Methods
Participants
Out of 1042 older patients who met the inclusion criteria of the initial studies, 241 had no information on family caregiving, and 139 were excluded because of low cognitive status, resulting in a final sample of 662 participants. The functional status before hospitalization, number of children, education, perceived economic status, family status, level of depression and anxiety, gender, and length of hospital stay of the excluded participants did not differ significantly from those of participants retained in the final sample. However, the excluded individuals were older (p<.05), had lower cognitive status (p<.05), and a higher level of comorbidity (p<.05).
Measures
Informal support was assessed with the ICHOA, the Informal Caregiving for Hospitalized Older Adults scale [4]. This measure consists of four subscales: instrumental support (e.g. help with eating or grooming), supervision of instrumental support (e.g. making sure staff help with eating), ensuring/supervision of medical care (e.g. keeping an eye on medical care or discussing condition with medical staff), and psychological support (e.g. comforting patients when sad or helping them get used to hospital). All subscales include 3-6 items evaluated on a Likert-type response scale from 1 (did not receive any help) to 5 (received help all the time). For this study, the reliability of scale items was good (from α=.83 to α=.88). Length of family caregivers' visits during the day was self-reported by patients and measured in hours. Family caregiver stay during the nightwas dichotomized as 0 (did not stay with the patient during the night) and 1 (accompanied during the night for at least four hours).
Ethno-Cultural Groups
Participants were divided into three ethno-cultural groups according to their self-identified native language (Hebrew, Arabic, or Russian), religion (Jews, Christian, or Muslim), country of birth (Israel, USSR, or other countries), and year of immigration (before or after 1989, with the former considered veteran immigrants). The three groups were as follows: (1) "Jews", Israeli-born with Hebrew as a mother language and/or veteran immigrants arriving before 1989; (2) "Arabs", Israeli-born with Arabic as a mother language and Christian or Muslim religion; (3) “FSU immigrants”, immigrants to Israel from FSU with Russian as their native language.
Background Characteristics
Sociodemographic variables and variables related to functional, cognitive, psychological, and health status were included in the data analysis as potential confounders. Functional status on admission was assessed using the 11-item Modified Barthel Index (MBI) [35], consisting of individuals’ self-assessment of their independence performing basic activities of daily living (ADLs). Cognitive status was measured using the Pfeiffer Short Portable Mental Status Questionnaire (SPMSQ) [34]. In this questionnaire, total scores range from 0 to 10; higher scores indicate better cognitive status. Level of anxiety was assessed using the Short Anxiety Screening Test (SAST) [36] a 10-item, 4-point Likert-scale questionnaire; anxiety is defined as receiving 24 points or more. Level of depression was measured using the 10-item Short Zung Interviewer Assisted Depression Rating Scale (Short Zung IDS) [37]. Responses are rated on a 4-point scale with the total score recoded from 0 to 75; depression is defined as scoring 70 points or more on the scale. Severity of chronic health condition was assessed using Charlson’s comorbidity index; the index weights 20 health conditions and their severity on a scale from 1 to 6 [38]. Sociodemographic variables: age, gender, family status, number of children and economic status were also included. Economic status was estimated by self-report on a scale from 1 (worst) to 5 (best).
Procedure
Functional and cognitive status, as well as background characteristics, were assessed within the first 24 hours of hospital admission. In-hospital informal support was assessed via in-person interviews at the end of hospitalization. Data on chronic morbidity and length of stay were retrieved from the hospitals’ electronic medical records.
Results
Participants’ Characteristics
The mean age of participants was 78.1 (SD = 5.7); half were male (51.4%, N = 340), and half were married (54.7%, N = 362). The mean number of children was 2.8 (SD = 2.1). Average cognitive and functional statuses were relatively high, indicating a mostly independent sample (cognitive status, M = 8.7, SD = 1.5; functional status before hospitalization, M = 91.7, SD = 15.4). About one fifth screened positively for anxiety (16.8%, N = 110) and depression (20%, N = 131). Subjective economic status of most participants was "like others" (42.1%, N=270) or higher (39.3%, N=252). Average length of hospital stay was 6.3 days (SD = 5.2). Average comorbidity score was 2.5 (SD = 2.1) (see Table 1.1).
Preliminary Analysis
All background characteristics (age, gender, family status, number of children, and functional, cognitive, psychological, and health status) were significantly related to at least one of the study variables and thus were included as control variables.
Ethno-Cultural Sample Characteristics
The majority of participants were non- or veteran immigrant Jews (63.5%, N=420); more than a quarter were “FSU immigrants” (30.5%, N=202); and 6% (N=40) were “Arabs”.
Informal Caregiving during Hospitalization
Hours of day visits. Almost all participants (97%, N= 622) were accompanied by family caregivers during their hospitalization. The average length of day visits was 4.7 hours (SD=3.5).
Night stay. Only 7.9 % of participants (N=52) were accompanied by their family caregivers during the night (“stay” was defined as at least four hours).
Type of support. Less than half of the participants (34.3%) received any instrumental support from their family caregivers during hospitalization, and only 11.6% reported that their family caregivers supervised instrumental support from the hospital staff. Psychological support was the most prevalent: 94.5% of hospitalized elders reported receiving this kind of support from family caregivers during hospitalization. The majority (83.1%) reported their family caregivers ensured/supervised medical care.
Ethno-Cultural Differences in Informal Caregiving
Hours of day visits. UNIANCOVA analyses controlling all covariates revealed significant differences between the ethno-cultural groups (F(2, 615)=13.86, p<.0005). Bonferroni post-hoc analysis found the “FSU immigrants” had the shortest visits (estimated average 3.6 hours per day), shorter than “Jews” (5.2 hours per day, p<.0005) and “Arabs” (5.7 hours per day, p=.007) (Table 1.2).
Night stay. Logistic regression analysis showed "Arabs" were more likely to be accompanied by family caregivers during night hours than “Jews” (OR=3.12; CI 95% (1.03-9.41); p=.044) or “FSU immigrants” (OR=7.32; CI 95% (1.62-33.0); p=.01).
Differences in kinds of support. Linear regressions showed “FSU immigrants” received significantly more psychological support from family caregivers than "Jews" (Beta = .11, t (628) = 2.73, p = .007). There were no significant differences between the three ethno-cultural groups in other kinds of informal support (Table 1.2).