BACKGROUND: It is generally assumed that patient compliance with adjuvant chemotherapy is superior after video-assisted thoracoscopic surgery compared with open lobectomy for non-small cell lung cancer (NSCLC). The level of evidence for this assumption, however, is limited to single-institution, case-control studies. We used a complete national lung cancer registry. METHODS: For better comparison and reduction of selection bias, we analyzed only patients who underwent standard lobectomy for clinical stage 1 NSCLC and subsequently had unsuspected lymph node metastases discovered at final histopathology. A clinical oncologist, who was blinded to the surgical approach, reviewed all medical oncology charts for types of adjuvant chemotherapy, reasons for not initiating or stopping treatment, number of cycles delivered, and time interval from surgery to initial chemotherapy. RESULTS: During a 6-year period (2007 to 2012), 1,968 patients underwent standard lobectomy for clinical stage 1 NSCLC by video-assisted thoracoscopic surgery (n=990; 50.3%) or thoracotomy (n=978; 49.7%). Unsuspected nodal upstaging was later found in 341 patients (17.3%), and 313 were analyzed: 189 patients (60.4%) received adjuvant chemotherapy and 121 (38.7%) completed all four cycles. Ordinal logistic regression revealed that chemotherapy compliance (none, partial, and full chemotherapy) was significantly reduced by the patient's age (p<0.001) and comorbidity index (p=0.003) but increased with N2 status (p=0.02). No significant difference between video-assisted thoracoscopic surgery and thoracotomy was seen regarding chemotherapy compliance (p=0.17), number of chemotherapy cycles (p=0.60), or time from surgery to chemotherapy (p = 0.41). CONCLUSIONS: Complete national data do not support the widespread assumption that adjuvant chemotherapy compliance is superior after thoracoscopic lobectomy for NSCLC. Instead, significant predictors of chemotherapy compliance are patient's age, comorbidity, and pathologic N status.