Development And Evaluation Of Quality Of Life Scale For Patients With Bladder Cancer

[abstract]Objective To develop a Chinese version scale of quality of life (QOL) instrument and to sinicize EORTC QLQ-BLM30 for forming homologous Chinese version, which can be used to appraise the quality of life for patients with bladder cancer, and evaluate the reliability, validity, responsibility and criterion. To explore the influence factors of bladder cancer patients' QOL, in order to provide the reference of therapy.Methods QLICP-GM(V1.0) was usd as general module. Based on the definition of quality of life from WHO, the research team referred to the accepted cancer special scales, related literature and consulted the experts'group (including the clinical experts on bladder cancer, Psychology and sociology experts), we drew off the items. Then we adopted the method of qualitative interview combining significance score to form a questionnaire. Chinese version's EORTC QLQ-BLM30 was developed by translation and back-translation and cultural adaptation. The validity, reliability and responsibility of the questionnaire were evaluated by a survey of QOL among 120 bladder cancer in-patients. The statistical methods used in this paper included statistical descriptions, Pearson Linear Correlation, Exploratory Factor Analysis(EFA),Structural Equation Modeling(SEM), Hierarchical Cluster Analysis and Paired-samples t Test, Chi-square Test, one-way ANOVA, Multivariate Regression.Results (1) One 27-items specific module for bladder cancer was constructed. There were five factors:urinary symptoms, special physical functioning, special psychology functioning, difficult of urine bag, sexual problems and ureter use problem.(2) QLICP-BL Reliability:The coefficients of internal consistency Cronbach's Alpha of all domains were greater than 0.7 except social function domain.The split-half were greater than 0.5 except social function domain. The results of test-retest reliability indicate that reliability of all domains correlation coefficients of first and second were greater than 0.75, scores of all domains correlation coefficients of first and second had no statistical.The reliability of the questionnaire was good.(3) QLICP-BL Validity:Because the item pool were discussed by panels and expert in urology and it could reflect the definition of health related QOL proposed by WHO, the QLICP-BL had good content validity. The analysis for the special module of QLICP-BL follows. Most of the correlation coefficients between item and its domain were greater than the correlation coefficients between item and other domains. Four components were explored by using factor analysis with varimax rotation,4 components which reflected content of urinary symptoms, special physical functioning, difficult of urine bag and sexual problems of special module were identified and explain 77.89% of the variance. Deleting item 13, Structural Equation Modeling establish modeling, Chi-square=89.64, Degrees of freedom=62, RMSEA=0.061, NNFI(TLI)=0.971, CFI=0.977, SRMR=0.0463.The mentioned three methods indicate the special module of QLICP-BL had good structure validity. We used the score of a domain which reflected whole quality of life in EORTC QLQ-C30(V3.0) as criterion, the correlation coefficients were greater than 0.6.The result showed good criterion validity of the instrument.(4) QLICP-BL Responsiveness:The differences of each domain between first and third time point had statistical significance. So the scale had good responsibility.(5) EORTC QLQ-BLM30:Method is same as that of QLICP-BL. The coefficients of internal consistency Cronbach's Alpha and The split-half of all domains were greater than 0.7 except body image domain. The results of test-retest reliability indicate that reliability of all domains correlation coefficients of first and second were greater than 0.9, scores of all domains correlation coefficients of first and second had no statistical. The reliability of the questionnaire was good. Four components were explored by using factor analysis with varimax rotation,4 components which reflected content of urinary symptom, future perspective, abdominal bloating and flatulence body image of EORTC QLQ-BLM30 were identified and explain 73.74% of the variance. Structural Equation Modeling establish modeling was not fitted. To sum up, Reliability, validity and responsiveness of EORTC QLQ-BLM30 was good.(6) About 15-30 minutes were spent in completing the instrument and the completed rate over 93% demonstrated that good feasibility of the instrument.(7) The multivariate regression analysis of influence factors of QOL show that age,education,occupation surgical methods, neutrophil, nation, urea, sex, tumour grade affect different domains of QOL of bladder cancer.Conclusion The QLICP-BL and EORTC QLQ-BLM30 were valid, reliable, sensitive and could be used to assess quality of patients with bladder cancer. The Quality of life scores did not have the close linear correlation with objiectivity items. The influence factors of patients'Quality of life were age, education, occupation, surgical methods, neutrophil, nation, urea, sex, tumour grade. However, there were still other influence factors.