Reliability and validity of self-rated Family Burden Scale in families of the patients with anorexia nervosa / 上海交通大学学报（医学版）

ABSTRACT

Objective • To explore the reliability and validity of self-rated Family Burden Scale (FBS) evaluating caregiver burden in families of the patients with anorexia nervosa (AN). Methods • The study included female AN patients (n=103) treated in the Eating Disorder Center of Shanghai Mental Health Center, Shanghai Jiao Tong University School of Medicine from Jul. 2017 to Jul. 2019 and their caregivers (parents, n=148). General demographic information of the patients and their caregivers, the course of illness and body mass index (BMI) of patients, and the time each caregiver spent in caring for the patient per day were recorded. FBS, Beck Depression Inventory (BDI), Beck Anxiety Inventory (BAI), and the MOS item short form health survey (SF-36) were self-rated by the caregivers. One-third of the caregivers were selected by random number method for FBS rating by specialists. The reliability of the scale was evaluated by calculating Cronbach's α coefficient, self-rating and other-rating consistency and the consistency between husband and wife (41 couples). The validity of the scale was evaluated by calculating the correlation of FBS score with the course of disease, patients' BMI, the time spent in caring per day, scores of BAI and BDI, and vitality, social function, role-emotion and mental health domain of SF-36. Results • The Cronbach's α coefficient of FBS was 0.921, the correlation coefficient between self-rating and other-rating by specialists was 0.705 (P=0.000), and the correlation coefficient of consistency score between couples was 0.547 (P=0.000). FBS score showed no correlation with the course of disease, %BMI and the time spent in caring per day. Scores of BAI and BDI, and vitality, social function, role-emotion and mental health domain of SF-36 showed low correlation with FBS score (all P<0.05). Conclusion • Self-rated FBS shows good reliability when used in families of AN patients. The validity is not as ideal, which suggests further revision of the scale.