Background: Different approaches for an effective quality management are funded by the Ministry of health to verify, and if necessary to optimize, the quality of health care using the tracer diagnosis breast, rectal and lung cancer in 8 regions in Germany. The aim of the study is to develop a model for description and support of high quality health care for cancer patients. The conception, initial findings and inter-hospital variations are shown for some aspects of the primary therapy of rectal cancer in the region of Munich (population 2.3 million).
Patients and methods: The field study is a population based cohort study. The recruitment phase started in April 1996 and finished March 1998. Established documentation sheets of the tumor registry along with original reports (reports of the doctors, pathologists etc.) are used for documentation.
Results: 26 surgical departments, 12 pathology departments, 9 radiotherapy departments and about 300 general practitioners have documented for this study. So far a sample of 809 have been analyzed. After an almost complete documentation the crude incidence will be about 23/100.000. 57% of all patients were men and 43% were women. 765 patients have been operated, 44 not. PT-categories (for the operated patients) are distributed as follows: pTis and pT1 10.7%, pT2 25%, pT3 54.5%, pT4 9.8%. Distribution of tumor localization: < 4 cm 6.7%, 4 bis < 8 cm 36.2%, 8 bis < 12 cm 30.8%, > or = 12 cm 26.3%. Local excision was performed in 4%, sphincter conserving therapy (SCT) in 71.5% (protective stoma: 23.7%), not sphincter conserving therapy (nSCT) in 18.4% and a palliative operation in 6%. The large surgical departments (> 25 patients per year) treated 53.8% of all patients. The proportion of older patients is significant higher in the smaller departments (< 10 patients per year). There are no differences in pTNM-distribution. No variations exist in the operation method between the department groups. More or less varying inter-hospital differences exist between the individual departments for all epidemiological and therapeutical parameters.
Conclusions: Variations of medical care exist more between the individual hospitals than between department groups, divided by workload. To support the quality of health care a feedback of the results has to be available for each physician and each department.