Background: Current TNM non-small cell lung cancer (NSCLC) staging uses only the anatomic location of lymph nodes to define N status. Several other cancer staging systems have found lymph node ratio (LNR) - the number of positive lymph nodes/total lymph nodes resected - to be a better predictor of survival after resection. The purpose of this study is to evaluate LNR as a predictor of recurrence and survival after R0 resection for NSCLC. Methods: A total of 1,143 consecutive patients underwent R0 resection for NSCLC between 1999 and 2008 at a high-volume single institution with 26% (n = 302) having N1 and N2 disease. The primary endpoints of the study were long-term survival and recurrence as a function of LNR. Cox proportional hazard models and Kaplan-Meier survival analyses were utilized to assess associations between LNR, N status, and pathologic stage with survival and recurrence after lung cancer resection. Results: Median follow-up was 44 months and was complete in 97% of patients. Nodal status of patients in this study was as follows: N0 disease, 73.5%; N1 disease, 18.7%; and N2 disease, 7.8%. There were 132 recurrences in patients with nodal disease (43.7%). The pathologic stage of patients in the study was as follows: stage IIA, 47%; stage IIB, 17%; stage IIIA, 35%; and stage IIIB, 1%. Mean total number of lymph nodes sampled was 11.1 ± 6.0 and mean number of positive lymph nodes 2.4 ± 2.0. Upon statistical modeling, LNR was found to be independently associated with decreased survival after resection for NSCLC (hazard ratio 2.63, confidence interval: 1.41 to 4.91, p = 0.002). Conclusions: In patients undergoing resection for NSCLC, increasing lymph node ratio is independently associated with decreased survival and decreased time to recurrence after R0 resection.