R E S E A R C H A R T I C L E
Effectiveness of an in
‐
home respite care program to support
informal dementia caregivers: A comparative study
Sophie Vandepitte
1|
Koen Putman
2|
Nele Van Den Noortgate
3|
Sofie Verhaeghe
1|
Lieven Annemans
11
Faculty of Medicine and Health Sciences, Department of Public Health and Primary Care, Ghent University, Ghent, Belgium
2
Faculty of Medicine and Pharmacy, Department of Public Health, Vrije Universiteit Brussel, Brussels, Belgium
3
Faculty of Medicine and Health Sciences, Department of Internal Medicine, Ghent University, Ghent, Belgium
Correspondence
Drs Vandepitte Sophie, Department of Public Health and Primary Care, Ghent University, Corneel Heymanslaan 10, 9000 Ghent, Belgium.
Email: [email protected]
Funding information
Baillet‐Latour Fund, Grant/Award Number: E/01447/15
Objectives:
Dementia is a major public health problem with important physical,
psychosocial, emotional, and financial consequences for patients, their caregivers,
and society. Since patients prefer to be managed at home, extensive research has
been conducted into effectiveness of psychosocial interventions to support informal
caregivers. The aim of this study was to assess the effectiveness of an in
‐
home
respite care program.
Methods:
In a prospective quasi
‐
experimental study, 99 dyads who received an in
‐
home respite care program were compared at 6 months post
‐
baseline, with 99
matched dyads receiving standard dementia care. Additionally, the short
‐
term effect
of the program was evaluated 14 to 15 days post
‐
intervention. The primary outcome
was caregiver burden. The secondary outcomes were: desire to institutionalize the
patient, caregiver quality of life, and frequency and impact of behavioral problems.
Mixed model analyses were performed to evaluate the impact of the intervention.
Results:
After 6 months, no significant difference on caregiver burden was
observed, but intervention group caregivers had a significant lower desire to
institu-tionalize the patient compared with control group caregivers (adj.diff =
−
0.51;
p = .02). Shortly after the program, intervention group caregivers also had a
signifi-cant lower role strain (adj.diff = 0.75; p = .05), and a lower burden on social and family
life (adj.diff = 0.55; p = .05) compared with baseline.
Conclusions:
This study was the first comparative study to investigate
effective-ness of an in
‐
home respite care program to support informal caregivers of persons
with dementia. The results partly confirm earlier positive findings from explorative
studies.
K E Y W O R D S
dementia, effectiveness, informal care, respite care
1
|I N T R O D U C T I O N
Dementia is a major public health issue with important physical, psychosocial, emotional, and financial consequences for patients,
caregivers, and society.1 In 2016, Alzheimer's Disease International estimated the worldwide economic cost of dementia to be 818 billion dollars and expected an increase towards a trillion dollars by 2018.2 Extensive research has been conducted into (cost)‐effectiveness of pharmacological interventions in dementia treatment.3Because a cure is not available in the near future and patients prefer to be managed at Trail registration: clinicaltrials.gov, ID: NCT02630446
DOI: 10.1002/gps.5164
home,4,5the focus has shifted towards effectiveness (and to a lesser extent cost‐effectiveness) of psychosocial interventions to support informal caregivers. Indeed, informal caregivers play a crucial role in the care and support of community‐dwelling patients with dementia.6As a result of this demanding and often highly burden-some and stressful task, they are at risk of developing several health problems.7
A recently conducted systematic review of comparative studies investigated the effectiveness of psychosocial support interventions on the informal caregiver and the care‐recipient. Psychosocial support was divided into the following four main categories: psychoeducation, cognitive behavioral therapy, occupational therapy, and respite care.5 It was concluded that psychoeducational support, cognitive behavioral therapy, and occupational therapy were effective strategies in improv-ing well beimprov-ing of both caregiver and recipient, and resulted in addi-tional societal benefits. However, research into the comparative effectiveness of respite care, and especially in‐home respite care— which is defined as a supportive service to give the informal caregiver a temporary relieve or break from caregiving duties—was rare because of methodological challenges.5,8-10Therefore, an additional review,11 which also allowed recent observational designs without control group, was conducted to investigate the effectiveness of different types of respite care. In this review, it was concluded that day care services were effective in decreasing caregiver burden and behavioral problems in people with dementia, but also accelerated time to nursing home placement. Results of temporary residential admission were rather mixed and showed unexpected adverse effects on both caregivers and care‐recipients (such as reduced sleep quality in care‐recipients while being admitted, and increased burden and distress in caregivers shortly after the respite period). Evidence on community‐based respite care such as in‐home respite care and host family respite care remained rare. Only one quantitative in‐home respite care intervention was identified indicating some ben-efits for caregivers.12
Nevertheless, qualitative research into the potential impact of in‐ home respite care was promising.11,13,14 The provision of respite responded to an unmet need of many caregivers to take a break.4,15 Caregivers were in general highly satisfied with the provided respite care.15-19They described benefits such as lower perceived burden,20 being able to get things done,14 and the possibility to escape from their duties enabled them to continue their demanding role as care-giver.16,17Additionally, some caregivers pointed out their preference for in‐home respite care, especially because their loved ones could remain in their trusted environment.14,16
Because explorative research suggests that in‐home respite ser-vices can actually be important for caregivers but the level of evidence should be improved, we have designed the (to our knowledge) first comparative study to assess the effectiveness of a specific 24‐hour/ 24‐hour in‐home respite care program. On the basis of the current evidence, our hypothesis is that in‐home respite care can potentially reduce caregiver burden, as well as the caregiver's intention to institu-tionalize the person with dementia, and that it can help manage behavioral problems.
2
|M E T H O D S
Detailed information about the study protocol can be consulted in a previous publication.21 Ethical approval (B670201526906) was obtained, and the trial was registered on clinicaltrials.gov, ID: NCT02630446.
2.1
|Design
A quasi‐experimental study was designed to investigate the effective-ness of an in‐home respite care program on caregivers of persons with dementia. A quasi‐experimantel design was chosen for practical and ethical reasons (see further) and because this design was earlier iden-tified as a favorable and valid strategy to conduct research into respite care efficacy.8 Comparison between both study arms was done by collecting health‐related and economic data.
2.2
|Participants
To be eligible for the study, the caregivers had to provide informal care. Hence, professional healthcare workers in the caregiving role under investigation were not recruited. The caregivers had to identify themselves as the main person responsible for the informal care (pri-mary caregiver) and had to fluently understand Dutch or French. Finally, caregivers with severe cognitive impairment or psychiatric comorbidity were excluded. The care‐recipient had to be formally diagnosed with dementia and to reside in the community.
Key points
• Research into the comparative effectiveness of in‐home respite care—a supportive service to temporary relieve the informal caregivers from their care responsibilities— is extremely rare. Therefore, a quasi‐experimental study was designed to assess the effectiveness of an in‐home respite care program to support informal caregivers of persons with dementia.
• Shortly after the program, intervention group caregivers demonstrated a significant lower role strain and a lower burden on social and family life compared with baseline. After 6 months, there was no significant difference in caregiver burden, but caregivers in the intervention group had a significant lower desire to institutionalize the patient.
• This study is the first comparative study to investigate effectiveness of an in‐home respite care program to support informal caregivers of persons with dementia. The results partly confirm earlier positive findings from explorative studies.
2.3
|Recruitment
Informal caregiver and dementia patient dyads for the intervention arm, ie, receiving the in‐home respite care program (intervention) were recruited via an independent supportive healthcare service at home called Baluchon (see further). Informal caregiver and dementia patient dyads for the control arm were broadly recruited across Belgium, ie, from six different memory clinics, a geriatric daycare clinic, several Alzheimer cafés (monthly meeting places for peers), local info points, and expert centers for dementia, the Flemish Alzheimer's league, a sickness fund, and 16 general practitioners. Recruitment started in January 2016 and was completed in December 2017.
2.4
|Setting
In each setting, eligible participants were identified by the health care professionals who then described the study, and asked informal care-givers if they were willing to participate. After their verbal consent, contact information was forwarded to the research team who then contacted them by phone. During this call, a home visit was planned to sign the informed consent and complete the baseline assessment.
2.5
|Intervention
The intervention consisted of a 24‐hour/24‐hour in‐home respite care program (called Baluchonnage), alongside standard dementia care, and was provided by an organization named “Baluchon Alzheimer Belgium.”In this respite program, caregivers could be relieved from their caregivers's task for at least 5 days while a trained employee took their place (5 d were offered for free). For the patient, all daily habits/activities and resource use remained unchanged.
The program also included caregiver support by keeping a support diary. The latter can be seen as a communication tool between the caregiver and the Baluchonneuse. In this diary, the trained employee noted all daily experiences and strategies to better deal with the diffi-cult behaviors the caregivers listed before. In this way, caregivers could validate their perceptions, learn to better cope with difficult behaviors, and felt understood by somebody. Also, after the the respite period, the trained employee remained available for further counsel indefinitely.
2.6
|Treatment as usual
The control group received usual dementia care (including medical, psychological, and other health and social services) and other support-ive initiatsupport-ives (such as support groups). Only dyads who already benefitted from the in‐home respite service offered by Baluchon and those stating at baseline not willing to ever consider this type of sup-port at baseline were excluded.
2.7
|Outcome measures
Assessments were carried out in both groups at baseline during a home visit. The follow‐up assessment at 6 months was made over the phone. In the intervention group, an extra assessment was carried out by phone 14 to 15 days after the last intervention day to evaluate the short‐term impact.
In both study arms, the baseline data were collected after inclu-sion. During this baseline assessment, background characteristics (sociodemographics, clinical data, and resource use of the patient) were collected, as well as the baseline values on the various outcomes (see below).
The primary outcome was caregiver self‐perceived burden col-lected with the Zarit Burden Interview (ZBI). This 22‐item validated self‐report questionnaire uses a 5‐point Likert scale, ranging from “never”to“nearly always present.”The final score ranges between 0 (no burden) to 88 (high burden).22
Secondary outcomes were:
• Desire of the caregiver to institutionalize the patient, measured with a modified version of the Desire‐to‐Institutionalize scale,23 which is known as a reliable predictor for future institutionaliza-tion.24,25 The scale contains six“yes/no” questions. The higher the total score (maximum score = 6), the greater the desire. • Caregiver health–related quality of life was collected with the EQ‐
5D‐5 L, a standardized non‐disease specific value‐based instrument on the self‐perceived health status.26 The instrument consists of five health‐related domains (mobility, self‐care, daily activities, pain, and depression/anxiety), with each domain having five answer cat-egories, resulting in a 5‐digit health profile. The profile was trans-lated into a utility value based on a published algorithm.27 • Frequency of behavioral problems in the dementia patients and
impact of those behaviors on the caregiver was measured with the Revised Memory and Behavior Problems Checklist (RMBPC).28 This validated caregiver self‐report measure contains 24 items, including three domains (depression, memory‐related problems, and disruptive behaviors) comprising two scales. One scale mea-sures, the frequency of problem behaviors of the person with dementia (ranging fromnever occurstooccurs daily or more often), and the other measures, the reactions of the caregiver to these problematic behaviors (ranging fromnot at alltoextremely).
A timeline of the different assessment moments and outcomes can be found in Figure 1.
2.8
|Power calculation
A sample size calculation was executed in SPSS Sample Power 3 for the primary outcome caregiver perceived burden. An effect size of 0.4 was defined based on a similar high‐quality trial, which implies a difference of six points on the ZBI. The standard deviation (SD) was set at 15.29 To obtain a power of 80% and a significance level at
0.05, each group had to contain 100 dyads. Next, to reduce selection‐ bias inherent to quasi‐experimental studies, an allocation ratio of 1:2 was used to allow propensity score matching (see below) and addition-ally to deal with drop‐out, an average estimated drop‐out rate of 20% was considered as well. As such, it was estimated that 124 dyads should be recruited in the intervention group and 248 in the control group in order to finally compare 100 dyads in each study arm after 6 months.
2.9
|Statistical analyses
First, descriptive statistics were carried out to draw a clear profile of the characteristics of study participants. To determine possible baseline differences between the groups, mean values of baseline characteristics were compared using independent sample t‐tests for continuous variables if normally distributed or by performing Pearson's Chi‐square test for categorical variables. To help control for bias and confounding, propensity score matching (nearest neighbor method) was carried out.30Matching between both groups at baseline was based on variables for which significant baseline differences were found between both study arms and on variables for which it was demonstrated in the literature that these were associated with self‐perceived burden.31,32These were: age of the informal caregiver, relationship between caregiver and patient, being on a waiting list for nursing home placement, severity of dementia, level of depen-dency, desire to institutionalize the patient, total time spent in caregiv-ing per day, caregiver burden, frequency of behavioral problems, impact of behavioral problems on the caregiver, and educational level of the caregiver.
To investigate possible effects of the intervention at 6 months of the matched sample, mixed model analyses were carried out for each outcome separately (defined as dependent variable in the model).
To check whether the assumptions had been met, graphical plots of residuals and predicted values were investigated. A P value of.05 was considered significant. Time (baseline & 6 months' outcomes) and group (intervention or control group) were defined as fixed factors in the mixed model, while caregiver unique ID was defined as the ran-dom factor. When a significant difference in a certain variable was still present after matching, this was defined as a covariate in the model. Finally, the propensity score was also included as a covariate. In liter-ature, this method is described as double propensity‐score adjustment and is known as a solution for the problem design bias or bias due to incomplete matching.33
3
|R E S U L T S
3.1
|Sample description
A response rate of 77% was reached meaning that, of the 465 care-givers recruited in the participating settings, 355 eventually completed the baseline evaluation during a home visit. In total, 252 dyads were allocated to the control group and 103 to the intervention group.
After propensity score matching, a total of 198 participants remained eligible for the effectiveness analysis: 99 caregiver/patient dyads in each study arm. Six months post‐baseline, 149 dyads com-pleted the primary endpoint. Figure 2 displays the recruitment process and flow of participants.
3.2
|Characteristics of the study participants and
matching
The descriptive statistics of the 198 matched dyads characteristics are outlined in Table 1. Most caregiver and patient characteristics were FIGURE 1 Timeline of assessments within the trial
well‐balanced after conducting the matching procedure, meaning that no significant difference remained between both study arms. Patient age (not used as matching variable) was included in the mixed model as a covariate because a significant difference between both groups remained. The overall balance test of Hansen and Bowers34(to evalu-ate the matching quality) indicevalu-ated good overall balance (x2= 13.08, p = .60) as well. Only some minor imbalance remained for the level of dependency of the patient: the standardized mean difference after matching was 0.27 while the aim is to be below 0.25.35Therefore, level of dependency was also included in the mixed model as a covariate.36
3.3
|Effectiveness of the in
‐
home respite care
program
Table 2 summarizes average scores at baseline and 6 months post‐ baseline and gives the adjusted estimated effect of the in‐home respite care program compared with standard dementia care for the primary and secondary outcomes.
The caregiver burden (primary outcome) did not show a significant mean difference between intervention and control group (−0.13; 95% CI,−3.27 to 3.02; p = .94) at 6 months post‐baseline. However, the
caregivers' desire to institutionalize the patient (secondary outcome) showed a significant mean difference of −0.51 (95% CI, −0.97 to 0.55; p = .023), in favor of the intervention group. More specifically, in the intervention group, the desire to institutionalize the patient in residential care decreased after 6 months (from 1.94 [SD = 0.18] to 1.77 [SD = 0.20]) while in the control group, the desire increased (from 2.09 [SD = 0.17] to 2.42 [SD = 0.20]). The effect size of this statisti-cally significant difference was calculated by dividing the mean differ-ence between both groups by the pooled standard deviation. On this basis, this Cohen'sdof.3, it can be concluded that the effect is small but clinically relevant.37
There was no significant mean difference at 6 months post‐ base-line in the caregiver's quality of life (0.01; 95% CI,−0.04 to 0.07; p = .59). Although caregiver quality of life obtained by the EQVAS in the intervention group increased (from 67.38 [SD = 2.00] to 70.06 [SD = 1.93]) while it decreased in the control group (from 68.82 [SD = 1.99] to 67.50 [SD = 1.97]), there was only a trend towards a signif-icant mean difference after 6 months post‐baseline (4.00; 95% CI,
−1.29 to 9.28; p = .137). Finally, no significant mean difference was detected for frequency of behavioral problems (−0.01; 95% CI,−0.15 to 0.14; p = .95) and for the impact of those problematic behaviors on the caregiver (0.01; 95% CI,−0.19 to 0.21; p = .89).
TABLE 1 Characteristics of the study participants Matched Control Group N = 99 Matched Intervention Group N = 99 Total Matched Group N = 198 Caregiver age in years, mean (SD) 67.15 (12.16) 64.79 (12.17) 65.97 (12.17) Caregiver gender, % (n) Female 72% (71) 75% (74) 73% (145) Region, % (n) Flanders 67% (66) 22.% (22) 44% (88) Wallonia 31% (31) 53% (52) 42% (83) Brussels 2% (2) 25% (25) 14% (27)
Caregiver professional situation, % (n)
Active on the labor market 69% (68) 57% (56) 23% (45)
Non‐active on the labor market 12% (12) 17% (17) 15% (29)
Retired 19% (19) 26% (26) 63% (124)
Multiple caregivers, % (n)
Yes 36% (36) 37% (37) 37% (73)
Change in professional situation (working less hours) to deal with caregiving tasks 9% (9) 22% (22) 16% (31) Caregiver educational level, % (n)
Low 15% (15) 14% (14) 15% (29)
Middle 30% (30) 21% (22) 26% (52)
High 55% (54) 64% (63) 59% (117)
Caregiver civil state, % (n)
Alone 11% (11) 8% (8) 10% (19)
Married/legal cohabitation 77% (76) 72% (71) 74% (147)
Unmarried 5% (5) 8% (8) 7% (13)
Widow 2% (2) 1% (1) 2% (3)
Divorced 5% (5) 11% (11) 8% (16)
Type of relationship caregiver/person with dementia, % (n)
Partner/spouse 63% (62) 57% (56) 60% (118)
Daughter/son 26% (26) 34% (34) 30% (60)
Other family member or friend 11% (11) 9% (9) 10% (20)
Cohabitation with person with dementia, % (n)
Yes 74% (73) 84% (83) 79% (156)
Caregivers burden, mean (SD) 39.76 (14.36) 41.21 (14.67) 40.48 (14.50)
Caregiver desire to institutionalize the patient, mean (SD) 2.02 (1.84) 2.03 (1.60) 2.03 (1.72)
EQ‐5D index, mean (SD) 0.70 (0.214) 0.72 (0.22) 0.71 (0.22)
EQ‐VAS, mean (SD) 69.2 (18.8) 66.92 (20.37) 68.06 (19.58)
Time spent in caregiving hours/day, mean (SD) 8.82 (6.46) 9.03 (5.75) 8.93 (6.10) Person with dementia age in years, mean (SD) 77.55 (9.87) 81.35 (8.29) 79.45 (9.29)
Waiting list for residential care, % (n) 43% (43) 47% (46) 44% (89)
Person with dementia gender, % (n)
Female 48% (47) 54% (53) 51% (100)
Person with dementia educational level, % (n)
Low 39% (39) 30% (30) 35% (69)
Middle 28% (28) 20% (20) 25% (48)
High 30% (30) 50% (49) 40% (79)
3.4
|Short
‐
term effect of the in
‐
home respite care
program
Table 3 summarizes average scores at baseline and 14 to 15 days post‐intervention in the intervention group and gives the adjusted estimated time effect for the primary and secondary outcomes.
There was no significant effect over time on the total caregiver burden (−0.01, 95% CI, −1.85 to 1.85; p = .99) between baseline and 14 to 15 days post‐intervention. However, two burden sub-scales did show a significant difference over time between baseline and 14 to 15 days post‐intervention. First of all, there was a signif-icant time effect of burden on social and family life (0.55; 95% CI, TABLE 1 (Continued) Matched Control Group N = 99 Matched Intervention Group N = 99 Total Matched Group N = 198 Person with dementia civil state, % (n)
Alone 2% (2) 1% (1) 2% (3) Married/legal cohabitation 67% (66) 57% (56) 62% (122) Unmarried 3% (3) 3% (3) 3% (6) Widow 26% (26) 35% (35) 31% (61) Divorced 2% (2) 4% (4) 3% (6) Severity of dementia, % (n)
Moderate cognitive decline 12% (12) 10% (10) 11% (22)
Moderately severe cognitive decline 41% (41) 35% (35) 38% (76)
Severe cognitive decline 41% (41) 43% (43) 42% (84)
Very severe cognitive decline 5% (5) 11% (11) 8% (16)
Level of dependency, mean (SD) 19.17 (5.87) 20.47 (6.43) 19.82 (6.18)
Frequency of behavioral problems in the last week, mean (SD) 1.61 (0.50) 1.58 (0.54) 1.59 (0.52) Impact of behavioral problems on the caregiver in the last week, mean (SD) 1.69 (0.71) 1.78 (0.71) 1.73 (0.71) Use of Home aid services in the last month (district nurse, home aid, sitting service), % (n)
Yes 67% (66) 84% (83) 75% (149)
Use of accommodation in the last 6 months (short stay, day care, host family), % (n)
Yes 42% (42) 43% (43) 43% (85)
Hospitalization or emergency room visit in the last 6 months, % (n)
Yes 28% (28) 29% (29) 29% (58)
Outpatient visits in the last month, % (n)
Yes 92% (91) 85% (84) 88% (175)
TABLE 2 Primary and secondary outcomes in intervention and control group (results for the mixed model)
Outcomes
Control Group Intervention Group In‐Home Respite Versus Standard Care Baseline N = 99 6 mo N = 73 Baseline N = 99 6 mo N = 76
Treatment effect adjusted difference,a (95% CI),Pvalue
Caregiver burden 40.73 (1.33) 41.03 (1.35) 39.95 (1.34) 40.12 (1.33) −0.13 (−3.27 to 3.02), .94 Desire to institutionalize 2.09 (0.17) 2.42 (0.20) 1.94 (0.18) 1.77 (0.20) −0.51 (−0.97 to 0.55), .023b Caregiver quality of life
EQ 5D (utilities) 0.69 (0.02) 0.67 (0.02) 0.73 (0.02) 0.72 (0.02) 0.01 (−0.04 to 0.07), .59 EQVAS 68.82 (1.99) 67.50 (1.97) 67.38 (2.00) 70.06 (1.93) 4.00 (−1.29 to 9.28), .137 Frequency of behavioral problems 1.61 (0.05) 1.71 (0.06) 1.57 (0.05) 1.67 (0.06) −0.01 (−0.15 to 0.14), .95 Impact of behavioral problems 1.71 (0.06) 1.62 (0.08) 1.74 (0.06) 1.67 (0.08) 0.01 (−0.19 to 0.21), .89 Data are mean (Standard deviation) unless otherwise specified.
aAdjusted for propensity score, level of dependency (KATZ), and age of dementia patient. bSignificant difference
0.01‐1.09; p = .047). Another significant effect was found for the subscale role strain. The caregivers, on 14 to 15 days post‐ intervention, scored significantly lower on this subscale than before obtaining the in‐home respite support (0.75; 95% CI, −0.01 to 1.50; p = .052). Desire to institutionalize the patient was also signif-icantly lower 14 to 15 days after obtaining the in‐home respite pro-gram compared with before (0.31; 95% CI, 0.05‐0.58; p = .023). Frequency of behavioral problems remained the same before and
14 to 15 days post‐intervention (0.01; 95% CI, −0.06 to 0.08; p = .82). Finally, the decrease in the extent to which the caregiver was affected by the behavioral problems shortly after the interven-tion compared with before was not significant (0.11; 95% CI,
−0.03 to 0.25; p = .134).
Figure 3 displays the changes over time in the intervention and control group for the primary and secondary outcomes adjusted for propensity score, patient age, and patient level of dependency. TABLE 3 Short‐term effects of time in intervention group (results for pairwise comparisons in mixed model)
Outcomes
Intervention Group Post‐Intervention Versus Pre‐Intervention Baseline N = 99 14‐15 d post‐intervention N = 98 Time effect adjusted differencea, (95% CI),Pvalue Caregiver burden
Total score 40.35 (1.32) 40.36 (1.34) −0.01 (−1.85 to1.85), .99 Burden on social & family life 7.65 (0.38) 7.10 (0.38) 0.55 (0.01 to 1.09), .047b Role strain 12.65 (0.53) 11.90 (0.53) 0.75 (−0.01 to 1.50), .052b Desire to institutionalize 1.96 (0.17) 1.65 (0.18) 0.31 (0.05 to 0.58), .023b Frequency of behavioral problems 1.58 (0.05) 1.57 (0.05) 0.01 (−0.06 to 0.08), .82 Impact of behavioral problems 1.75 (0.06) 1.64 (0.08) 0.11 (−0.03 to 0.25), .134 Note. Data are mean (standard deviation) unless otherwise specified.
aAdjusted for propensity score, level of dependency (KATZ), and age of dementia patient. bSignificant difference
FIGURE 3 Changes over time in the intervention and control group for the primary and secondary outcomes (mean scores, adjusted for patient age & dependency, propensity score)
4
|D I S C U S S I O N
Our study shows that informal caregivers of persons with dementia benefitted partially from a temporary in‐home respite care program shortly after the respite episode (14‐15 d) and 6 months later. After 6 months, the intervention group had a significant lower desire to institutionalize the person with dementia. Shortly after the respite episode, the intervention group reported not only a sig-nificant lower desire to institutionalize the person with dementia, but also a significant lower role strain and a lower burden on social and family life.
The finding that the desire to institutionalize (DIS) significantly decreased in the intervention group after 6 months is important. Indeed, DIS is a well‐known valid proxy for predicting time to actual placement.38 When caregivers have a lower DIS, they can better maintain their role, and dementia patients can remain longer in their trusted environment. The latter is preferred because staying home is associated with better patient quality of life and less morbidity. Moreover, institutionalization puts more financial pressure on the healthcare budget. Research has proven also that lower DIS is associated with lower burden, and with being able to better cope with the behavioral problems of the dementia patient.39 However, our trial did not reveal a significantly reduced caregiver burden nor improved coping with behavioral problems. One may assume that with a larger sample size, such a significant effect might have been found. But when considering the pvalues, it is unlikely that a significant or clinically meaningfull effect would have been found in the primary outcome. Only for the quality of life outcome, for which a clear trend of improvement in the inter-vention arm was observed, a larger sample size might have led to a significant result.
This study was the first comparative study to investigate effec-tiveness of this specific in‐home respite care program to support informal caregivers of persons with dementia. The results partly con-firmed earlier positive findings from explorative qualitative studies into the potential benefits of respite care (ie, reduced stress, being able to maintain role, and lower strain).14,16,19,20 We believe the study has moreover some methodological strengths. First, a quasi‐ experimental trial was set up for practical and ethical reasons and because this design was earlier identified as a favorable and valid strategy to conduct research into respite care efficacy.8 Indeed, it can be seen as ethically unacceptable to randomly assign dyads to one of the trial arms withholding control group dyads to receive the program for 1 year after first informing them about the possibil-ity of getting this support for free. Also, because there is no plausi-ble placebo alternative for the intervention, participants allocated to a control group in a randomized way would have been more prone to drop out. Second, to avoid bias and confounding inherent to quasi‐experimental studies that lack random assignment of partici-pants, a propensity score matching was performed. In this way, we tried to come close to an RCT setting and effects could be assigned to the intervention. Third, to further improve internal validity, we have only included participants in the control group that had
indicated that they were prepared to utilize in‐home respite care. Fourth, internationally validated questionnaires were used. These were adapted to the Belgian context when needed, and forward‐ back translations were performed when no Dutch or French version existed. Fifth, active components of the intervention are clearly described, and drop‐out or termination of participation was docu-mented. A sixth methodological strength was the fact that the first assessment was conducted during a home visit. Face‐to‐face interviewing remains the preferred type of data collection for this type of projects and could be the explanation for the very low drop‐out rate present in this trial.
Some limitations also need to be mentioned. First, although a pro-pensity score matching was performed, there remains a risk towards internal validity because unmeasured confounders could still be pres-ent. For example, in this trial, the duration since diagnosis of dementia was not considered. A second limitation is that only participants will-ing to consider in‐home respite were allowed in this trial. As such, we cannot speculate whether effects would have been different for caregivers not willing or not asking for this kind of support on their own initiative. Third, as in many trials, participants with lower socio-economic status and low educational level may be underrepresented. Possibly, we were not able to include those most in need for support. However, the fact that intervention group participants got 5 days of the program for free might have alleviated this problem to a certain level and have attracted families who would probably not have consid-ered the service before because of financial reasons. Indeed, as this service is not reimbursed, this would have come with a considerable cost for the caregiver. A fourth limitation is that we did not control for the exact duration of the intervention. Even by setting a minimum duration of 5 days as an inclusion criterion, differences in outcome variables can still be caused by differences in duration. However, the variation in duration remained limited: 65% of the sample received the intervention during 5 to 10 days. The average was 9.2 days, and the maximum duration was 17 days. Fifth, as in many supportive care-giver interventions, the rather limited duration of the trial and short follow‐up can be considered a weakness. Because dementia is a slow degenerative process, it is to be expected that differences in time to placement are not to be found on such short notice. However, to alleviate this problem we have used the desire to institutionalize scale, which is a valid proxy for actual placement. Sixth, as discussed above, the trial did only reach 73% of the intended sample size of 100 dyads in each study arm at 6 months. Seventh, in contrast to earlier respite care studies (especially into temporary residential admission11), we did not detect any adverse effects of the program in the primary and secondary research outcomes. Nevertheless, it could be interesting to expand current evidence with an additional quantitative and/or qualitative study into the potential negative effects (such as reduced sleep and increased deterioration) and nega-tive experiences of caregivers and care‐recipients. Finally, in order to allow policy makers to make an informed decision on whether to adopt this service into standard dementia care in the future, a cost‐ effectiveness analysis of this program as well as a budget impact anal-ysis should be conducted.
5
|C O N C L U S I O N
In conclusion, the investigated in‐home respite care program was effective in decreasing the desire of the caregiver to institutionalize the care‐recipient after 6 months, but not the caregiver burden. In a next step, longer‐term evidence is necessary to investigate if this decreased desire leads to delayed institutionalization of the persons with dementia. Furthermore, there will be a need for a health eco-nomic evaluation of the program.
A C K N O W L E D G E M E N T S
The authors would like to thank all participants and the recruiting organizations for their collaboration in this study. They would also like to thank the volunteers, jobstudents, and especially Lisa Van Wilder for their technical support and contribution to the data collection. This study was funded by the Baillet‐Latour Fund (E/01447/15). The funding source had no role in the design, conduct, or reporting.
C O N F L I C T O F I N T E R E S T None declared.
D A T A A V A I L A B I L I T Y S T A T E M E N T
The data that support the finding of this study are available from the corresponding author upon reasonable request.
O R C I D
Sophie Vandepitte https://orcid.org/0000-0002-6965-6775
R E F E R E N C E S
1. Brodaty H, Donkin M. Family caregivers of people with dementia. Dia-logues Clin Neurosci. 2009;11(2):217‐228.
2. Prince MR, Adelina C‐H, Knapp M, Guerchet M, Karagiannidou M. World Alzheimer report 2016: improving healthcare for people living with dementia: coverage, quality and costs now and in the future. London, UK 2016.
3. Knapp M. The cost‐effectiveness challenge: is it worth it?Alzheimer's Res & Ther. 2015;7(1):10‐10.
4. Koopmanschap MA, van Exel NJ, van den Bos GA, van den Berg B, Brouwer WB. The desire for support and respite care: preferences of Dutch informal caregivers.Health Policy. 2004;68(3):309‐320. 5. Vandepitte S, Van den Noortgate N, Putman K, Verhaeghe S, Faes K,
Annemans L. Effectiveness of supporting informal caregivers of people with dementia: a systematic review of randomized and non‐ randomized controlled trials.J Alzheimers Dis. 2016;52(3):929‐965. 6. Zwaanswijk M, Peeters JM, van Beek AP, Meerveld JH, Francke AL.
Informal caregivers of people with dementia: problems, needs and sup-port in the initial stage and in subsequent stages of dementia: a questionnaire survey.Open Nurs J. 2013;7:6‐13.
7. Sorensen S, Duberstein P, Gill D, Pinquart M. Dementia care: mental health effects, intervention strategies, and clinical implications.Lancet Neurol. 2006;5(11):961‐973.
8. Zarit SH, Bangerter LR, Liu Y, Rovine MJ. Exploring the benefits of respite services to family caregivers: methodological issues and current findings.Aging Ment Health. 2017;21(3):224‐231.
9. Maayan N, Soares‐Weiser K, Lee H. Respite care for people with dementia and their carers.Cochrane Database Syst Rev. 2014;16(1): Cd004396.
10. Neville C, Beattie E, Fielding E, MacAndrew M. Literature review: use of respite by carers of people with dementia.Health Soc Care Commu-nity. 2015;23(1):51‐63.
11. Vandepitte S, Van Den Noortgate N, Putman K, Verhaeghe S, Verdonck C, Annemans L. Effectiveness of respite care in supporting informal caregivers of persons with dementia: a systematic review. Int J Geriatr Psychiatry. 2016;52(3):929‐965.
12. Grant I, McKibbin CL, Taylor MJ, et al. In‐home respite intervention reduces plasma epinephrine in stressed Alzheimer caregivers. Am J Geriatr Psychiatry. 2003;11(1):62‐72.
13. Arksey H, Jackson K, Croucher K, et al. Review of respite services and -short‐term breaks for carers for people with dementia. London: London School of Hygiene and Tropical Medicine, National Co‐ ordinating Centre for NHS Service Delivery and Organisation R & D (NCCSDO). 2004.
14. O'Connell B, Hawkins M, Ostaszkiewicz J, Millar L. Carers' perspectives of respite care in Australia: an evaluative study. Contemp Nurse. 2012;41(1):111‐119.
15. van Exel J, Moree M, Koopmanschap M, Goedheijt TS, Brouwer W. Respite care—an explorative study of demand and use in Dutch infor-mal caregivers.Health Policy. 2006;78(2‐3):194‐208.
16. Ashworth M, Baker AH. 'Time and space': carers' views about respite care.Health Soc Care Community. 2000;8(1):50‐56.
17. Parahoo K, Campbell A, Scoltock C. An evaluation of a domiciliary respite service for younger people with dementia.J Eval Clin Pract. 2002;8(4):377‐385.
18. van Exel J, de Graaf G, Brouwer W. Give me a break! Informal caregiver attitudes towards respite care.Health Policy. 2008;88(1):73‐87. 19. Zarit SH, Gaugler JE, Jarrott SE. Useful services for families: research
findings and directions.Int J Geriatr Psychiatry. 1999;14(3):165‐177. 20. Stirling CM, Dwan CA, McKenzie AR. Why carers use adult day respite:
a mixed method case study.BMC Health Serv Res. 2014;6(14):245. 21. Vandepitte S, Van Den Noortgate N, Putman K, Verhaeghe S,
Annemans L. Effectiveness and cost‐effectiveness of an in‐home respite care program in supporting informal caregivers of people with dementia: design of a comparative study.BMC Geriatr. 2016;16(1):207. 22. Zarit SH, Reever KE, Bachpeterson J. Relatives of the impaired elderly:
correlates of feelings of burden.Gerontologist. 1980;20(6):649‐655. 23. Morycz RK. Caregiving strain and the desire to institutionalize family
members with Alzheimer's disease. Possible predictors and model development.Res Aging. 1985;7(3):329‐361.
24. McCaskill GM, Burgio LD, Decoster J, Roff LL. The use of Morycz's desire‐to‐institutionalize scale across three racial/ethnic groups. J Aging Health. 2011;23(1):195‐202.
25. Pruchno RA, Michaels JE, Potashnik SL. Predictors of institutionaliza-tion among alzheimer‐disease victims with caregiving spouses. J Gerontol. 1990;45(6):S259‐S266.
26. Herdman M, Gudex C, Lloyd A, et al. Development and preliminary testing of the new five‐level version of EQ‐5D (EQ‐5D‐5 L).Qual Life Res. 2011;20(10):1727‐1736.
27. Devlin NJ, Shah KK, Feng Y, Mulhern B, van Hout B. Valuing health‐ related quality of life: an EQ‐5D‐5 L value set for England.Health Econ. 2018;27(1):7‐22.
28. Teri L, Truax P, Logsdon R, Uomoto J, Zarit S, Vitaliano PP. Assessment of behavioral problems in dementia: the revised memory and behavior problems checklist.Psychol Aging. 1992;7(4):622‐631.
29. Martin‐Carrasco M, Dominguez‐Panchon AI, Gonzalez‐Fraile E, Munoz‐Hermoso P, Ballesteros J. Effectiveness of a psychoeducational intervention group program in the reduction of the burden experi-enced by caregivers of patients with dementia: the EDUCA‐II randomized trial.Alzheimer Dis Assoc Disord. 2014;28(1):79‐87. 30. Austin PC. An introduction to propensity score methods for reducing
the effects of confounding in observational studies. Multivar Behav Res. 2011;46(3):399‐424.
31. Etters L, Goodall D, Harrison B. Caregiver burden among dementia patient caregivers: a review of the literature.J Am Acad Nurse Pract. 2008;20(8):423‐428.
32. Cheng ST. Dementia caregiver burden: a research update and critical analysis.Curr Psychiatry Rep. 2017;19(9):64.
33. Austin PC. Double propensity‐score adjustment: a solution to design bias or bias due to incomplete matching. Stat Methods Med Res. 2017;26(1):201‐222.
34. Hansen BB, Bowers J. Covariate balance in simple, stratified and clus-tered comparative studies.Stat Sci. 2008;23(2):219‐236.
35. Austin PC. Balance diagnostics for comparing the distribution of base-line covariates between treatment groups in propensity‐score matched samples.Stat Med. 2009;28(25):3083‐3107.
36. Austin PC, Grootendorst P, Anderson GM. A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: a Monte Carlo study.Stat Med. 2007;26(4):734‐753.
37. Sullivan GM, Feinn R. Using effect size—or why theP value is not enough.J Grad Med Educ. 2012;4(3):279‐282.
38. Gallagher D, Ni Mhaolain A, Crosby L, et al. Determinants of the desire to institutionalize in Alzheimer's caregivers.Am J Alzheimers Dis Other Demen. 2011;26(3):205‐211.
39. Vandepitte S, Putman K, Van Den Noortgate N, et al. Factors associ-ated with the caregivers' desire to institutionalize persons with dementia: a cross‐sectional study. Dement Geriatr Cogn Disord. 2018;46(5‐6):298‐309.
How to cite this article: Vandepitte S, Putman K, Van Den Noortgate N, Verhaeghe S, Annemans L. Effectiveness of an in‐home respite care program to support informal dementia caregivers: A comparative study. Int J Geriatr Psychiatry. 2019;1–11.https://doi.org/10.1002/gps.5164
