of dementia were significantly more likely to be admitted to a LTC fa-cility12 but no more likely than residents without dementia to be hospitalized.27 This is in line with reports illustrating a less pronounced impact of dementia on hospitalization among older populations with greater levels of co-morbidity.41 Findings of the current study suggest that particular sub-groups of DAL residents with dementia (including those aged 90+ years, with poor social relationships, moderate to severe fatigue, health instability, high medication

use, and 2+ recent hospitalizations) are at greater risk for admission to an acute care hospital. Not surprisingly, the relative magnitude of risk associated with selected clinical factors (many of which were correlated with previous hospitalizations) was greater in the age-adjusted models. Increased age, higher levels of fatigue and health instability, and 2+ previous hospitalizations likely indicate greater frailty and/or predisposition to being hospitalized.18,64 These characteristics could be used to define at risk target groups for interventions.

For vulnerable older AL residents with dementia, it may be important to consider end-of life care priorities including enhanced access to palliative care.34 Preliminary data suggest that increasing the availability of palliative care services within the AL sector may significantly reduce the likelihood of hospitalization and LTC admissions.38

While the relationship between medication use and hospitalization risk may reflect the number of drugs acting as a marker of comorbidity and/or illness severity, many hospital admissions of older individuals are drug-related.65 Optimizing medication prescription and administration in AL, particularly for frail residents with self-care and/or communication difficulties, have been highlighted as key areas

requiring improvement.5,10,24,66,67

The finding of a significant independent association between poor social relationships and/or engagement and increased hospitalization risk among DAL residents with dementia may reflect the importance of close social ties in slowing symptom progression in dementia.68 Higher levels of participation (in group and/or solitary activities) have also been associated with a longer time to discharge from AL.12,69 Although the mechanisms remain unclear, the potential for interventions aimed at increasing staff awareness of social vulnerability and opportunities for resident engagement to reduce adverse care transitions represent important areas for future research.

The higher hospitalization risk observed for the small number of DAL dementia residents with a CPS score of 0 (relatively intact) may be due to an increased likelihood for less impaired residents to exhibit clearer illness or symptom presentations (prompting a hospital transfer), a greater hesitancy among facility staff and family members to hospitalize more impaired residents, and/or the presence of other unique characteristics among this subgroup. Those with a CPS score of 0 were generally less impaired on all other health and functional characteristics.

Few studies have investigated the impact of facility-level factors on AL residents' quality of care or health outcomes and findings remain unclear.9,16,17,40 Residents of facilities with certain characteristics (eg, rural, smaller, not affiliated with a ownership chain of AL and/or LTC facilities) were more likely to be admitted to hospital. Given these facility characteristics are inter-related, it is difficult to delineate the underlying mechanisms associated with these factors. All 3 may be linked to lower levels of clinical oversight and services.70 DAL facilities from rural regions were also those with lower levels of professional staffing and service availability. Others have shown that a higher proportion of licensed nursing staff hours (whether RN or LPN) or more hours of RN staff time per resident might reduce hos-pitalization risk in residential care and AL.16,21 This effect has also been reported to be stronger with increasing dementia case mix.21 The intensity and skill mix of nursing staff used in AL should be commensurate with the health needs of residents with dementia. Our study suggests that greater access to skilled nursing care may be needed to both monitor for early manifestations of declining health and ensure the capacity to provide additional care on site.

Strengths of our study include the large sample, diverse range of resident- and facility-level characteristics examined, and comprehensive prospective data collection. Some limitations warrant consideration. Approximately 28% and 42% of eligible DAL and LTC

residents were not enrolled, respectively. Although their age and sex distributions were similar to participants, this may limit the gener-alizability of our findings. Our study was restricted to residents of publicly subsidized AL spaces in Alberta. These settings are subject to provincial care standards with admission through a single point of entry. Caution is warranted in generalizing our results to private-pay institutions or to AL facilities elsewhere. Our data do not permit any clear interpretation of the importance of specific care practices or services (present or absent in a facility) to residents' risk of hospi-talization. We also did not have access to comparable data on hos-pitalization rates for older adults with dementia receiving care in their own homes. Finally, the data collection took place between 2006 and 2009. Since then, changes have taken place within the Alberta AL sector, and their possible effect on hospitalization is unknown.

Conclusions

Nearly 40% of DAL residents with dementia in Alberta were hospitalized over a year, a rate substantially higher than that observed for dementia residents in LTC. Many of the resident and facility level factors predictive of hospitalization parallel those reported for nursing homes29-36,71 suggesting similar opportunities for targeting and/or developing interventions to reduce potentially inappropriate care transitions.18,19,39,72,73 Although we believe that a proportion of the hospital admissions for DAL residents with dementia are potentially preventable, it is important to note that not all of these hos-pitalizations are inappropriate. As noted for the nursing home sector,29 further research in

AL39,74

is required to better define strategies to prevent potentially inappropriate hospitalizations among this vulnerable population.

Acknowledgments

The authors thank Deanna Wanless, Anna Charlton, Cheri Komar (Study Coordinators), Drs David Zimmerman and Jean Parboosingh (Study Advisors), our research staff, and the facilities, residents and their family members who participated in ACCES.

Supplementary Data

Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/jjamda.2015.01.079.

References

1. Alzheimer's Association. 2013 Alzheimer's disease facts and figures. Alzheimer Dement 2013;9:208-245.

2. Smetanin P, Kobak P, Briante C, et al. Rising tide: The impact of dementia in Canada 2008 to 2038. Toronto (ON): RiskAnalytica; 2009.

3. Callahan CM, Schubert CC. Dementia: The complexities of comorbidity in dementia. Nat Rev Neurol 2014;10:184-186.

4. Vu M, Hogan DB, Patten SB, et al. A comprehensive profile of the sociodemo-graphic, psychosocial and health characteristic of Ontario home care clients with dementia. Chronic Dis Inj Can 2014;34:132-144.

5. Zimmerman S, Sloane PD, Reed D. Dementia prevalence and care in assisted living. Health Aff 2014;33:658-666.

6. Golant SM. Do impaired older persons with health care needs occupy U.S. assisted living facilities? An analysis of six national studies. J Gerontol B Psychol Sci Soc Sci 2004;59:S68-S79.

7. Rosenblatt A, Samus QM, Steele CD, et al. The Maryland Assisted Living Study: Prevalence, recognition, and treatment of dementia and other psychiatric disorders in the assisted living population of central Maryland. J Am Geriatr Soc 2004;52:1618-1625.

8. Zimmerman S, Sloane PD, Heck E, et al. Introduction: Dementia care and quality of life in assisted living and nursing homes. Gerontologist 2005;45:5-7.

9. Sloane PD, Zimmerman S, Gruber-Baldini AL, et al. Health and functional outcomes and health care utilization of persons with dementia in residential care and assisted living facilities: Comparison with nursing homes. Gerontologist 2005;45:124-132.

10. McNabney MK, Samus QM, Lyketsos CG, et al. The spectrum of medical illness and medication use among residents of assisted living facilities in Central Maryland. J Am Med Dir Assoc 2008;9:558-564.

11. Samus QM, Onyike CU, Johnston D, et al. Twelve-month incidence, prevalence, persistence, and treatment of mental disorders among individuals recently admitted to assisted living facilities in Maryland. Int Psychogeriatr 2013;25: 721-731.

12. Maxwell CJ, Soo A, Hogan DB, et al. Predictors of nursing home placement from assisted living settings in Canada. Can J Aging 2013;32:333-348.

13. Rockwood JKH, Richard M, Garden K, et al. Precipitating and predisposing events and symptoms for admission to assisted living or nursing home care. Can Geriatr J 2014;17:16-21.

14. Canadian Centre for Elder Law. Discussion paper on assisted living: Past, present and future legal trends in Canada. Available at: http://www.bcli.org/ project/assisted-living-past-present-and-future-legal-trends-canada; 2008. Accessed June 11, 2014.

15. Alberta Health and Wellness. Continuing Care Strategy: Aging in the Right Place. Alberta Health and Wellness, December 2008. Available at: http://www. health.alberta.ca/documents/Continuing-Care-Strategy-2008.pdf. Accessed June 11, 2014.

16. Zimmerman S, Sloane PD, Eckert JK, et al. How good is assisted living? Findings and implications from an outcomes study. J Gerontol B Psychol Sci Soc Sci 2005;60:S195-S204.

17. Zimmerman S, Anderson WL, Brode S, et al. Systematic review: Effective characteristics of nursing homes and other residential long-term care settings for people with dementia. J Am Geriatr Soc 2013;61:1399-1409.

18. Callahan CM, Arling G, Tu W, et al. Transitions in care for older adults with and without dementia. J Am Geriatr Soc 2012;60:813-820.

19. Bellantonio S, Kenny AM, Fortinsky RH, et al. Efficacy of a geriatrics team intervention for residents in dementia-specific assisted living facilities: Effect on unanticipated transitions. J Am Geriatr Soc 2008;56:523-528.

20. Strain LA, Maxwell CJ, Wanless D, et al. Designated Assisted Living (DAL) and Long-Term Care (LTC) in Alberta: Selected Highlights from the Alberta Continuing Care Epidemiological Studies (ACCES). Edmonton, AB: ACCES Research Group, University of Alberta. Available at: http://hdl.handle.net/ 10402/era.23779; 2011. Accessed June 11, 2014.

21. Stearns SC, Park J, Zimmerman S, et al. Determinants and effects of nurse staffing intensity and skill mix in residential care/assisted living settings. Gerontologist 2007;47:662-671.

22. Dobbs D, Hayes J, Chapin R, et al. The relationship between psychiatric disorders and the ability to age in place in assisted living. Am J Geriatr Psychiatry 2006;14:613-620.

23. Phillips CD, Munoz Y, Sherman M, et al. Effects of facility characteristics on departures from assisted living: Results from a National Study. Gerontologist 2003;43:690-696.

24. Sloane PD, Gruber-Baldini AL, Zimmerman S, et al. Medication undertreatment in assisted living settings. Arch Intern Med 2004;164:2031-2037.

25. Sloane PD, Zimmerman S, Perez R, et al. Physician perspectives on medical care delivery in assisted living. J Am Geriatr Soc 2011;59:2326-2331.

26. Zimmerman S, Gruber-Baldini AL, Hebel JR, et al. Nursing home characteristics related to Medicare costs for residents with and without dementia. Am J Alzheimers Dis Other Demen 2008;23:57-65.

27. Hogan DB, Amuah JE, Strain LA, et al. High rates of hospital admission among older residents in assisted living facilities: Opportunities for intervention and impact on acute care. Open Med 2014;8:33-45.

28. Phelan EA, Borson S, Grothaus L, et al. Association of incident dementia with hospitalizations. JAMA 2012;307:165-172.

29. Ouslander JG, Maslow K. Geriatrics and the triple aim: Defining preventable hospitalizations in the long-term care population. J Am Geriatr Soc 2012;60: 2313-2318.

30. Givens JL, Selby K, Goldfeld KS, et al. Hospital transfers of nursing home residents with advanced dementia. J Am Geriatr Soc 2012;60:905-909.

31. O'Malley AJ, Caudry DJ, Grabowski DC. Predictors of nursing home residents' time to hospitalization. Health Serv Res 2011;46:82-104.

32. Becker MA, Boaz TL, Andel R, et al. Predictors of preventable nursing home hospitalizations: The role of mental disorders and dementia. Am J Geriatr Psychiatry 2010;18:475-482.

33. Grabowski DC, Stewart KA, Broderick SM, et al. Predictors of nursing home hospitalization: A review of the literature. Med Care Res Rev 2008;65:3-39.

34. Mitchell SL, Teno JM, Intrator O, et al. Decisions to forgo hospitalization in advanced dementia: A nationwide study. J Am Geriatr Soc 2007;55: 432-438.

35. Gruneir A, Miller SC, Intrator O, et al. Hospitalization of nursing home residents with cognitive impairments: The influence of organizational features and state policies. Gerontologist 2007;47:447-456.

36. Carter MW, Porell FW. Vulnerable populations at risk of potentially avoidable hospitalizations: The case of nursing home residents with Alzheimer's disease. Am J Alzheimers Dis Other Demen 2005;20:349-358.

37. Lyketsos CG, Samus QM, Baker A, et al. Effect of dementia and treatment of dementia on time to discharge from assisted living facilities: The Maryland assisted living study. J Am Geriatr Soc 2007;55:1031-1037.

38. Dobbs D, Meng H, Hyer K, et al. The influence of hospice use on nursing home and hospital use in assisted living among dual-eligible enrollees. J Am Med Dir Assoc 2012;13:189e9-189e13.

39. Becker M, Boaz T, Andel R, et al. Predictors of avoidable hospitalizations among assisted living residents. J Am Med Dir Assoc 2012;13:355-359.

40. Hedrick SC, Guihan M, Chapko MK, et al. Assisted living pilot program: Health outcomes. J Aging Health 2009;21:190-207.

41. Bynum JPW, Rabins PV, Weller W, et al. The relationship between a dementia diagnosis, chronic illness, Medicare expenditures, and hospital use. J Am Ger-iatrSoc 2004;52:187-194.

42. Draper B, Karmel R, Gibson D, et al. The Hospital Dementia Services Project: Age differences in hospital stays for older people with and without dementia. Int Psychogeriatr 2011;23:1649-1658.

43. Canadian Institute for Health Information. Analysis in Brief: Alternate Level of Care in Canada. 2009. Available at: https://secure.cihi.ca/free_products/ALC_ AlB_FlNAL.pdf. Accessed February 9, 2015.

44. Lin PJ, Fillit HM, Cohen JT, et al. Potentially avoidable hospitalizations among Medicare beneficiaries with Alzheimer's disease and related disorders. Alzheimers Dement 2013;9:30-38.

45. Seitz DP, Gill SS, Gruneir A, et al. Effects of dementia on postoperative outcomes of older adults with hip fractures: A population-based study. J Am Med Dir Assoc 2014;15:334-341.

46. Teno JM, Gozalo P, Mitchell SL, et al. Survival after multiple hospitalizations for infections and dehydration in nursing home residents with advanced cognitive impairment. JAMA 2013;310:319-320.

47. Fong TG, Jones RN, Marcantonio ER, et al. Adverse outcomes after hospitali-zation and delirium in persons with Alzheimer disease. Ann Intern Med 2012; 156:848-856.

48. Wanless D, Strain LA, Maxwell CJ. Designated Assisted Living (DAL) and Long-Term Care (LTC) in Alberta: Alberta Continuing Care Epidemiological Studies (ACCES) Methodology. Edmonton, AB: ACCES Research Group, University of Alberta. Available at: http://hdl.handle.net/10402/era.23788; 2011. Accessed June 11, 2014.

49. Hirdes JP, Ljunggren G, Morris JN, et al. Reliability of the interRAl suite of assessment instruments: A 12-country study of an integrated health information system. BMC Health Serv Res 2008;8:277.

50. Poss JW, Jutan NM, Hirdes JP, et al. A review of evidence on the reliability and validity of Minimum Data Set data. Healthc Manage Forum 2008;21:33-39.

51. Hartmaier SL, Sloane PD, Guess HA, et al. Validation of the Minimum Data Set Cognitive Performance Scale: Agreement with the Mini-Mental State Examination. J Gerontol A Biol Sci Med Sci 1995;50:M128-M133.

52. Morris JN, Fries BE, Morris SA. Scaling ADLs with the MDS. J Gerontol A Biol Sci Med Sci 1999;54:M546-M553.

53. Burrows AB, Morris JN, Simon SE, et al. Development of a minimum data set-based depression rating scale for use in nursing homes. Age Ageing 2000;29:165-172.

54. Hirdes JP, Frijters DH, Teare GF. The MDS-CHESS Scale: A new measure to predict mortality in institutionalized older people. J Am Geriatr Soc 2003;51: 96-100.

55. Perlman CM, Hirdes JP. The aggressive behavior scale: A new scale to measure aggression based on the minimum data set. J Am Geriatr Soc 2008;56: 2298-2303.

56. Wodchis WP, Naglie G, Teare GF. Validating diagnostic information on the Minimum Data Set in Ontario hospital-based long term care. Med Care 2008; 46:882-887.

57. Canadian Institute for Health Information. Final Report—The Canadian Enhancement of 1CD-10. Canadian Institute for Health Information, 2001.

58. Berry SD, Ngo L, Samelson EJ, et al. Competing risk of death: An important consideration in studies of older adults. J Am Geriatr Soc 20l0;58: 783-787.

59. Murphy TE, Han L, Allore HG, et al. Treatment of death in the analysis of longitudinal studies of gerontological outcomes. J Gerontol A Biol Sci Med Sci 2011;66:109-114.

60. Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc 1999;94:496-509.

61. Rudolph JL, Zanin NM, Jones RN, et al. Hospitalization in community-dwelling persons with Alzheimer's disease: Frequency and causes. J Am Geriatr Soc 2010;58:1542-1548.

62. Fillenbaum G, Heyman A, Peterson BL, et al. Use and cost of hospitalization of patients with AD by stage and living arrangement: CERAD XXI. Neurology 2001;56:201-206.

63. Kane RL, Atherly A. Medicare expenditures associated with Alzheimer disease. Alzheimer Dis Assoc Disord 2000;14:187-195.

64. Hogan DB, Freiheit EA, Strain LA, et al. Comparing frailty measures in their ability to predict adverse outcomes among older residents of assisted living. BMC Geriatr 2012;12:56.

65. Canadian 1nstitute of Health 1nformation. Adverse drug reaction-related hos-pitalizations among Canadian seniors, 2006-2011. Canadian 1nstitute of Health Information. Available at: https://secure.cihi.ca/free_products/Hospitalizations %20for%20ADR-ENweb.pdf 2013. Accessed June 11, 2014.

66. Young HM, Gray SL, McCormick WC, et al. Types, prevalence, and potential clinical significance of medication administration errors in assisted living. J Am Geriatr Soc 2008;56:1199-1205.

67. Zimmerman S, Love K, Sloane PD, et al. Medication administration errors in assisted living: Scope, characteristics, and the importance of staff training. J Am Geriatr Soc 2011;59:1060-1068.

68. Norton MC, Piercy KW, Rabins PV, et al. Caregiver-recipient closeness and symptom progression in Alzheimer disease: The Cache County Dementia Progression Study. J Gerontol B Psychol Sci Soc Sci 2009;64:560-568.

69. Tighe SK, Leoutsakos JM, Carlson MC, et al. The association between activity participation and time to discharge in the assisted living setting. 1nt J Geriatr Psychiatry 2008;23:586-591.

70. Beeber AS, Zimmerman S, Reed D, et al. Licensed nurse staffing and health service availability in residential care and assisted living. J Am Geriatr Soc 2014;62:805-811.

71. Cherubini A, Eusebi P, Dell'Aquila G, et al. Predictors of hospitalization in Italian nursing home residents: The U.L.I.S.S.E. Project. J Am Med Dir Assoc 2012;13: 84e5-84e10.

72. Bynum JPW, Andrews A, Sharp S, et al. Fewer hospitalizations result when primary care is highly integrated into a continuing care retirement community. Health Aff 2011;30:975-984.

73. Loeb M, Carusone SC, Goeree R, et al. Effect of a clinical pathway to reduce hospitalizations in nursing home residents with pneumonia. A randomized controlled trial. JAMA 2006;295:2503-2510.

74. Levenson SA. On assisted living, are we hitting the mark or missing the boat? J Am Med Dir Assoc 2012;13:314-315.