This is a cross-sectional study conducted in 2018 in Iran.
Sample
The study population was all the older people under hemodialysis who referred to medical centers in the western cities of Mazandaran province and the eastern part of Gilan province and their family caregivers. Convenience sampling was applied and the number of samples was determined using the following formula.
Where is the needed sample size, s=24.5 and d=0.2. The needed sample size was stimated to be 341 participants by accepting of sample power and Taking into account the 10 percent probability of sample loss, the number of samples in both groups of hemodialysis elderly and their family caregivers was considered to be 370 people.
Inclusion criteria
For elderly people: being 60 and older, history of hemodialysis for at least one year, not having confirmed mental illness, lack of sensory disturbances such as blindness and deafness and receiving the score of 7 and above at the abbreviated mental test (23).
For caregiver: caring for an older adult under hemodialysis, being the patient’s relative, not caring another elderly and not having physical or mental illnesses (self-report).
Exclusion criteria
Unwillingness for cooperation.
Data collection
Data was collected by the following tools. All tools were completed by samples from both groups:
1- Individual-social information questionnaire related to older people under hemodialysis:
Age, sex, marital status, level of education, number of children, occupation, roommate , frequent hospitalization due to chronic illness, having other chronic diseases, duration of hemodialysis, being the head of family, ability of doing personal activities, level of need for daily care, Member of the Hemodialysis Association, Drug Abuse, Financial Sufficiency and type of Medical Insurance.
2- Individual-social information questionnaire related to family caregivers:
Age, sex, marital status, level of education, medical education, being patient’s relative, length of care for the elderly, chronic illness, care for another patient, being the source of family income, place of residence, housing status, occupation, number of family members, living with the patient.
3- The questionnaire of Elder Abuse to the patients under hemodialysis by Family Caregivers:
This tool has been designed and psychometrically assessed by Mahmoudian et al. (2018). It has an acceptable internal validity (α =0.98) for measuring the construct of elder abuse by family caregivers. It includes 57 items and 7 subscales including psychological misbehavior (6items), authority deprivation (7items) physical misbehaviours (2 items), financial misbehaviours (11 items), being abandoned (4 items), caring neglect (8 items) and emotional misbehavior (19 items). The tool is scored through a 4 point likert scale with the options of never (1), sometimes (2), often (3) and always (4). Scores range between 1 to 228. Receiving the score of 1-75 indicates low severity of abuse, 76-152 average and 153-228 extreme severity of abuse(13).
4- Zarit Burden Interview:
This tool was designed by Zarit et al. (1988). It includes 22 items and three dimensions of role-playing stress, Intra-psychic stress, and Competencies and expectations. Caregiver’s responses are assessed through a five point likert scale (never to always) and the scores range between 0-88. Getting a score of 0 to 20 showed little or no care burden, score of 21 to 40 a moderate, and the score of 41 to 88 a severe care burden (24, 25). The validity of this tool was investigated and confirmed by Talebi et al. (2016). Its reliability is also confirmed by Cronbach's alpha coefficient of 0.86 (26).
5-Instrumental Activities of Daily Living (IADL):
This tool was first developed by Lawton & Brody (1969) and includes 8 dimensions of Using of telephone, Shopping, Meal preparation, Housekeeping, Laundry, Mode of transportation, Medication management and Money management for women. Women are scored on all 8 areas of function; historically, for men, the areas of food preparation, housekeeping, laundering are excluded. Independent (no help=2), needs help (with a little help=1) and dependent (cannot do that=0) based on a 3 point likert scale is used for responding to the questions. A summary score ranges from 0 (low function, dependent) to 8 (high function, independent) for women, and 0 through 5 for men.
For women, the score of zero indicates a completely dependent situation, 1 to 15 needs for assistance, and the score of 16 indicates a completely independent situation. For men, the score of zero indicates a completely dependent situation, 1 to 9 needs for assistance, and the score of 10 indicates a completely independent situation. The reliability of this tool was controlled by Soltan mohammadi et al. (2014). The results showed that the scale has an acceptable validity and the intra-class correlation coefficient of the two stages of the test with r=0.993 is confirmed(27).
Data Analysis
The data obtained from this study were analyzed using LISREL 8.5, SPSS 18 softwares and SEM method. SEM is a general and very powerful multivariate analysis technique of the multivariate regression family. More precisely, it is a general linear model, which allows the researcher to test a set of regression equations simultaneously and to Examine the relationships between different variables at the same time (28). The fitness of the proposed model with data was measured using the Chi-square/degree of freedom ratio (CMIN/DF), Parsimonious Fit Index (PFI), Comparative Fit Index (CFI), Parsimonious Comparative Fit Index (PCFI), Incremental Fin Index (IFI), Goodness of Fit Index (GFI), and Root Mean Square Error of Approximation (RMSEA). P-Value less than 0.05 was considered as significant level.