Skip to main content

Table 1 Baseline characteristics of the studies included

From: Hospitalization at the end of life among nursing home residents with dementia: a systematic review

First author, year

Country

Data source

Year of data

Sample (with dementia)

Inclusion/ exclusion criteria

How was dementia assessed?

Mean age at death

(% females)

Aaltonen,

2014 [33]

Finland

Nationwide registry data (Care register for health care and Care register for social care)

Causes of Death Register

2002–2008

13,159a NHR

Died at 70 years or older

In NH during their last months of life (in care both 6 months and 3 months before death, stayed there for ≥90 days during the last 6 months of life)

Any cause of death (immediate, underlying, intermediate, contributing) with ICD-10 codes: F00, F01, F02, F03, G30

87 years (76%)

Agar,

2017 [34]

Australia

Face-to-face or telephone interviews

Nursing home and medical records

Questionnaires

2013–2014

64b NHR from 10 NH

Nursing homes with ≥50% of residents with dementia providing intensive levels of care

Residents with advanced dementia

Medical records, FAST, AKPS

Advanced dementia:

- documented diagnosis of dementia

- and FAST (≥6a, stable for 1 month)

- and AKPS ≤50

85.8 years (58%)

Allers, 2018 [13]

Germany

Health insurance claims data (DAK)

Long-term care insurance data

2010–2014

not reported

≥65 years old, newly admitted to a NH

Insured continuously for at least 365 days without NH placement before

ICD-10 codes in the quarter of NH admission: F00.x, F01.x, F02.0, F02.3, F03, G30.x, G31.0, G31.1, G31.82, G31.9, R54

not reported

Cai, 2016 [35]

USA

MDS 2.0

Medicare beneficiary summary file

Medicare claims

2007–2010

293,967 NHR

≥65 years old

In NH ≥90 days before death

Continuously enrolled in Medicare fee-for-service Plans

Medicare–Medicaid dually eligible during the last 30 days of life

CPS constructed of MDS 2.0 data

- mild cognitive impairment: CPS 0–2

- moderate cognitive impairment: CPS 3–4

- severe cognitive impairment: CPS 5–6

CPS 3–4:

85.9 years

(70.1%)c

CPS 5–6:

85.7 years

(76.3%)c

Gessert, 2008 [36]

USA

Administrative databases from Centers for Medicare and Medicaid services

MDS

2000–2001

3703 NHR

Urban and rural NH

≥67 years old

Not enrolled in Health Maintenance Organization

No hospice benefits during 2 years prior to death

Not comatose

Severe and persistent cognitive impairment based on the CPS

- CPS = 6 on ≥2 consecutive MDS reports at least 60 days apart

- absence of MDS reports with CPS ≤4 there after

87.1 years

(76.4%)c

Hendriks, 2017 [20]

Netherlands

Questionnaires from the Dutch end-of-life in dementia study (DEOLD)

2007–2011

330 residents from 34 LTCF

Residents admitted to psychogeriatric wards

Known diagnosis of dementia upon nursing home admission (by a physician)

Advanced dementia: CPS 5–6 + GDS 7

85.2 years (not reported)

Houttekier, 2014 [21]

Belgium

Questionnaires (with access to medical files) from the “Dying well with Dementia” study

2010

195 NHR from 69 NH

Inclusion process in two steps:

First: residents had to be completely care dependent for ADL and disoriented in time and space or had to have a Katz scale score ≥ 3

Second: Resident had to have dementia (reported by nurse or GP)

After inclusion CPS and GDS was assessed:

- mild/moderate: CPS < 5, GDS < 7

- severe: CPS ≥5 and GDS < 7 or CPS < 5 and GDS = 7

-very severe/ advanced: CPS ≥5, GDS = 7

< 85 years: 31.7%

85–90 years: 37.7%

> 90 years: 30.6%

(61.4%)

Krishnan, 2015 [37]

Canada

MDS 2.0

Medical charts

Death certificates

2010–2013

58 NHR from 1 NH

All residents were included

Dementia recorded on death certificate as underlying or immediate cause of death

not reported

Lamberg, 2005 [30]

USA

MDS

Medical long-term care records

2001–2003

240 NHR from 1 NH

Long term residents (stay ≥30 days)

Advanced dementia (CPS 5–6)

CPS 5–6

Long term care medical records to identify cause of cognitive impairment

92 years (median)

(75.8%)

Li, 2013 [22]

USA

MDS 2.0

Medicare beneficiary file

Hospice and hospital claims

2003–2007

143,980 NHRc

≥65 years old

> 3 months in nursing home

No rehabilitation or postacute stay

Not comatose

No transfers to another NH after last assessment

Not enrolled in managed care in last 30 days of life

Diagnosis of Alzheimer’s disease or other dementia on the last full MDS assessment

87.4 years

(71.9%)c

Livingston, 2013 [31]

UK

Care home resident records

Interviews

Questionnaires

 

30 NHRd from 1 NH

Living in the NH for at least 1 month before death

Resident records had to be available

Medical records

Diagnosis of dementia or suspected dementia (symptoms fulfilling standard dementia criteria)

not reported

Sloane, 2008 [15]

USA

(Telephone) interviews with staff and family

Study cohort from the Collaborative Studies of long-term care (CS-LTC)

2002–2005

247 NHR

NHR who spent 15 out of 30 days in a NH

Died no more than 3 days after leaving the NH

NH staff members were asked whether the decedent was an Alzheimer’s type resident 3 months before death and if dementia was a contributing factor towards the resident’s death

not reported

Temkin-Greener, 2013 [32]

USA

Nationwide administrative data from the chronic condition data warehouse

Medicare denominator files

MDS

2003–2007

384,355 NHR

Died in NH or within 8 days of discharge to a different care setting

Not enrolled in managed care in last 30 days of life

Not from Virgin Islands or Puerto Rico

not reported

not reported

  1. Proportions are reported with one decimal place (provided decimal places were given or could be calculated)
  2. Abbreviations: ADL Activities of daily living, AKPS Australia-modified Karnofsky Performance Status, CPS Cognitive Performance Scale ranging between 0 (intact cognitive performance) and 6 (very severely impaired), FAST Functional Assessment Staging Tool, GDS Global Deterioration Scale ranging from 1 (no cognitive decline) to 7 (very severe cognitive decline), LTCF Long-term care facility, MDS Minimum Data Set, NH Nursing home, NHR Nursing home residents
  3. aSample size given in this table does not represent the whole study population but only the number of nursing home residents reported
  4. bStudy design was an intervention study with a control group, the data reported in this table refers to the control group
  5. cNumbers refer to the latest year studied, which was 2007
  6. dNumbers refer to residents who died prior to the intervention