Performance indicators for lung cancer surgery in the Netherlands

In the Netherlands, surgery for lung cancer is traditionally performed in low-volume hospitals. To assess the need for centralization, we examined early outcome measures and compared results between hospitals and with other European countries.

Data on patient, tumour and treatment characteristics were retrieved from the Netherlands Cancer Registry. Results were tabulated for 30-day postoperative mortality (POM), major morbidity rate (intrathoracic empyema, bronchopleural fistula or rethoracotomy) and pneumonectomy proportion. Hospital variation was projected using funnel graphs in which the results for individual hospitals are plotted against volume.

The study comprised a series of 9579 patients with primary non-small cell lung cancer, diagnosed from 2005 through 2010 and operated in 79 different hospitals. The POM was 2.7% on average and age, gender, period and type of surgery were determined as prognostic factors. Multivariable analysis did not reveal an association with hospital volume (P = 0.34). The POM was higher for operations on Fridays (4.0%) or during weekends (6.8%). Major morbidity was observed after 8.3% of operations and was more frequent after bilobectomy (11.6%) or right pneumonectomy (22%). The pneumonectomy proportion decreased from 18% in 2005 to 11% in 2010. Funnel plots revealed a limited number of significant outliers, despite combining data over a 6-year period.

Results for the Netherlands were similar to those from other European countries. Hospital volume was not associated with early outcome indicators. Quality assessment at the hospital level remains a major challenge given the low frequency of adverse events and the impediments of chance variation.