FO032THE PHYSICAL ACTIVITY COMPONENT OF THE SF36 IS THE QoL DOMAIN HOLDING THE HIGHEST PREDICTIVE POWER FOR SURVIVAL IN THE DIALYSIS POPULATION

Abstract

Introduction and Aims: Established instruments of Health Related Quality of life (QoL) incorporate various QoL domains which capture emotional and physical well-being. These instruments provide information potentially useful to guide clinical decisions in patients with chronic kidney failure on dialysis (stage G5D). However, due to time constraints, QoL records are rarely registered in clinical practice in this population. Detailed studies of the link of the domains of QoL with survival may be useful to further simplify these instruments and to explore their relevance for prognosis. Methods: We systematically measured QoL by the Short Form 36 (Rand corporation) instrument validated for Italian people in a cohort study including nine-hundred and eighty-one patients on chronic dialysis. These patients represented about the 75% of the dialysis population in a public regional dialysis network. The predictive power for survival of the eight domains (physical functioning, role limitation due to physical health, role limitation due to emotional problems, energy/fatigue, emotional well-being, social functioning, pain, general health) composing the SF36 was tested in unadjusted analyses and in statistical models incorporating established predictors of survival in this population i.e. age, sex, smoking, diabetes, systolic BP, serum cholesterol, previous cardiovascular events, serum albumin, CRP, Hb, serum phosphate and dialysis vintage. Results: All the eight QoL domains predicted survival in unadjusted analyses. All but two of these domains (role limitation due to physical health and due to emotional problems), maintained an independent predictive power for survival in a statistical model including the full set of established death predictors in dialysis patients listed above. These predictors explained 28% of the variability in survival. Physical functioning was by far the QoL domain adding the highest explanatory power to the model based on these factors (added explanatory power 6%) and such an explanatory power was higher than that of age (4%), which ranked as the second predictor in the same model. Physical functioning improved model calibration by the 63% and, importantly so, the reclassification power by the 9% (P=0.003).The strength (R2) of the association of the variability in the SF36 domains with physical functioning ranged from 36% (Energy/fatigue) to 11%(role limitation, emotional) and collectively these domains explained only the 44% of the variability in physical functioning documenting that this component conveys unique information for the interpretation of QoL. Conclusions: The physical function component of the SF36 is the QoL domain holding the highest predictive power for mortality among the eight domains of this instrument and the top rank predictor among all risk factors currently applied to model survival in the dialysis population. The discrimination power and reclassification ability of physical functioning is of degree relevant for clinical practice. Physical functioning by the SF36 can be reliably obtained just in 2 minutes. Such a measure has potential for refining prognosis and informing exercise programs in the dialysis population.