Treatment-Related Morbidity and Hospital League Tables: Experience from a National Audit of Radiotherapy-Induced Morbidity in Cervical Carcinoma

• Published in: Clinical oncology (Royal College of Radiologists (Great Britain)) Vol. 14; no. 1; pp. 40 - 42
• Main Authors: Denton, A.S., Bond, S.J., Matthews, S., Bentzen, S.M., Maher, E.J.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Ltd 01.02.2002; Elsevier
• Subjects: Biological and medical sciences; Epidemiology; Female; Female genital diseases; Gynecology. Andrology. Obstetrics; Hospitals; Humans; Medical Audit; Medical sciences; Radiation Injuries - epidemiology; Radiation Injuries - etiology; Retrospective Studies; Tumors; United Kingdom; Uterine Cervical Neoplasms - radiotherapy; United Kingdom; Europe; Human; Methodology; Treatment efficiency; Relational database; Uterine cervix; Malignant tumor; Radiotherapy; Epidemiology; Morbidity; Female genital diseases; Audit; Female; Medium term; Uterine cervix diseases; Survival curve
• ISSN: 0936-6555; 1433-2981
• DOI: 10.1053/clon.2001.0022

Data are now available from a U.K. audit of survival and late morbidity following curative radiotherapy for cancer of the cervix treated in 1993. The complication rate per centre ranges from 0 to 67%. Although the frequency of complications following curative radiotherapy for cancer of the cervix might be considered to be an indicator of clinical performance, variation in treatment outcomes can be explained by sampling variability rather than real differences in quality of care. In the present study we have asked the question: could the disparity in complication rates between centres be no more or less than would be expected by chance? Our analysis suggests that this is the case, and for this reason it would be premature to use such outcome data to produce league tables or to assess institutional differences. Thus, ranking centres according to complication rate would not be valid, as the differences in rates observed are probably not significantly different from the national average. It is important that audit data are not used inappropriately and this analysis further highlights the need for reliable prospective collection of clinical information and the importance of considering sampling variability in interpreting the results of such studies.