Introduction Left untreated, malignant pleural mesothelioma (MPM) is associated with uniformly poor prognosis. Better survival has been reported with surgery-based multimodality therapy, but to date, no trial has demonstrated survival benefit of surgery over other therapies. We evaluated whether cancer-directed surgery influenced survival independently from other predictors in a large population-based dataset. Methods The SEER database was explored from 1973 to 2009 to identify all cases of pathologically-proven MPM. Age, sex, race, year of diagnosis, histology stage, cancer-directed surgery, radiation, and vital status were analyzed. The association between prognostic factors and survival was estimated using Cox regression and propensity matched analysis. Results There were 14,228 patients with pathologic diagnosis of MPM. On multivariable analysis, female gender, younger age, early stage, and treatment with surgery were independent predictors of longer survival. In comparison to no treatment, surgery alone was associated with significant improvement in survival [adjusted hazard ratio (adj HR) 0.64 (0.61-0.67)], but not radiation [adj HR 1.15 (1.08-1.23)]. Surgery and radiation combined had similar survival as surgery alone [adj HR 0.69 (0.64-0.76)]. Results were similar when cases diagnosed between 1973 and 1999 were compared to cases diagnosed between 2000 and 2009. Conclusions Despite developments in surgical and radiation techniques, the prognosis for MPM patients has not improved over the past 4 decades. Cancer-directed surgery is independently associated with better survival, suggesting that multimodal surgery-based therapy can benefit these patients. Further research in adjuvant treatment is necessary to improve prognosis in this challenging disease.