Symptom Clusters Predict Mortality Among Dialysis Patients in Norway: A Prospective Observational Cohort Study

Abstract

ContextPatients with end-stage renal disease on dialysis have reduced survival rates compared with the general population. Symptoms are frequent in dialysis patients, and a symptom cluster is defined as two or more related co-occurring symptoms. ObjectivesThe aim of this study was to explore the associations between symptom clusters and mortality in dialysis patients. MethodsIn a prospective observational cohort study of dialysis patients (n = 301), Kidney Disease and Quality of Life Short Form and Beck Depression Inventory questionnaires were administered. To generate symptom clusters, principal component analysis with varimax rotation was used on 11 kidney-specific self-reported physical symptoms. A Beck Depression Inventory score of 16 or greater was defined as clinically significant depressive symptoms. Physical and mental component summary scores were generated from Short Form-36. Multivariate Cox regression analysis was used for the survival analysis, Kaplan-Meier curves and log-rank statistics were applied to compare survival rates between the groups. ResultsThree different symptom clusters were identified; one included loading of several uremic symptoms. In multivariate analyses and after adjustment for health-related quality of life and depressive symptoms, the worst perceived quartile of the “uremic” symptom cluster independently predicted all-cause mortality (hazard ratio 2.47, 95% CI 1.44–4.22, P = 0.001) compared with the other quartiles during a follow-up period that ranged from four to 52 months. The two other symptom clusters (“neuromuscular” and “skin”) or the individual symptoms did not predict mortality. ConclusionClustering of uremic symptoms predicted mortality. Assessing co-occurring symptoms rather than single symptoms may help to identify dialysis patients at high risk for mortality.