Search

This study is intended to describe the cancer-related hospitalisations in the thirty French University and Regional Hospitals (CHR&U). The analysis is carried out on the years 2001 and 2002 databases and uses a relevant PMSI (Medicalisation Program Information System, the French DRG) data analysis program, which is an adapted version of the DAC (Cancer Activity Description) analysis program. Cases are selected from cancer-related diagnoses and procedures. The results provided have been classified according to paediatric and adult populations, to the organ anatomical localisations and to the different types of care : one-day hospitalisations or hospitalisations of more than one day, medical or surgical care, cares which include chemotherapy, radiation therapy or palliative procedures. Cancer-related hospitalisations amount to almost 25 % of the total CHR&U hospitalisations and add up to over a million stays per year. One-day hospitalisations amount to 25% of the cancer-related stays and essentially consist in radiation therapy (50% within the adult category) and for chemotherapy (30%). 26% of the hospitalisations of more than one day are surgical. This study is a first descriptive analysis of cancer-related hospital activity in CHR&U. The next stage will see this analysis applied to patients, using the anonymous patient identification number contained in the PMSI coding.

Recently, to answer patients, caregivers and professionals needs, the "Plan Cancer" has been presented by the French Government. This plan is intended to improve quality of care in cancer patients and finally, patients' survival and quality of life. This planned strategy stresses the importance of organized interactions between hospitals and between the various health professionals. Measuring the number of patients with cancer and the activity related to cancer in large networks of multidisciplinary hospitals has became a real challenge in France for organizational, quality of care and economic reasons. Many University Hospitals in France have chosen to face this question by using the French DRG based information system called PMSI. It allows estimating the proportion of hospital stays concerned by cancers that are identified with algorithms based on ICD 10. However, French databases of hospital discharges do not allow patients identification. We collected data on hospital stays and patients in a subset of an organized network focused on cancer care and composed of 55 public or private hospitals in the Rhone-Alpes area. We used these data to estimate the number of patients who had been hospitalized within the network in 2000. Approximately 110,000 hospital stays were related with a diagnostic of cancer, corresponding to a number of patients within a range of 30345 to 35700. In absence of communicating files between hospitals, claims databases are an interesting source of information for cancer burden. The recent implementation of a procedure allowing the linkage of data concerning each patient should permit better estimates in the future. The main limitation will remain the possibility of a hospital to participate to more than one network.

In an era where health care expenditure control is a necessity, weighing a new treatment's cost against its added benefits is of crucial importance. These medico-economic analyses may be planned for the evaluation of new molecules in their daily use. Different methods may be used. The medico-economic evaluation of a drug can be performed with the use of clinical studies or modelization techniques. "Clinical studies" include the following methods: randomized trials, so-called "experimental", non-randomized studies, and observational studies. This classification follows a downward evolution of both the investigator's intervention power and the level of proof. Finally, modelization is becoming a commonly used tool. Randomized trials are the "gold standard" but the other, more pragmatic study types, providing a lesser degree of proof can complete the evaluation of a new molecule. One must find the best compromise between the study's objectives and the study type.

The increasing costs of care make it important to identify those strategies of greatest value from both an effectiveness and cost perspective. Economic analysis is characterized by a simultaneous consideration of alternatives costs and outcomes, and can provide useful data for managerial decision making. In this paper, methods of economic evaluations in general and in cancer in particular is reviewed. In cancer treatment, preventive, curative or palliative strategies can be concerned. Economic evaluation have become increasingly important in oncology because of the proliferation of expensive new treatments. Furthermore, considering quality of life effects is particularly important in oncology, where many treatments obtain modest improvements in response or survival. Quality of life measurements are also reviewed.

Measuring the burden of cancer activity in big teaching multidisciplinary hospitals becomes a real challenge in France for organizational, quality of care and economic reasons. At the "Hospices Civils de Lyon" University Hospital in France, we have chosen to face this question by using the french DRG based information system called PMSI. It allows to identify hospital stays related to cancer care through the use of an algorithm based on selected ICD 10 codes for hospital stay. We then estimate the proportion of patients concerned by cancer and classify hospital stays and patients according to various epidemiological and clinical parameters. The results for the year 1999 taken as an example showed that 43,883 out of 293,827 (15%) of hospital stays were related with a diagnostic of cancer, corresponding to 12,777 different patients. The concordance study on 154 patients between the data from the electronic PMSI files and medical paper records showed a 97% concordance for cancer diagnosis and 93% for the precise type of tumor. In absence of hospital based cancer registries, the PMSI data base gives an accurate source of information for cancer burden and provides many potential applications in defining hospital policies for cancer management and resource allocations.

The therapeutic interest of gemcitabine both as a single drug and in combination is now well known and applied in various malignant diseases. However, it would be worth being further explored in two tumor models for which its efficacy has been suspected and which have been somehow "abandoned" by modern oncology after the demonstration of their high sensitivity to platinum derivatives, namely epidermoid tumours of the upper aero-digestive tract and uterine cervix. Combination studies should be encouraged in this setting.

Primary peritoneal tumors are rare and yet not well-known malignancies. We report here the clinical cases of 8 patients who were referred to our institution with a primary peritoneal tumor, 4 serous carcinomas (PSC) and 4 malignant mesotheliomas (PMM). Diagnosis was available in all cases and reviewed by two pathologists. A battery of various immunomarkers contributed to confirm the morphological diagnosis. All patients underwent initial surgery with or without optimal debulking. All of them received a platinum-based systemic chemotherapy. Intraperitoneal treatment with cisplatin and hyperthermia was performed in two cases. Four patients are dead (3 PSC, 1 PMM), two are alive (one in partial remission (1 PSC) and one is having treatment (1 PMM)), one is in complete remission (1 PMM) and one is lost to follow up (1 MMP). PSC and PMM are difficult to distinguish since they share the embryonic cOElomic epithelium as a common origin. The clinical presentations are similar and usually include the detection of a pelvic mass and/or diffuse peritoneal carcinomatosis with or without malignant ascites. Therefore, the accuracy and reliability of the initial histological examination is of crucial importance. Further attempts will be required to better identify the most active therapeutic post-surgical combinations, especially for PMM.