Evaluation of surgical practice in the treatment of lung cancer in France from the PMSI national database

Background. - In recent years, improving the quality of care has been a concern for health professionals in France, through the certification of institutions, accreditation and continuous professional development. Evaluation of these different measures has rarely been carried out. The objective of the study was to evaluate the quality of surgical management of lung cancer in different regions using hospital mortality as an indicator. Method. - From the national database of the Program of Medical Information Systems (PMSI), data on all patients who had undergone surgery for lung cancer were extracted as well as the characteristics of the centers. The main outcome criterion was hospital mortality. Logistic models allowed an estimation of the risk standardized mortality rate for each center. Results. - From January 1, 2015 to December 31, 2015, 10,675 patients underwent surgery for lung cancer in 158 French centers. The hospital mortality rate was 3.43% (n=366). Thirty-nine facilities (25%) performed fewer than 15 pulmonary resections. The minimum activity volume was a single pulmonary resection during the year and the maximum was 300 interventions with a coefficient of variation estimated at 147%. Hospital mortality ranged from 0 % to 50% depending on the entries with a coefficient of variation of 112%. For some regions, it is possible to count up to 5 centers per million inhabitants (Languedoc-Roussillon) or 4 centers per million inhabitants (Limousin, Pays-de-Loire). The majority of regions had 3 centers per million inhabitants. Eleven regions have no centers with a standardized mortality rate below 3%. Five regions (Languedoc-Roussillon, Pays-de-Loire, Aquitaine, Brittany and Provence Alpes Cote d'Azur) have at least two centers with a risk standardized rate of mortality above 4%. Among the academic centers, 20% have a risk standardized mortality rate of less than 3%. Among the centers with a risk standardized rate of mortality < 3%, 20% performed more than 39 pulmonary resections, 7% between 39 and 15 procedures and 0% for centers with < 15 interventions. Conclusion. - This work confirms that hospital volume is one of the components of quality of care. The number of centers should be adapted to the actual needs of the population in order to enable patients to access effective services. (C) 2018 SPLF. Published by Elsevier Masson SAS. All rights reserved.