The Development Of Burden Evaluation System For Family Caregivers Of Chronic Diseases

ObjectiveTo analyse the characteristics of family caregivers of chronic diseases in both rural and urban areas and develop a multi-dimensional caregiver burden evaluation system on the basis of a short version of ZBI scale which was efficient to evaluate and compare direct caregiver burden in Chinese background.MethodsA cluster sampling selecting 794 primary family caregivers of different types of chronic patients from both urban and rural communities in Hubei province were interviewed. The family caregiver'direct burden, depression,quality of life of family caregivers were evaluated by ZBI (Zarit Caregiver Burden Interview), CES-D(Center for Epidemiological Studies Depression Scale), WHOQOL-BREF respectively.Other aspect of burden such as self-care, satisfaction, competence, safety were evaluated by self-made items. The functions of daily activities of chronic illness patients were assessed by ADL (Activities of Daily Living).The items of the short version selected were these had high item-total relations across all kinds of diseases, and then the items of other dimensions of caregiver burden entered to the evaluation system to build a multi-dimensional family caregiver burden evaluation system and its reliability and validity was analyzed.Results1. The character of family caregivers of chronic diseasesThere were more female caregivers than males with an average age of 55.40±11.71 years and more than half of the caregivers were spouses of the patients, followed by their children. In contrast to rural caregivers, urban caregivers were more prone to be female,and with higher income (P<0.0001).Most of the caregivers were the spouses of patients in both urban and rural areas but among the urban caregivers there were more children who took care of the patients. There were no difference between urban and rural caregivers in age, education level, marital status, whether living with the patients (P>0.05).The ZBI score of urban caregiver burden was 26.41which was significantly lower than the 41.87 of rural caregivers and the rural-urban difference also existed in different chronic disease (P<0.05).In addition, the ZBI of caregivers of different diseases had significant difference (P<0.001)among which caregivers of mental illness had greatest ZBI scores in both urban and rural. The characteristics of caregivers integrated by cluster analysis were divided into five categories, which showed the urban caregivers generally had lower direct caregiver burden, which was much higher in rural areas.2. The short version ZBI scaleSeven items (2,3,4,10,11,17,19) retained in the short version ZBI (ZBI₇). Compared with other simplified version, the Chronbach's a coefficient of ZBI₇ was 0.90, only slightly lower than 0.91of ZBI₁2, and were higher than 0.84 in various chronic diseases which indicated its good internal consistency. The correlation of short version with the total scale correlation coefficients were all ideal but ZBI₇ had a correlation coefficients as high as 0.95 in the whole sample and 0.90 to 0.96 in different diseases which proved it better represents the information of the original scale. The correlations of ZBI₇ with the ADL and its associated subscale were also highly consistent with ZBI₂2 and it divided the direct caregivers'burden into three groups of low, median, high which indicated it had a good discrimination of direct burden of different types of diseases just as the original scale. On the other side, it had the highest effect size in distinguishing different level of direct caregiver burden. ROC curve was used to compare the specificity and sensitivity of different short version ZBI and found that area under the curve ZBI₇ were significantly higher than other versions except ZBI₁2 but the relatively fewer items of ZBI₇ indicated it's an acceptable scale with good reliability, validity and feasibility in practical use.3. The development family caregivers burden evaluation system of chronic diseaseThe ZBI₇ and the 12 items from the dimensions of self-care, satisfaction, competence, and safety formed the caregiver burden assessment system. The exploratory factor analysis resulted in a scale with 5 factors structure including 10 items. The cumulative contribution rate was 58.01% and the structure was consistent with the theoretical proposed dimensions. The Chronbach's a of the scale was 0.74 and all the subscales were significantly related to the total scale (P<0.01). The caregiver burden evaluation system had a significant correlation with CES-D(r=0.63), WHOQOL (r=-0.51) and correlated with psychological dimensions most. Among the sub-scales, the direct burden and self-care were significantly correlated the CES-D, WHOQOL and all its subscales and ADL, but the satisfaction was only significant correlated physical, social dimensions of WHOQOL, the competence correlated with the CES-D, physical and psychological dimensions of WHOQOL significantly. The sub-scale scores of different diseases were different significantly except self-care, and the caregivers of mental illness had least self-care behaviors, competence, the lowest satisfaction, and the heaviest burden. Caregivers with different demographic characteristics and needs of caregiving related services had different correlation with the sub-scales of the caregiver evaluation system.ConclusionThe direct burden measured by ZBI of family caregivers of chronic diseases in rural area was generally higher than the urban caregivers. The short version ZBI scale (ZBI₇) represented the original scale well. The evaluation system of caregiver burden including 10 items and 5 dimensions of direct burden, self-care, satisfaction, competence and safety had an ideal reliability and validity in caregiver burden appraisal of different types of chronic diseases, and can be used for evaluation and screen the burden of family caregivers of chronic disease in community.