The SF36 as an outcome measure of services for end stage renal failure

Objective-To evaluate the use of the short form 36 (SF36) as a measure of h ealth related quality of life of patients with end stage renal failure, doc ument the results, and investigate factors, including mode of treatment, wh ich may influence it. Design-Cross sectional survey of patients with end stage renal failure, wit h the standard United Kingdom version of the SF36 supplemented by specific questions for end stage renal failure. Setting-A teaching hospital renal unit. Subjects and methods-660 patients treated at the Sheffield Kidney Institute by haemodialysis, peritoneal dialysis, and transplantation. Internal consi stency, percentage of maximal or minimal responses, SF36 scores, effect siz es, correlations between independent predictor variables and individual dim ension scores of the SF36. Multiple regression analysis of the SF36 scores for the physical functioning, vitality, and mental health dimensions agains t treatment, age, risk (comorbidity) score, and other independent variables . Results-A high response rate was achieved. Internal consistency was good. T here were no floor or ceiling effects other than for the two "role" dimensi ons. Overall health related quality of Life was poor compared with the gene ral population. Having a functioning transplant was a significant predictor of higher score in the three dimensions (physical functioning, vitality, a nd mental health) for which multiple regression models were constructed. Ag e, sex, comorbidity, duration of treatment, level of social and emotional s upport, household numbers, and hospital dialysis were also (variably) signi ficant predictors. Conclusions-The SF36 is a practical and consistent questionnaire in this co ntext, and there is evidence to support its construct validity. Overall the health related quality of life of these patients is poor, although transpl antation is associated with higher scores independently of the effect of ag e and comorbidity. Age, comorbidity, and sex are also predictive of the sco res attained in the three dimensions studied. Further studies are required to ascertain whether altering those predictor variables which are under the influence of professional carers is associated with changes in health rela ted quality of-life, and thus confirm the value of this outcome as a measur e of quality of care.