Table 2 Case management

Gagnon (1999)

Quebec, Canada RCT 10 months

≥70 years who had visited an emergency department in the previous year

n = 427

(58.1% F)

= 81.6

Participants were assigned nurse case managers who operated using the Promotion of Autonomy Framework. Case managers created and implemented a care plan and coordinated the work of all healthcare and service providers involved in care. There were approximately 28 recorded telephone contacts and 36 home visits per person.

For controls, hospital and community services were provided separately.

Over 10-months, participants in the intervention group were readmitted to the emergency department significantly more frequently than controls (p = 0.041, d = 0.2).

No significant differences were found between the two groups on quality of life, satisfaction with care, functional status, admission to hospital, and length of hospital stay from baseline to 10 months.

14

Vickrey (2006)

Dementia care quality intervention trial

California, USA

Cluster RCT

1 year

≥65 years with dementia receiving Medicare with an informal caregiver

n = 408 (54.9% F)

= 80.1 ± 6.6

Case managers trained to use care management software and provided with a care plan manual conducted assessments and 6 monthly reassessments of participants, designed and implemented care plans in collaboration with caregivers, taught skills and provided ongoing follow-up. Seminars were held for primary care providers at participating health care organisations.

Controls received usual care.

After 18 months the proportion of guidelines adhered to was significnatly higher in intervention (64%) compared to controls (24%; p ≤ 0.001). Intervention participants had higher rates of receiving information or services from ≥1 community agency (RR = 1.5, 95% CI 1.0-1.9), respite care (p ≤ 0.03), home health aide services (p ≤ 0.03), professional carer services (p ≤ 0.03), enrollment in a wandering program (p = 0.001), cholinesterase inhibitor use (p = 0.032), health related quality of life (p = 0.034) and health care quality (p ≤ 0.011). Intervention caregivers had higher confidence in caring (p ≤ 0.01), caregiving mastery (p ≤ 0.01), social support (p = 0.029) and met needs for problem behaviours (p ≤ 0.012).

There were no differences in caregiver health related quality of life.

14

Alkema (2007), Shannon (2006)

California, USA

RCT

1 year

>65 years, enrolled in Medicare health plan, rated as being at risk of future healthcare service use

n = 781, 823

(65.3% F)

= 83.3

A care manager (care advocate) operating via telephone evaluated needs, made referrals to additional services and called monthly to moitor progress, offer support and coaching, provide additional information and assistance and follow-up to ensure linkages were establised.

Controls received usual care including Medicare managed care.

After the 12-month intervention, the case managed care group had lower mortality than controls (OR = 0.45; p = 0.006). However, at 24-month follow-up, mortality differences between the groups were not significant (p = 0.198).

After 12 months participants were more likely to use primary care physicians (OR = 2.05, 95% CI 1.28-3.28), were less likely to be admitted to hospital (OR = 0.43, 95% CI 0.22-0.84) and had fewer hospital days (OR = 0.39, 95% CI 0.17-0.86) compared to controls.

There were no differences between groups on emergency department and specialist use.

13

Bernabei (1998)

Rovereto, Italy

RCT

1 year

≥65 years

n = 199

(70.9% F)

= 81.0 ± 7.3

Participants received case management and care planning from a community geriatric evaluation unit and general practitioners. Case managers conducted assessments every 2 months, monitored the provision of services, provided extra help as requested and were available to deal with problems. Controls received usual care including non-case managed community services

Over 1 year the intervention group improved on function (ADLs, p < 0.001, d = 6; IADLs, p < 0.05, d = 3) and depression (p < 0.05, d = 4) and declined less on cognition (p < 0.05, d = 4), compared to the control group.

Over 1 year, the intervention group had lower risk of admission to a nursing home (HR = 0.81, 95% CI: 0.57-1.16), acute hospital (HR = 0.74, 95% CI: 0.56-0.97), or emergency (HR = 0.64, 95% CI: 0.48-0.85) compared controls.

There were no differences in 1 year mortality.

12

Shapiro (2002)

USA

RCT

1 year 6 months

≥60 years on waiting list to receive social services

n = 105

(85.7% F)

= 77.2

Individualized care plans were developed by a geriatric nurse together with participants and caregivers after a thorough in-home geriatric assessment. Case managers coordinated the delivery of services which were prescribed and changed to address specific needs and problems.

Controls received usual care.

After 18 months, participants in the intervention group were less likely to be institutionalized or die than those in the comparison group (combined as a single endpoint, OR = 0.18, p = 0.029). The intervention group had improved on Satisfaction with Social Relationships (F = 2.59, p < 0.05, d = 0.45), Environmental Mastery (F = 3.71, p < 0.01, d = 0.54), and Life Satisfaction (F = 3.18, p < 0.05, d = 0.53). No statistically significant difference was found for depression.

12

Eloniemi-Sulkava (2001)

Finland

RCT

2 years

≥65 years with dementia and caregivers

n = 100

(53.0% F)

= 79.4

A nurse case manager with access to a physician provided advocacy, round the clock comprehensive support, continuous and systematic counseling, annual training courses, follow-up calls, in-home visits and assistance with arrangements for social and healthcare services. The frequency of contacts varied from 5 times a day to once a month.

Controls received usual care.

During the first 6 months, the rate of institutionalization was significantly lower in the intervention group than in controls (HR = 0.12, 95% CI: 0.02-0.93) but this benefit decreased over time (HR = 1.18, 95% CI: 1.02-1.36). The estimated probability of staying in community care for 6, 12, and 24 months was 0.98, 0.92, and 0.63 in the intervention group and 0.91, 0.81, and 0.68 in the control group, respectively.

12

Miller (1999), Newcomer (1999a, 1999b), Shelton (2001)

Medicare Alzheimer's Disease Demonstration (MADDE)

USA

RCT

3 years

Persons diagnosed with dementia enrolled in Medicare A and B

n = variable (see results)

For Newcomer 1999a:

40% F

= 79 ± 8 years

MADDE participants received case management (with a ratio 1:30 for Model A or 1:100 for Model B) and 80% subsidy of service costs (up to $489 for Model A or $799 for model B).

Controls received usual care.

After 1 year there was increased use of any home care service (OR = 2.77, 95% CI 2.40 - 3.0) and adult day care (OR = 2.23, 95% CI 1.92-2.60) [n = 5209]

Over 3 years there were no differences on nursing home entry rates (n = 8095).

After 3 years there were no differences in the change in caregiver burden or depression. [n = 5307].

Over 3 years caregivers in MADDE had a lower likelihood than controls of any hospitalization (OR = 0.58, 95% CI 0.35-0.97), but not of emergency department use, length of hospital stay or number of hospitalizations (n = 412).

There were no differences between Model A and Model B in any of the outcomes tested.

12

Kinney (2003)

USA

RCT

2 years

Enrollees of Indiana's state case management program and/or the Medicaid home and community-based services waiver program for the aged (≥65 years) and disabled n = 1006

(77.5% F)

= 67.7

The intervention involved two computer-assisted methods for individualized care planning. The Normative Treatment Planning (NTP) program assessed needs and prescribed services using a standard set of algorithms. The Client Feedback System (CFS) program provided systematic feedback on participant satisfaction to service providers. Participants were randomly assigned to receive none, one or two of the interventions.

The control group had case managers who prepared a non-computerized care plan.

Over 2 years perception of needs met (p < 0.05, d = 0.027) and service satisfaction (p < .05, d = 0.027) improved in the NTP compared to the control group. The CFS group had significantly higher satisfaction than the control group (p < 0.05, d = 2.7) but not greater perception of needs met. There were no statistically significant differences in perception of needs met and satisfaction between the group that used both NTP and CFS and the control group.

11.5

Marek (2006)

USA

NRCT

1 year

≥64 years

n = 85 (80% F)

= 77.1

Participants received nurse care coordination in addition to a local care program, Missouri Care Options (MCO), which included basic and advanced personal care, nurse visits, homemaker care, and respite care. Care coordinators conducted a comprehensive admission assessment, created a care plan and coordinated health and social services.

Controls were recruited from a similar neighborhood and received the basic MCO program with limited nursing visits and were more likely to be white.

After 12 months, the intervention group improved significantly more than the control group on pain (OASIS M00420; <0.01), dyspnea (p = 0.03), and function (p = 0.01). No significant differences were found over time between groups in emotional stability, medication management, cognition and incontinence.

9

Morales-Asencio (2008)

Spain

NRCT

6 months

Homebound persons requiring assistance for daily activities

n = 258

= 76.3

A case manager made home visits, conducted assessments, established links with and coordinated other health institutions and professionals, arranged technical assistance at home, provided education telecare for the participants and education and support for caregivers.

The control group received visits according to their health demands and at baseline had fewer functional limitations than participants.

The intervention group had significantly lower scores on activities of daily living function and family function compared to the control group at baseline (p = 0.021; p = 0.023 respectively). These differences no longer occurred at six months (p = 0.222; p = 0.142). Cognitive status and instrumental activities of daily living were lower in the intervention than the controls at both baseline (p = 0.042; p = 0.008) and 6 months (p = 0.008; p = 0.007).

8

Gravelle (2007)

Evercare, England

Longitudinal observational study

1 year 9 months

≥65 years

n = ~7000 Evercare practices

(mean age not reported)

Participants were monitored by advanced practice nurses who developed individualized care plans with the participant, general practitioner and other staff.

Control data were gathered from all non-Evercare practices in England. At baseline, intervention practices had significantly higher rates of admission and use of emergency bed days and faster growth rates in admissions for the general population aged ≥65.

Over 21 months, the intervention had no significant effect on rates of emergency admission, emergency bed days, and mortality for the whole Evercare sample or a high risk subsample with a history of two or more emergency admissions in the preceding 13 months in comparison to the control group.

7

Onder (2007)

Aged in Home Care Project (AdHoC), Europe

Retrospective observational study

1 year

≥65 years and receiving home care services

n = 3292

(73.6% F)

= 82.3 ± 7.3

The case management group comprised participants living in Finland, Iceland, Italy, Sweden & the UK. Participants in these countries had case managers who conducted assessments, dealt with problems that arose, monitored the provision of services, worked with geriatric evaluation units to design and implement individualized care plans and who provided additional services as needed.

Controls lived in Czech Republic, Denmark, France, Germany, the Netherlands and Norway where case management was not available. Controls were more likely to be women, live alone, be physically active, had more severe cognitive impairment and a lower prevalence of daily pain and a number of chronic diseases and unexpected weight loss than participants.

During the 1-year follow-up, the risk of nursing home admission was significantly lower in the case management group compared to controls (OR = 0.56, 95% CI: 0.43-0.63).

7

Bierlein (2006)

Canada

Longitudinal observational

6 months

>65 years, 22% were cognitively impaired

n = 179 (65% F)

= 80 ± 7.38

Participants were assigned case coordinators and had access to various community health services.

There was no control group.

After 6 months, participants' scores improved on the physical (p < 0.001, d = 0.4) and mental health subscales (p < 0.001, d = 0.4) of the SF-8. Risk of institutionalization decreased significantly (p < 0.03, d = 0.1). However there was a statistically significant deterioration on social interaction (p < 0.04, d = 0.2) and instrumental support (p < 0.001, d = 0.3). Subjective support scores (p = 0.88) and cognitive scores (p = 0.68) did not change significantly.

7

Onder (2008)

AdHOC

Retrospective observational

≥65 years already receiving home care services

n = 4007

(74.1% F)

= 82.3 ± 7.3

See Onder, 2007 above

Compared to the control group, more participants in the case management group had blood pressure measured in previous 2 years (OR = 1.31, 95% CI 1.08-1.59), received influenza vaccination in the last 2 years (OR = 2.08, 95% CI: 1.81-2.39) and had medication reviewed in the last 6 months (OR = 1.69, 95% CI: 1.42-2.01).

Compared to the control group, caregivers of participants in the intervention group were more likely to be able to continue in caring activities (OR = 0.49, 95% CI: 0.35-0.69) and were less dissatisfied (OR = 0.47, 95% CI: 0.29-0.73). There was no significant difference for caregiver distress (OR = 1.04, 95% CI: 0.78-1.38).

6.5