Intervention Review

You have free access to this content

Respite care for people with dementia and their carers

  1. Nicola Maayan1,
  2. Karla Soares-Weiser1,
  3. Helen Lee2,*

Editorial Group: Cochrane Dementia and Cognitive Improvement Group

Published Online: 16 JAN 2014

Assessed as up-to-date: 3 DEC 2012

DOI: 10.1002/14651858.CD004396.pub3


How to Cite

Maayan N, Soares-Weiser K, Lee H. Respite care for people with dementia and their carers. Cochrane Database of Systematic Reviews 2014, Issue 1. Art. No.: CD004396. DOI: 10.1002/14651858.CD004396.pub3.

Author Information

  1. 1

    Enhance Reviews Ltd, Wantage, UK

  2. 2

    Oxford, Oxfordshire, UK

*Helen Lee, Hidcote, Radley, Oxford, Oxfordshire, OX14 3BL, UK. helencharlottelee@yahoo.co.uk.

Publication History

  1. Publication Status: New search for studies and content updated (no change to conclusions)
  2. Published Online: 16 JAN 2014

SEARCH

 
Characteristics of included studies [ordered by study ID]
Grant 2003

MethodsRCT comparing one intervention group with a control in a parallel group design

Duration: 2 weeks


ParticipantsN=55
Country: USA

Mean age ˜73 years

62% female

Inclusion criteria: spousal caregivers of people with a diagnosis of "probable" or "possible" AD


InterventionsIntervention group: 10 days of in-home respite of up to 6 hours per day over a 2 week period (N=27)
Control group: no respite provided (N=28)


OutcomesFor the caregiver:
1. Structured Interview Guide for the Hamilton Depression and Anxiety Scales
2. Brief Symptom Inventory
3. Physiological measures


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Randomly assigned with a table of random numbers"

Allocation concealment (selection bias)Unclear riskNo information reported

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskParticipants were not blinded

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNo information reported

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskLeaving the study early not reported and Ns not reported in the results

Selective reporting (reporting bias)Low riskAll outcomes stated in the study are reported

Other biasLow riskSupported by a National Institute on Aging grant

No conflicts of interest reported

Korn 2009

MethodsRCT comparing one intervention group with an alternative treatment group in a parallel group design

Duration: 8 weeks


ParticipantsN=42

Country: USA

Mean age 50 years, range 27-69

90% female

Inclusion criteria: American Indians or Alaskan Natives who had been the primary caregiver of a family member with dementia for at least 6 months, currently providing at least 4 hr of direct assistance per day, access to a telephone, and plan to remain in the community for at least 6 months. No medical conditions that would preclude the use of polarity therapy including acute infection, deep vein thrombosis, diabetic neuropathy, current substance abuse, cardiac arrhythmia, or other conditions associated with severe disability or high risk of death.


InterventionsIntervention group: enhance respite care, eight sessions ranging from 60 to 120 minutes (N=
Control group: polarity therapy, trained practitioners administered the standard 21-point protocol to participants during eight 50-min sessions (N=

Both PT and ERC provided the same amount of time (3 hr) of paid care for the care recipient.


OutcomesFor the caregiver:
1. Perceived stress scale
2. Center for Epidemiological Studies–Depression
Scale
3. SF-36
4. Quality of Life–AD (Caregiver Version)
5. Pittsburgh Sleep Quality Index
6. Penn State Worry Questionnaire


NotesAll participants who enrolled in the study received a choice of a fresh salmon or a small gift basket (value $30.00) following their participation.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear risk"Randomization was carried out separately in each of two strata defined by baseline scores on the Perceived Stress Scale", no further details reported

Allocation concealment (selection bias)Unclear riskNo information reported

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskParticipants were not blinded

Blinding of outcome assessment (detection bias)
All outcomes
Low risk"Caregivers were instructed not to reveal which treatment they had received; a protocol deviation log was maintained by the nurse and clinical coordinator to record if blinding was broken and in no case did that occur."

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskFive people from the enhanced respite group and two form the polarity therapy left the study early due to lack of time
"All outcomes were analyzed on an intent-to-treat basis using data from all participants who could be assessed. Of 42 participants, 35 completed the entire course of PT or ERC and the posttreatment assessment, three dropouts did not complete their assigned treatment but did complete the posttreatment assessment (and were included in the outcome analysis), and four dropouts completed neither their assigned treatment nor the posttreatment assessment. Thus, the change in outcome measures from baseline until the end of the trial is based on 38 of 42 participants"

Selective reporting (reporting bias)Low riskAll outcomes stated in the study are reported

Other biasLow riskFunding from the National Institutes of Health, National Center for Complementary and Alternative Medicine (NIH-NCCAM)

No conflicts of interest reported

Lawton 1989

MethodsRCT comparing one intervention group with a control in a cluster randomised design
Duration: 12 months


ParticipantsN=632
Country: USA

Mean age: caregiver 60 years, person with dementia 76 years

79% female caregivers, 60% female with dementia

Inclusion criteria: people with "AD and related disorders" and their caregivers


InterventionsIntervention group: access to institutional respite, daycare and in-home respite over a 12 month period. The choice of which type or types of respite used was made by the caregiver
Control group: no access to respite via the programme


OutcomesFor the person with dementia:
1. Amount of formal respite used
2. Amount of informal respite used
3. Institutionalisation
4. Severity of Illness
5. Mortality

For the caregiver:
1. Caregiver wellbeing
2. Physical health - Philidelphia Geriatric Center Multilevel Assessment Instrument
3. Depression - Center for Epidemiological Studies Depression Scale (CESD)
4. Bradburn Affect Balance Scale (ABS)


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Assignment of a given participant by a random number"; "randomisation of people from support groups was accomplished by using the random number to assign each whole support group either to the E or C condition"

Allocation concealment (selection bias)Unclear riskNo information reported

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskParticipants were not blinded

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNo information reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk"Ten experimental subjects dropped out too early to serve as subjects and one control subject heard about, requested, and was given the experimental respite experience, requiring deletion from the study. Over the course of the year 19% of the experimental and 21% of the control subjects died"

Selective reporting (reporting bias)High riskData was not reported in a useable form, and the two different approaches to randomisation means that any statistics must use the support group as the unit of analysis and not the individual. It was not reported how many support groups there were and how much of the sample came from this source

Other biasLow riskSupported by a grant from the John A Hartford Foundation inc of New York and by the Pew Charitable Trusts of Philadelphia

No conflicts of interest reported

Wishart 2000

MethodsRCT comparing one intervention group with a control in a parallel group design

Duration: 6 weeks


ParticipantsN=24
Country: Canada

Mean age 58 years

86% female

Inclusion criteria: caregivers of clients with cognitive impairment referred to the Special Steps Program who were able to go on outings


InterventionsIntervention group: weekly 2-hour visit by trained volunteers taking the person with dementia out of the house on a walk or another activity (N=13)
Control group: waiting list - people in this group received the intervention 6 weeks later (N=11)


OutcomesFor the caregiver:
1. Caregiver burden - Zarit
2. Social support


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Randomisation was carried out by computer-generated random assignment to group", "grouping was blocked after every four assignments so that groups would not differ greatly in

Allocation concealment (selection bias)Low risk"Group numbers were placed in sealed opaque envelopes"

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskParticipants were not blinded

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNo information reported

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk"At 6 weeks there were three dropouts due to death or illness, two in the experimental group and one in the control group", ITT was not used

Selective reporting (reporting bias)Low riskAll outcomes stated in the study are reported

Other biasLow riskSupported by a New Horizons Grant, Partners in Aging Project, Health Cananda, and the System-Linked research Unit, McMaster University.

No conflicts of interest reported

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Beattie 2012Cross-sectional study about respite use and non-use in carers of people with dementia

Brodaty 1989Allocation to intervention was sequential by date of postal application

Burdz 1988Experimental and control groups were not assigned by the experimenters

Cameron 2011Review article about assessing and helping carers of older people

Conlin 1992Allocation to experimental and control groups was by alternation

Droes 2000Assignment to groups was not random

Engedal 1989The intervention did not fit the inclusion criteria for the review because it wasn't designed to provide temporary periods of rest or relief for the caregivers

Hedrick 1993Participants did not have dementia. Inclusion was based on those elderly people who met criteria predicting who would be admitted to a nursing home

Howe 2009Commentary on suboptimal take-up of respite care

Kosloski 1993Non-equivalent control group design

Lee 2007Prospective case series study with no control group

Lukas 2012Intervention was individual advice about available treatment options for dementia patients

Mavall 2007Observational study with no control group

Mohide 1990Respite given as one part of a multi-component caregiver support programme

Montgomery 1989The participants did not meet the inclusion criteria because only a small proportion were diagnosed with dementia

Neville 2006Observational study with no control group

Newcomer 1999Intervention was case management with community care service benefit

Quayhagen 2000Caregivers randomised to the daycare group were also enrolled in support groups

Stirling 2012Intervention was decision aids for respite service choices

Thiel 2012Not randomised

Wells 1987Compared respite care with nursing home placement

Wells 1990The experimental and control groups were not assigned by the researchers. They were made up of a group that were already receiving daycare and a group that were about to receive it

Whitebird 2011Intervention was mindfulness-based stress reduction and control group was community caregiver education and support. Participants could apply for additional financial help up to $200 to obtain respite care or travel assistance

Wimo 1993Assignment to groups was not random

Zarit 1998Assignment to groups was not random

 
Comparison 1. Respite care versus no respite care

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Caregiver Burden121Mean Difference (IV, Fixed, 95% CI)-5.51 [-12.38, 1.36]

 2 Hamilton-Depression155Mean Difference (IV, Fixed, 95% CI)-0.18 [-3.82, 3.46]

 3 Hamilton-Anxiety155Mean Difference (IV, Fixed, 95% CI)0.05 [-3.76, 3.86]

 4 Brief Symptom Inventory155Mean Difference (IV, Fixed, 95% CI)0.04 [-0.29, 0.37]

 5 Affective Support119Mean Difference (IV, Fixed, 95% CI)-0.44 [-2.85, 1.97]

 6 Confidant Support119Mean Difference (IV, Fixed, 95% CI)1.3 [-1.04, 3.64]

 
Comparison 2. Respite care versus polarity therapy

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Perceived Stress Scale138Mean Difference (IV, Fixed, 95% CI)5.80 [1.43, 10.17]

 2 Center for Epidemiological Studies - Depression Scale138Mean Difference (IV, Fixed, 95% CI)6.0 [0.31, 11.69]

 3 Penn State Worry Questionnaire138Mean Difference (IV, Fixed, 95% CI)8.1 [-3.14, 19.34]

 4 SF-36 Mental component summary138Mean Difference (IV, Fixed, 95% CI)-0.90 [-6.35, 4.55]

 5 SF-36 Physical component summary138Mean Difference (IV, Fixed, 95% CI)-4.5 [-9.69, 0.69]

 6 Pittsburgh Sleep Quality Index138Mean Difference (IV, Fixed, 95% CI)1.70 [-0.55, 3.95]

 7 Quality of Life - AD138Mean Difference (IV, Fixed, 95% CI)-1.80 [-5.74, 2.14]

 
Summary of findings for the main comparison. Respite care versus no respite care for people with dementia and their carers

Respite care versus no respite care for people with dementia and their carers

Patient or population: patients with people with dementia and their carers
Settings: outpatients
Intervention: Respite care versus no respite care

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

ControlRespite care versus no respite care

Rate of institutionalisationSee commentSee commentNot estimable_See commentNo studies reported data for this outcome

Mortality of people with dementia - not reportedSee commentSee commentNot estimable-See commentNo studies reported data for this outcome

Physical health of people with dementia - not reportedSee commentSee commentNot estimable-See commentNo studies reported data for this outcome

Quality of life of people with dementia - not reportedSee commentSee commentNot estimable-See commentNo studies reported data for this outcome

Caregiver Burden
Zarit's Caregiver Burden Scale
Follow-up: 6 weeks
The mean caregiver burden in the intervention groups was
5.51 lower
(12.38 lower to 1.36 higher)
21
(1 study)
⊕⊝⊝⊝
very low1,2,3

Caregiver psychological stress and health
Various scales
Follow-up: 2 weeks
See commentSee commentNot estimable55
(1 study)
⊕⊝⊝⊝
very low2,3,4
Grant 2003 measured this outcome on 3 scales, none of which showed a significant difference between respite care and no respite care.5

Caregiver quality of life - not reportedSee commentSee commentNot estimable-See commentNo studies reported data for this outcome

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 Risk of bias - serious: Wishart 2000 had an unclear risk of bias for blinding and incomplete data.
2 Imprecision - serious: this outcome had very wide confidence intervals.
3 Publication bias - strongly suspected: only one study reported data for this outcome.
4 Risk of bias - serious: Grant 2003 had an unclear risk of bias for allocation concealment, blinding and incomplete data.
5 The scales used were: Hamilton Depression scale, Hamilton Anxiety scale and the Brief Symptoms Inventory.
 
Summary of findings 2. Respite care versus polarity therapy for people with dementia and their carers

Respite care versus polarity therapy for people with dementia and their carers

Patient or population: people with dementia and their carers
Settings: outpatients
Intervention: Respite care versus polarity therapy

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

ControlRespite care versus polarity therapy

Rate of institutionalisation - not reportedSee commentSee commentNot estimable-See commentNo studies reported data for this outcome

Mortality of people with dementia - not reportedSee commentSee commentNot estimable-See commentNo studies reported data for this outcome

Physical health of people with dementia - not reportedSee commentSee commentNot estimable-See commentNo studies reported data for this outcome

Quality of life of people with dementia - not reportedSee commentSee commentNot estimable-See commentNo studies reported data for this outcome

Caregiver burden - not reportedSee commentSee commentNot estimable-See commentNo studies reported data for this outcome

Caregiver psychological stress and health
Various scales
Follow-up: 8 weeks
See commentSee commentNot estimable38
(1 study)
⊕⊝⊝⊝
very low1,2,3
Korn 2009 measured this outcome on 4 scales, one of which showed a significant difference favouring polarity therapy between respite care and no respite care.4

Caregiver quality of life
Quality of Life - AD (Caregiver version)
Follow-up: 8 weeks
The mean caregiver quality of life in the intervention groups was
1.8 lower
(5.74 lower to 2.14 higher)
38
(1 study)
⊕⊝⊝⊝
very low1,2,3

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 Risk of bias - serious: Korn 2009 had an unclear risk of bias for randomisation, allocation concealment, and incomplete data.
2 Imprecision - serious: this outcome had very wide confidence intervals.
3 Publication bias - strongly suspected: only one study reported data for this outcome.
4 The scales used were: Perceived Stress scale, CES Despression scale, Penn State Worry Questionnaire and SF-36 Mental component summary.
 
Table 1. EPICOT+ research recommendations

Element to consider in future researchImplications and suggestions for future research arising from Cochrane review

EvidenceCurrent evidence does not allow one to make any reliable conclusions about the efficacy of respite care for people with dementia and their caregivers. This reflects a lack of high quality research in this difficult area

Population
  1. People of any age with dementia of any type
  2. Full-time carers of people with dementia

InterventionRespite care, i.e. a service or group of services designed to provide temporary periods of relief and/or rest for caregivers

ComparisonAn alternative intervention, waiting list or no respite care

OutcomesFor people with dementia - rate of institutionalisation, mortality, physical health, quality of life; for caregivers – caregiver burden, psychological stress and health, and quality of life

Time stampNovember 2013

Disease burdenDementia is a common and serious mental health problem affecting 6.4% of the population, and increasing in prevalence with age, from 0.8% in 65 to 69 year olds to 28.5% in people aged 90 years or older. In the coming years an exponential increase in numbers of people affected is anticipated as populations age (Lobo 2000). Providing care for a person with dementia in the community commonly places stress on the primary caregiver. Such stress can have a range of adverse effects including the breakdown of the relationship between patient and caregiver, a poorer quality of care, and physical and psychological morbidity for both patient and caregiver (Neufield 2003; Parks 2000).

TimelinessMean age of population: over 65 years
Duration of intervention: minimum 1 month
Length of follow-up: minimum 3 months. Dementia is a chronic condition, and most studies in the review were between two and six weeks long, and showed no difference between groups, which could be due to the short duration

Study typeRandomised controlled trial