Background
Methods
Search of the literature and study inclusion
Data extraction
Statistical analysis
Results
Identification of studies
Data source characteristics
Data Source | Sampling Frame | Sample Size | Data Collection | Duration of Follow-Up | Number of Nursing Home Admissions/Rate | Predictor Variables Available |
---|---|---|---|---|---|---|
Study of Assets and Health Dynamics of the Oldest Old (AHEAD) | National panel study, 70 years of age and over in U.S. | Up to 8,222 | Baseline face-to-face interviews, telephone follow-ups every other year with respondent or proxy | Up to 7 years (1993–2000) | Up to 1135/ | Cognitive performance, physical and functional health, economic status, family structure, demographics, service use, caregiving, financial resources |
Established Populations for Epidemiological Studies of the Elderly-East Boston (EPESE-EB) | All non-institutionalized persons 65 years age and over in East Boston, Massachusetts enumerated via community census | 3,545 | In-person and telephone interview with respondent and/or proxy | 1982–1985 | 142 | Demographic characteristics, social and physical functioning, chronic Conditions, related health problems, self-reported Service use, depression |
Established Populations for Epidemiological Studies of the Elderly-Iowa (EPESE-IOWA) | All non-institutionalized persons 65 years age and over in Washington and Iowa counties, enumerated via all Area Agencies on Aging and local informants | Up to 3,097 | " | Up to 4 years (1982–1986) | Up to 369 | " |
Established Populations for Epidemiological Studies of the Elderly-New Haven (EPESE-NH) | All non-institutionalized persons 65 years age and over in New Haven, Connecticut enumerated via stratified random sample of household clusters | Up to 2,812 | " | Up to 9 years (1982–1991) | Up to 935 | " |
Government Accounting Organization-Cleveland Study (GAO) | Older adults, 65 years and age and over and living independently, were randomly selected in Cleveland, OH | 1,598 | In-person interviews with respondent or proxy | 1975–1984 | 406 | Sociodemographics, mental health, physical health, disability, available informal support |
Longitudinal Study on Aging (LSOA) | All civilian non-institutionalized persons in the U.S. 70 years of age and over in 1984 | Up to 7,541 | In-person interviews at baseline, telephone interviews at each 2-year follow-up with respondent or proxy | Up to 6 years (1984–1990) | Up to 1100 | Cognitive performance, physical and functional health, economic status, family structure, demographics, service use, caregiving, financial resources |
National Long-Term Care Survey (NLTCS) | Random sample from Medicare enrollment files, 65 years of age and over in U.S. with any ADL or IADL dependency | Up to 5,851 | In-person interviews with older respondent and/or proxy | 1982–1984 | Up to 911 | Cognitive performance, physical and functional health, economic status, family structure, demographics, service use, caregiving, financial resources |
Medicare Survey | People living in the community during the 1977 Current Medicare Survey | 4400 | Interviews with respondents initially and one year later | 1977–1978 | 127 | Sociodemographics, payment source, functional status, perceived health status, service use and physician contact |
Massachusetts Elder Health Project | A geographically stratified random sample of disabled elders living in Eastern Massachusetts | Up to 586 | In-person and telephone interviews with respondent or proxy | Up to 6 years (1984/85–1991) | Up to 143 | Informal and formal community care, caregiving burden, sociodemographics, physical disability |
Massachusetts Health Care Panel Study (MHCPS) | Statewide (Massachusetts) probability sample of non-institutionalized adults 65 years of age and over | 1625 | In-person interviews at baseline and first follow-up a year later; mailed surveys every 4 years thereafter | Up to 10 years (1974/75–1984/85) | Up to 148 | Demographics, social characteristics, functional disabilities, attitudinal variables |
Medicare Health Outcomes Study (MHOS) | Random sample of Medicare+Choice (Medicare managed care) enrollees over the age of 64 who were not institutionalized | 137,632 | Mail survey and telephone follow-up of non-respondents on an annual basis | 1998–2000 | 11,220 | Quality of life/psychosocial status, self-reported symptoms and diagnoses, functional disability, sociodemographics |
Monogohela Valley Independent Elders Survey (MoVIES) | Participants 65 years of age and over, with 6th grade or greater education, and living in the community were selected from voter registration lists in 23 communities in southwestern Pennsylvania | 1147 | Clinical assessments and in-person interviews in respondents' homes annually | 1987–2001 | 156 | Sociodemographics, functional and cognitive impairment, medication usage, depression |
Predictors of nursing home admission
Predictor | Data Sources Included (% of Participants Who Entered Nursing Home) | Pooled Odds Ratio (95% Confidence Interval) |
PValue for Heterogeneity |
---|---|---|---|
ADLs | LSOA(14.6) [10], EPESE-IOWA(11.9) [34], NLTCS (14.8) [24] | 1.11 (1.07–1.16) | .73 |
1–2 ADLs (versus 0 ADLs) | EPESE-EB(4.0) [21], EPESE-NH(9.0) [21], EPESE-IOWA(12.0) [21], NLTCS (14.8) [15], LSOA(n/a) [11] | 2.45 (2.02–2.97) | .87 |
ADLs >= 1 | MHCPS(9.1) [33], GAO(25.4) [35] | 1.88 (.86–4.08) | .01 |
ADLs >= 3 | EPESE-EB(4.0) [21], EPESE-NH(9.0) [21], EPESE-IOWA(12.0) [21], NLTCS(14.8) [15], LSOA(n/a) [11] | 3.25 (2.59–4.09) | .30 |
Age | MHCPS(9.1) [36], LSOA(14.6) [10], EPESE-EB(4.0) [21], EPESE-NH(9.0) [21], EPESE-IOWA(12.0) [21], MEHCP(24.4) [26], NLTCS(8.0) [23], AHEAD(9.7) [28], GAO(25.4) [35] | 1.11 (1.08–1.14) | .00 |
Age squared | MS(2.9) [32], EPESE-IOWA(11.9) [34], AHEAD [28] | 1.00 (1.00–1.01) | .00 |
Available informal caregiver | LSOA(14.6) [10], NLTCS(8.0) [23], GAO(25.4) [35] | 1.23 (1.04–1.46) | .57 |
Cognitive impairment | NLTCS(16.9) [13], MEHCP(24.4) [26], MHCPS(9.1) [37] | 2.54 (1.44–4.51) | .04 |
Education | MHCPS(9.1) [36], EPESE-IOWA(11.9) [34], GAO(25.4) [35] | 1.03 (.97–1.08) | .20 |
Female | MHCPS(9.1) [36], LSOA(14.6) [10], EPESE-EB(4.0) [21], EPESE-NH(9.0) [21], EPESE-IOWA(12.0) [21], NLTCS(14.8) [15], AHEAD(9.7) [28], GAO(25.4) [35] | .97 (.87–1.08) | .02 |
Formal help | LSOA(14.6) [10], NLTCS(14.8) [15] | 1.23 (.93–1.62) | .47 |
Subjective health | MHCPS(9.1) [36], EPESE-IOWA(11.9) [34], LSOA(3.9) [27] | .90 (.58–1.39) | .00 |
Homeowner | LSOA(14.6) [10], NLTCS(16.9) [13] | .82 (.71–.95) | .21 |
Prior hospitalization | LSOA(14.6) [10], NLTCS(16.9) [13], EPESE-IOWA(11.9) [34] | 1.19 (1.07–1.33) | .67 |
IADLs | LSOA(14.6) [10], NLTCS(8.0) [23] | .98 (.71–1.36) | .00 |
Income | NLTCS [11], EPESE-IOWA(11.9) [34], AHEAD(9.7) [28], GAO(25.4) [35] | .95 (.83–1.08) | .00 |
Annual income > $5,000 (vs. $5,000–$10,000) | EPESE-EB(4.0) [21], EPESE-NH(9.0) [21], EPESE-IOWA(12.0) [21] | 1.02 (.78–1.33) | .82 |
Annual income < $5,000 (vs. $5,000–$10,000) | EPESE-EB(4.0) [21], EPESE-NH(9.0) [21], EPESE-IOWA(12.0) [21] | 1.45 (1.15–1.83) | .70 |
Annual income missing (vs. $5,000–$10,000) | EPESE-EB(4.0) [21], EPESE-NH(9.0) [21], EPESE-IOWA(12.0) [21] | 1.72 (1.23–2.39) | .16 |
Lives alone | MHCPS(9.1) [36], EPESE-EB [21], EPESE-NH [21], EPESE-IOWA [21], NLTCS(15.7) [14], LSOA(3.9) [27], GAO(25.4) [35] | 1.91 (1.55–2.35) | .02 |
Lives with others, no spouse | EPESE-EB(4.0) [21], EPESE-NH(9.0) [21], EPESE-IOWA(12.0) [21] | 1.54 (.91–2.63) | .12 |
Married | LSOA(14.6) [10], EPESE-IOWA(11.9) [34], AHEAD(9.7) [28] | .63 (.41–.95) | .01 |
Medicaid-eligible | MHCPS(9.1) [36], MS(2.9) [32], LSOA(14.6) [10], NLTCS(14.8) [15] | 1.21 (.89–1.65) | .00 |
Number of children | LSOA(14.6) [10], NLTCS(14.8) [24] | .88 (.80–.97) | .00 |
Prior nursing home use | EPESE-EB(4.0) [21], EPESE-NH(9.0) [21], EPESE-IOWA(12.0) [21], MHCPS(2.9) [33], NLTCS(14.8) [15] | 3.47 (1.89–6.37) | .00 |
SPMSQ >= 4 errors | EPESE-EB(4.0) [21], EPESE-NH(9.0) [21], EPESE-IOWA(12.0) [21] | 2.33 (1.80–3.00) | .48 |
SPMSQ missing | EPESE-EB(4.0) [21], EPESE-NH(9.0) [21], EPESE-IOWA(12.0) [21] | 1.94 (.68–5.51) | .01 |
White (vs. all other races/ethnicities) | LSOA(14.6) [10], NLTCS(15.7) [14], AHEAD(9.7) [28] | 1.61 (1.22–2.11) | .04 |
Predictor | Data Sources Included (% of Participants Who Entered Nursing Home) | Pooled Hazards Ratio (95% Confidence Interval) |
PValue for Heterogeneity* |
---|---|---|---|
Age | LSOA(14.6a) [12], MHOS(9.2) [31], MoVIES(13.6) [38], NLTCS(7.6b) [22] | 1.07 (1.03–1.11) | .00 |
Arthritis | AHEAD(17.0) [29], MHOS(9.2) [31] | .86 (.58–1.29) | .00 |
Blood pressure | AHEAD(17.0) [29], MHOS(9.2) [31] | 1.04 (1.01–1.07) | .60 |
Cancer | AHEAD(17.0) [29], MHOS(9.2) [31] | 1.15 (1.11–1.19) | .68 |
Cardiovascular disease | AHEAD(17.0) [29], MHOS(9.2) [31] | 1.02 (.88–1.17) | .32 |
Diabetes | AHEAD(17.0) [29], MHOS(9.2) [31] | 1.35 (1.15–1.57) | .11 |
Falls | AHEAD(17.0) [29], MHOS(9.2) [31] | 1.16 (1.02–1.30) | .28 |
Female | AHEAD(17.0) [29], EPESE-NH(33.3) [39], LSOA(14.6) [12], MHOS(8.2) [31], MoVIES(13.6) [38], NLTCS(7.6) [22] | .87 (.81–.93) | .18 |
Homeowner | LSOA(14.6) [12], MHOS(8.2) [31] | .88 (.85–.92) | .61 |
Incontinence | AHEAD(17.0) [29], EPESE-NH(33.3) [39] | 1.05 (.95–1.17) | .34 |
Living with others | LSOA(14.6) [12], MoVIES(13.6) [38] | 1.23 (.93–1.62) | .89 |
Lung disease | AHEAD(17.0) [29], MHOS(8.2) [31], NLTCS(7.6) [22] | 1.01 (.70–1.46) | .00 |
Married | MHOS(8.2) [31], MoVIES(13.6) [38] | .90 (.87–.92) | .46 |
Number prescription drugs | EPESE-NH(33.3) [39], MoVIES(13.6) [38] | 1.12 (.96–1.29) | .002 |
Spouse present | AHEAD(17.0) [29], EPESE-NH(33.3) [39] | .66 (.56–.77) | .24 |
Stroke | AHEAD(17.0) [29], EPESE-NH(33.3) [39] | 1.24 (1.04–1.49) | .05 |
White (vs. all other races/ethnicities) | EPESE-NH(33.3) [39], MHOS(8.2) [31], NLTCS(14.0) [25] | 1.76 (1.35–2.30) | .00 |