Table 3 Findings of included systematic reviews

van Haastregt, 2000 [23]

P: Elders 65+ years living in the community I: Preventive home visitation programs provided by a nurse or equivalent C: Usual care O: Health indicators

Little evidence exists in favor of the effectiveness of preventive home visits to elderly people living in the community. None of the trials reported negative effects.

Cochrane Collaboration (1997)

Study scores ranges from 20% to 71% with a mean of 54%.

No difference

Markle-Reid, 2006 [21]

P: Elders 65+ years living at home and eligible for home care services I: Preventive home visitation programs provided by a nurse or equivalent C: Usual care O: Health indicators

Home visit interventions by nurses improved health and functional status, mortality rates, and use of hospitalization and nursing homes.

Tool developed by Jadad et al., 1996. Trials fulfilled most effectiveness criteria. Shortcomings were randomization, blinding and follow-up.

Home with support

Huss, 2008 [22]

P: Elders 70+ years living in the community I: Multidimensional preventive home visitation programs

C: Usual care O: Health indicators

Multidimensional preventive home visits may reduce disability burden among older adults when based on multidimensional assessment with clinical examination.

Evaluated concealment of allocation, and blinding of staff.

50% reported adequate blinding; 29% had adequate concealment.

Home with support

Beswick, 2010, [20]

P: Elders 65 + years living at home or preparing for hospital discharge to home I: Community-based multifactorial interventions with preventive strategies and subsequent active management C: Usual care O: Indicators of independence

Overall benefit of complex community interventions in helping older people to live at home and maintain independence.

Evaluated loss to follow-up and randomization. Specifics not reported. Same effect sizes in studies of different quality.

Home with support

Mehta, 2011, [18]

P: Patients who have had surgery for a hip fracture I: Home physiotherapy C: No physiotherapy and outpatient physiotherapy O: Indicators of health, quality of life, performance-based indicators

Home physiotherapy was better than no physiotherapy care for improving patient-reported health-related quality of life.

Cochrane risk of bias tool Overall moderate to low quality evidence

No difference

Turner, 2011 [28]

P: Older people living at home I: Modification of physical hazards in the home and related components (e.g., education) C: Usual care O: Injury rate

The effect of home modification on falls was either inseparable or insignificant.

EPOC checklist Inclusion criteria limited to higher quality RCTs.

Evidence was inseparable or insignificant

Fens, 2013, [19]

P: Post-stroke older patients discharged home I: Multidisciplinary care including home assessment, assessment with follow-up care, education.

C: Usual Care O: Indicators of health and well-being

Limited evidence for the effectiveness of multidisciplinary care for stroke patients being discharged home compared to usual care.

CONSORT Studies ranged from 35% to 62%, with a mean of 50%

No difference

You, 2013 [25]

P: Community- dwelling frail elders aged 65+ years I: Independent case management interventions applied in the community C: Usual care O: Health service use

Moderate evidence that case management interventions can improve clients’ use of some community care services and delays nursing home placement, reduces nursing home admission, and shortens length of nursing home stay.

Used a checklist informed by evidence. Moderate and lower quality studies.

Home with support

Bryant-Lukosius, 2015 [24]

P: Patients receiving care (subgroup analysis of elderly) I: Transitional support from a clinical nurse specialist C: Usual care O: Health system utilization, patient health outcomes, caregiver outcomes

Evidence that clinical nurse specialist transitional support reduced re-hospitalizations and improved caregiver depression. The effects on other outcomes were less clear.

Modified Cochrane risk of bias tool GRADE Moderate risk (n-3); high risk (n = 2) Low-quality evidence where GRADE could be applied

Home with support

Reilly, 2015 [26]

P: People with dementia living in the community and their carers I: Case management (planning and coordination of care) C: Usual care O: Health and health resource indicators

Some evidence that case management is beneficial for improving some outcomes at certain time points, both in the person with dementia and in their caregiver.

Cochrane risk of bias tool Low to moderate overall risk of bias

Home with support

Liu, 2016 [27]

P: Adults 60+ years with complex needs I: Supportive care environment at home, including technology use C: Usual care O: Aging in place indicators, technology readiness

Home health monitoring technologies reduce some negative health outcomes for the elders, however, elder technological readiness is low.

PEDro scale and Sackett criteria Quality of the individual studies were unclear

Home with support

Mottram, 2002 [7]

P: Older adults in need of care services I: Enhanced long-term home-care services C: Institutional long-term care O: Health indicators

Insufficient evidence to determine whether dependent older people fare better at home with care services compared to living in institutional long-term care.

Descriptive Study rated as small and of poor methodological quality

Insufficient evidence

Gomes, 2013 [30]

P: Older adults with a life-limiting chronic disease I: Home palliative care; reinforced home palliative care C: Usual care (varied depending on local context) O: Death at home

Home palliative care increases the chance of dying at home and reduces symptom burden, particularly for patients with cancer.

EPOC checklist Inclusion criteria limited to higher quality RCTs. 6/16 RCTs were high quality; 0/4 controlled clinical trials were high quality

Home with support

Wysocki, 2015, [29]

P: Adults aged 60+ years I: Home and community based care C: Institutional care O: Function, cognition, mental health, acute care use, mortality.

No difference between most outcomes for LOC. Insufficient evidence to draw conclusions about preferred LOC.

Agency for Healthcare Research and Quality Low methodological quality

Insufficient evidence

Britton, 2000, [32]

P: Elderly participants requiring rehabilitation services after stroke I: Home rehabilitation started after acute hospital stay. C: Conventional rehabilitation O: Health and functional indicators

No statistically significant differences in outcomes between home rehabilitation and conventional care for activities of daily living, depression, quality of life, social activities, stress, satisfaction, depression, and quality of life for family members.

Quality factors (Drummond & Jefferson, 1996) Moderate to high

No difference

Toussant, 2005, [34]

P: Elderly with a sustained a hip fracture I: Home based physical therapy rehabilitation to manage surgically treated hip fractures C: Conventional rehabilitation O: Health and functional indicators

Home-based rehabilitation programs involving physical therapy are as beneficial as intensive hospital rehabilitation programs.

Sackett’s rules of evidence and grades (1989) Grade B evidence (supported by at least 1 small RCT with low risk of false positive/ negative)

No difference

Hillier, 2010 [33]

P: Community-dwelling participants within 1-year post-stroke I: Stroke rehabilitation delivered at home C: Stroke rehabilitation delivered in a center O: Independence in function

Home-based rehabilitation is superior to centre-based for functional benefits in the early period post-discharge from an inpatient setting. There is conflicting evidence that the results remain in favour of home-based long-term (6 months)

PEDro Criteria Range 7 to 9 of 11; mean = 9/11

Home rehabilitation

Mehta, 2011, [18]

As above

Home physiotherapy was similar to outpatient physiotherapy in improving patient-reported health-related quality of life. Performance-based outcomes were marginally better following outpatient physiotherapy compared with home physiotherapy 3 and 6 months after surgery. Due to the poor methodological quality of included studies, the authors concluded insufficient evidence to recommend one setting over another.

As above

Insufficient evidence

Allen, 2012, [35]

P: Patients with hip-fracture and dementia I: Home rehabilitation C: Conventional rehabilitation O: Indicators of function

Comparable functional recovery outcomes in patients with dementia recovering from hip fracture across locations compared.

Newcastle–Ottawa Quality Assessment Scale for Cohort Studies Studies scored 5, 7 of 8

No difference

Fens, 2013, [19]

P: Post-stroke older patients discharged home

I: Home rehabilitation C: Conventional rehabilitation O: Indicators of function

No differences between rehabilitation at home versus conventional settings.

CONSORT Studies ranged from 54% to 73%, with a mean of 60%

No difference

Brown, 2015 [31]

P: Older people I: Medical day hospitals C: Domiciliary care

O: Indicators of health

No difference in health outcomes between elders that receive home rehabilitation services compared to those who receive rehabilitation services at a medical day hospital.

Cochrane risk of bias tool Low to moderate overall risk of bias Overall low quality evidence

No difference