Who are the bene ciaries and what are the reasons for non-utilization of respite care? A cross-sectional study on family caregivers


 BackgroundFamily caregivers assume substantial caregiving responsibilities for persons with chronic conditions, which leads to negative impact on their lives. Respite care is provided as a temporary relief for them. The design of appropriate respite care programs requires the identification of beneficiary subgroups for different types of services. The objectives of the study were to quantify the uptake of respite care services by family caregivers of persons with spinal cord injury, and to identify the main beneficiaries of the respective offers and the reasons for non-use.MethodsA cross-sectional survey of family caregivers of persons with spinal cord injury was conducted nationwide in Switzerland. The use of 11 different respite care services during the last 12 months was investigated, along with caregivers’ reasons for not using any respite service. Classification trees were used to characterize the beneficiaries and reasons for not using respite services.ResultsOne-third of family caregivers used at least one type of respite care service during the last 12 months. Utilization of respite care was found to be greater among those who employed professional home care (57% vs 24% among those without professional home care). There were marked cantonal differences in the utilization of respite care. The primary reason for not using respite services was “no demand” (80% among non-users of respite services), mainly among caregivers who were less emotionally affected by their caregiving tasks. Conclusions Utilization of respite care services primarily depends on living arrangements and place of residency and less on the functional status of the care recipient. Programs should thus be tailored to the cultural context of their potential users. This is best achieved through coordination with local health care professionals who can identify the needs, provide information, initiate referrals, and integrate the care into a larger support plan.


Abstract Background
Family caregivers assume substantial caregiving responsibilities for persons with chronic conditions, which leads to negative impact on their lives. Respite care is provided as a temporary relief for them. The design of appropriate respite care programs requires the identi cation of bene ciary subgroups for different types of services. The objectives of the study were to quantify the uptake of respite care services by family caregivers of persons with spinal cord injury, and to identify the main bene ciaries of the respective offers and the reasons for non-use.

Methods
A cross-sectional survey of family caregivers of persons with spinal cord injury was conducted nationwide in Switzerland. The use of 11 different respite care services during the last 12 months was investigated, along with caregivers' reasons for not using any respite service. Classi cation trees were used to characterize the bene ciaries and reasons for not using respite services.

Results
One-third of family caregivers used at least one type of respite care service during the last 12 months.
Utilization of respite care was found to be greater among those who employed professional home care (57% vs 24% among those without professional home care). There were marked cantonal differences in the utilization of respite care. The primary reason for not using respite services was "no demand" (80% among non-users of respite services), mainly among caregivers who were less emotionally affected by their caregiving tasks.

Conclusions
Utilization of respite care services primarily depends on living arrangements and place of residency and less on the functional status of the care recipient. Programs should thus be tailored to the cultural context of their potential users. This is best achieved through coordination with local health care professionals who can identify the needs, provide information, initiate referrals, and integrate the care into a larger support plan.

Background
Informal care can have positive effects on the well-being of persons with chronic conditions [1]. Family caregivers are mostly involved in providing nursing, emotional, and practical support [2,3]. They assume substantial responsibilities, which can have negative impacts on their lives [4,5]. Respite care is available to provide relief to family caregivers through short breaks, advice, or social connections [6]. Services vary depending on whether they are provided by local municipalities, non-pro t organizations, or private providers [7]. Utilization of respite care has been found to be generally low in different countries despite high levels of need [8,9]. The design of appropriate respite care programs requires the identi cation of bene ciary subgroups for different types of services [9,10]. Although variations in respite use have been found among caregivers in relation to their age, cultural orientation, nances, and need for respite [11][12][13][14], the commonly applied regression modeling is not su cient for identifying such subgroups. In regression models, a linear relationship is assumed, only additive effects are detected for average members of a population, and there is a risk of over tting when considering many predictors. The options for modeling non-linear relationships are limited, and the interaction between predictors is di cult to interpret [15]. To balance these limitations, the present study thus proposes the use of classi cation trees as a tool to perform a comprehensive exploration of respite care utilization, considering a wide spectrum of predictors.
This study aimed to quantify the uptake of different respite care services for family caregivers, the reasons for non-use, and the respective predictors. Spinal cord injury (SCI) in Switzerland was used as an example of a complex chronic condition requiring long-term care. Speci cally, the study aimed to 1) identify the utilization of various respite services and the characteristics of their main bene ciaries among family caregivers and 2) investigate the reasons for the non-use of respite services and the respective predictors.

Data collection
A survey of family caregivers for persons with SCI was conducted in Switzerland. Persons with SCI were contacted from August 2016 to July 2017 and asked to forward a questionnaire to their primary family caregiver. Their contact information was acquired from the Swiss Spinal Cord Injury Cohort Study (SwiSCI) database [16]. The study inclusion criteria for persons with SCI followed those of SwiSCI, focusing on persons with a chronic SCI for more than 2 years, aged over 16 years, and residing in Switzerland. The exclusion criteria were persons with SCI resulting from congenital conditions, new injury in end-of-life care, or neurodegenerative disorders. Persons with SCI who reported not having a family caregiver were excluded. "Family caregiver" was de ned as a partner, direct relative, or relative-in-law who cared for or assisted the person with SCI in daily living at home. The inclusion criteria for family caregivers were age over 18 years and ability to ll out the questionnaire in one of the survey languages (German, French, or Italian).

Questionnaire and variables
The family caregivers were asked if they used any of the 11 respite care services over the last 12 months.
For those who indicated not having used any respite care service, the reasons were classi ed among eight categories with multiple answers possible. Additional questions covered their socio-demographic characteristics, living situation, employment, nances, caregiving situation, and characteristics of the person with SCI. The caregiver's quality of life (QoL) was measured with a single item about overall quality of life (WHOQOL-BREF) [17]. Twelve items of the COPE-index [18] were used to capture the participant's perception of the positive and negative aspects of caregiving.

Data analysis
Odds ratios for respite care use were calculated between groups of study participants. Need for respite services was expressed as caregiving burden, operationalized by the caregiver's QoL, the care recipient's wheelchair dependency, and the caregiver's time investment in care.
Predictors of respite care service utilization were identi ed using classi cation and regression trees (CART) [15]. Separate classi cation trees were built for each of the 11 types of respite care services and for the use of any respite care service as outcomes. Another set of classi cation trees was built for the eight reasons for the non-use of respite services as outcomes. In all classi cation trees, 133 potential predictors were included (Supplementary Table 1). The results are summarized in tables with the most predictive subgroups for the respective outcomes. Data were prepared with Stata 16 for Windows (College Station, TX, USA). Classi cation trees were built with the R package "rpart" [19].

Sample characteristics
Among the 4502 invitations, 1259 persons with SCI did not live with a family caregiver. Those excluded also included 532 persons who were ineligible based on other criteria, 397 untraceable contacts, 110 participation refusals, and 1487 non-responses. A total of 717 questionnaires were returned (response rate 31%), and of those, 679 answered the question about respite care (95%). The primary family caregivers were mostly female (72%), and the average age was 57 years old (Table 1). They were mostly the spouse of the person with SCI (76%), and 84% lived in the same household as their care recipient. The median care duration was 9 years, with a median time investment of 12 hours per week. One-third of the participants reported having used insurance-covered professional home care, with a median use of 8 hours per week.  Table 2). One-third reported having used at least one type of respite care service during the last 12 months (Table 3). The most commonly used respite service was a driving service for the care recipients (16%), followed by household support (14%). Support groups and training courses were used by the fewest participants (1% each).  Wheelchair dependency: 0 = not fully wheelchair dependent (able to stand, able to walk); 1 = fully wheelchair dependent.
*=p < 0.05, **=p < 0.01, ***=p < 0.001. Table 3 Predictors of utilization of respite services during the last 12 months Utilization of respite care services was found to be higher among participants with professional home care (57% vs 24% among those without professional home care). Among those, up to 64% of the use was explained by the canton (region of residence). An association with professional home care was also found for speci c types of respite care services: driving services (29% vs 1% among those with less than 1.5 weekly hours of professional home care), household support (28% vs 8% among those with less than 0.75 weekly hours of professional home care), and relief offers for holidays (16% vs 3% among those without professional home care). Utilization of professional home care and canton of residence were the strongest determinants of respite care utilization (Supplementary Table 2), and they were the primary predictors of the utilization of ve types of respite care services and of the overall utilization of respite care. Receiving nancial compensation was the most decisive factor for using daytime respite services, while assisting in feet washing was the most decisive for social companionship. Family caregivers' time investment in care was a weak predictor of respite care utilization, as was the care recipient's functional status, caregiver's satisfaction about their nancial situation, and income.
The reason of "no demand" for respite services was primarily linked to a less negative impact of caregiving on the caregiver's emotional well-being and to higher satisfaction with their nancial situation. The vast majority (91%) of the participants with these characteristics reported not having a need for respite care, and this group constitutes 61% of the total non-user group. Participants who were negatively affected by caregiving and did not mention "no demand" for respite care were geographically clustered.
Half of them lived in eight-mainly French-speaking-cantons (out of 26) and mentioned barriers other than "no demand". Family caregivers who invested an excessive amount in care reported more often the care recipient's preference for family care as a reason for not using respite care services. The canton of residence was the most important factor for delineating the various reasons for not using respite care services (Supplementary Table 3). Table 4 Predictors of reasons for non-utilization of respite services during the last 12 months

Discussion
Regional variability was most decisive factor for respite care use and for the reasons for non-utilization. This nding agrees with that from a national study that found geographic variations in respite uptake among families with a member with an intellectual disability, as respite care is provided locally [20]. Similarly, health care use in Switzerland is linked to the cantonal supply of services and nancing [21].
Households with persons with SCI tend to cluster in cantons where the SCI-speci c services are well established [22]. However, the participants in the present study rarely mentioned unavailability as a reason for non-use. Although nancial support varies across cantons [6], this did not seem to be a main driver, as only a few participants mentioned cost as a barrier to access respite care services, contrary to a previous study conducted in Australia [12]. Cultural characteristics, as described by the language region, did not fully explain the unequal local utilization of respite care services, although health care use varies across the different language regions in Switzerland [23]. As there are no national regulations for respite care, other local characteristics may be more in uential, such as the passive views, family values toward respite care, and cantonal demographic features [9,14,21].
Family caregivers who employed professional home care were more likely to use respite care services. The two do not seem to be linked through the families' nancial capacity, as it was found to not have an impact. The predictive power of caregiver burden on respite care utilization was small, though other studies have identi ed this as a strong determinant [24,25]. As the use of home nursing has been associated with a higher physical dependency [26], these family caregivers seemed to need both sources of support. The results from the present study also con rm the previous nding that professional home care is the main source of temporary breaks and condition-speci c consultations for the family caregivers [6].
It was not surprising that "no demand" was the most common reason for non-utilization. The statement of "no demand" was linked to the subjective perception of burden. Similar to the ndings regarding the care of persons with dementia, the care for the person was the main consideration when deciding about utilization, and family caregivers rarely declared their needs for a break [8,9]. Family caregivers may experience feelings of guilt or fear when leaving the care recipients alone [11,26], or they may wish to maintain a positive aspect, such as close relationship with the care recipient [27]. Even when most family caregivers wish to continue the care by themselves, they trust health professionals as a reliable information source and acknowledge their central role in encouraging caregivers to use external support [27]. Because they are in close contact with the families, health professionals can identify the needs, if not explicitly expressed by the family caregivers, and provide information or initiate referrals [13]. As the perceived QoL of family caregivers might indicate the need for support, it is recommended that this indicator be included in clinical or routine assessments so services can be initiated for those who are in need.

Conclusion
Family caregivers provide high-intensity care over many years and prioritize the needs of the care recipients. Here, classi cation trees were used to identify that respite care use was primarily determined by contextual factors. The respite programs should thus be adapted to the cultural, regional, and personal contexts of their intended users. This is best achieved in coordination with local health care professionals who can identify the needs, provide individual information, initiate referrals, and integrate care activities into a larger support plan. informed consents along with the questionnaire, con rming that they understood the purpose of the study and their participation.

Consent for publication
Not applicable Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
All authors are salaried employees of organizations that are subsidized by the Swiss Paraplegic Foundation.

Funding
This study was part of the project "Informal health care by family members for persons with spinal cord injury in Switzerland", funded by the Swiss National Science Foundation (No. 160158) and Swiss Paraplegic Research. The funding agencies did not have any role in design and conduct of the study; collection, management, and interpretation of the data; or preparation, review, or approval of the manuscript. Description: The le contains the questionnaire, from which data of the present study were retrieved. It refers to an English translation of the original questionnaire.