ASSOCIATION BETWEEN PHYSICAL FRAILTY AND INSOMNIA SYMPTOMS AMONG PATIENTS ON HEMODIALYSIS

Abstract Background and Aims Community-dwelling older adults with physical frailty have been suffering from symptoms of insomnia, which have been associated with higher risk of falling and cardiovascular events. Therefore, insomnia symptoms should be assessed for disease management. Most of patients undergoing hemodialysis (HD) have both physical frailty and insomnia symptoms. However, only a few studies reported the association between physical frailty and insomnia symptoms in patients undergoing HD. Therefore, this study aimed to investigate whether insomnia symptoms were one of the useful factors to discriminate physical frailty in patients undergoing HD. Method This study included 231 outpatients undergoing HD three times a week. Physical frailty was measured using the frailty screening index, consisting five components (shrinking, physical function, physical activity, forgetfulness, and exhaustion). Insomnia symptoms were measured using the Athens Insomnia Scale (AIS), a self-administered questionnaire consisting of eight questions, each scored from 0 to 3. Age, sex, body mass index (BMI), HD vintage, comorbidity score, dialysis shift, depressive symptoms, use of sleep medications, usual gait speed, and physical activity (steps/non-HD day) were also investigated. Patients were classified as frailty (the frailty screening index score ≥ 3) and non-frailty (< 3). The analysis of covariance (ANCOVA) was used to compare the AIS scores between frailty and non-frailty. To examine whether the AIS score had the discriminative ability of the frailty complementary to clinical characteristics comparable to the usual gait speed and physical activity, receiver operating characteristic (ROC) curves were constructed, and the area under the curve (AUC) was compared using four models: clinical characteristics (age, sex, BMI, and comorbidity score) only (Model 1), Model 1 plus AIS score (Model 2), Model 1 plus usual gait speed (Model 3), and Model 1 plus physical activity (Model 4). Results The median age was 68 (interquartile range: 59–75) years, 59% were men and 37% had frailty. In ANCOVA, the AIS score in the group of frailty was significantly higher than those of non-frailty even after adjusting age, sex, BMI, HD vintage, comorbidity score, dialysis shift, depressive symptoms, and use of sleep medications (p = 0.004). Through the ROC analysis, Model 2 [AUC: 0.70, 95% confidence intervals (CI): 0.63–0.76], Model 3 (AUC: 0.70, 95%CI: 0.63–0.77), and Model 4 (AUC: 0.69, 95%CI: 0.61–0.76) showed significantly better discriminative ability as compared to Model 1 (AUC: 0.64, 95%CI: 0.56–0.71, all P < 0.05) (Figure). Moreover, AUCs were significantly increased by adding the AIS score to Model 3 (from 0.70 [95%CI: 0.63–0.77] to 0.74 [95%CI: 0.67–0.80], p < 0.05). In addition, AUCs were increased by adding the AIS score to Model 4 (from 0.69 [95%CI: 0.61–0.76] to 0.73 [95%CI: 0.65–0.80], p = 0.07) with a trend toward statistical significance. Conclusion This study examined the association between physical frailty and insomnia symptoms in patients undergoing HD. The AIS score had the discriminative ability of the frailty complementary to clinical characteristics comparable to the usual gait speed and physical activity. Future studies should investigate the causal relationship between physical frailty and insomnia symptoms in patients undergoing HD in detail.