Background: TNM classification has been the traditional cornerstone of decision-making in treatment choices of lung cancer. In addition to TNM stage and lung function, there is no other objective information to support this decision. Severe comorbidity affects overall survival, and may be an independent prognostic factor. In this study we quantified the effect of comorbidity on stage-specific survival in resected non-small cell lung cancer (NSCLC) patients in a large population-based setting. Methods: Among 3,152 NSCLC patients from the Danish Lung Cancer Registry who underwent surgical resection between 1 January 2005 and 31 December 2010, mortality hazard ratios were calculated during three consecutive time periods following surgery (0-1 month, 1 month - 1 year and >1 year) according to Charlson comorbidity score (CCS 0, 1-2, 3+), ECOG performance status, lung function, age, sex, pathological T and N stage according to the revised 7th edition TNM lung cancer staging system, using Cox proportional hazard modelling. The Kaplan-Meier method was used to describe stage-specific survival according to the Charlson comorbidity score. Results: Increasing severity of comorbidity was independently associated with significantly higher death rates throughout the three periods of follow-up [HR1.18 for CCS 1-2 and 2.06 for CCS 3+ in 0-1 month, 1.13 for CCS 1-2 and 1.57 for CCS 3+ during 1 month - 1 year and 1.14 for CCS 1-2 and 1.84 for CCS 3+ after 1 year]. Stage-specific five-year survival in patients with severe comorbidity (CCS 3+) was significantly lower than in patients without comorbid disease [e.g., 38% (95% CI 23-53%) for pT1 and CCS 3+ vs. 69% (62-75%) for pT1 and CCS 0]. Conclusions: This study shows that severe comorbidity affects survival of NSCLC patients undergoing surgical resection by as much as a single stage increment. As the survival of NSCLC patients after different treatment regiments is mainly based on the TNM classification, further research is necessary to fully identify which patients are most likely to benefit from surgery, but also to provide a basis for developing a better prediction instrument for the survival after combination treatment involving radiotherapy and chemotherapy.