Screening for cognitive symptoms in dialysis patients with an extended version of Kidney Disease Quality of Life Cognitive Function subscale (KDQOL-CF): a validation study

Abstract

Background: Cognitive impairment and cognitive complaints are highly prevalent in haemodialysis patients and are associated with adverse health outcomes. Currently, there is no established guideline on cognitive screening in this population. Although neuropsychological tests are the gold standard measure of cognition, they are time-consuming and require trained personnel. The Kidney Disease Quality of Life Cognitive Function subscale (KDQOL-CF), a self-administered questionnaire with only three items, may be a feasible alternative for busy renal settings. In this study, we validated an extended version of KDQOL-CF by including an additional memory item (i.e., “How much of the time during the past four weeks did you have memory difficulties?”) to improve its ability to capture memory impairments that are common in dialysis patients but missing in the original scale. Methods: A total of 268 haemodialysis patients treated in 10 dialysis centres in Singapore completed the extended KDQOL-CF and gold standard measures of objective cognition (Montreal Cognitive Assessment) and subjective cognition (Patient’s Assessment of Own Functioning Inventory). Patients also self-reported their functional impairment and treatment nonadherence. Statistical analyses were performed to determine the factor structure and psychometric properties of the extended KDQOL-CF. Receiver operating characteristic curve analyses were conducted to determine the diagnostic ability of the extended KDQOL-CF in identifying objective cognitive impairments and subjective cognitive complaints. Additionally, we examined associations between the extended KDQOL-CF and patients’ self-reported functional impairment and treatment nonadherence. Results: The extended KDQOL-CF can be explained by a one-factor model and has good internal consistency and convergent validity. Receiver operating characteristic curve analysis provided support for the diagnostic accuracy of the extended KDQOL-CF in identifying objective cognitive impairments (area under curve = 60.9%) and subjective cognitive complaints (area under curve = 76.2%). The extended KDQOL-CF also performed better than the original KDQOL-CF in predicting functional impairment and treatment nonadherence in the recruited patients. Conclusions: The extended KDQOL-CF may be used as a first-step cognitive screening tool in dialysis settings to offer a gateway for further diagnostic evaluation and preventive or rehabilitative programs.