Introduction To evaluate the clinical and dosimetric parameters that increase the risk of radiation pneumonitis (RP) in locally advanced non-small cell lung cancer (NSCLC) patients treated with concomitant chemoradiotherapy of nationwide multicentric data analysis. Methods All data of 268 patients who underwent definitive chemoradiotherapy were retrospectively collected from eight institutes participating in this study. Patient, tumor and treatment-related factors and dosimetric parameters were analyzed for grade >= 2 RP. The toxicity scoring system of The Radiation Therapy Oncology Group used for grading the severity of pneumonitis. A relationship with the risk of RP with potential predictive factors were evaluated by univariate and multivariate analyses. A recursive partition analysis (RPA) was applied to stratify patients according to the risk of developing RP. Results There were 90 (33.6%) patients who had grade >= 2 RP. The median time to pneumonitis after treatment was 4 months (range:1-6 months). In univariate analysis, diabetes mellitus (DM), use of cisplatin/etoposide, total and daily radiotherapy (RT) fraction dose, the planning target volume (PTV) size, mean lung dose, V5, V10 and RT technique were associated with the development of pneumonitis. In multivariate analysis, only DM (P = 0.008) was found to be independent risk factors for RP. According to RPA, the risk of developing RP was highest in patients with DM. Conclusions In our study, besides the known dosimetric factors, DM was found to be the most important risk factor causing RP development in multivariate analysis and RPA. The risk is tripled compared to patients without DM.