BACKGROUND Robotic and video-assisted thoracoscopic surgery (VATS) approaches for lung resection are associated with decreased inpatient opioid use compared with open surgery. Whether these approaches affect outpatient persistent opioid use remains unknown. METHODS Non-small cell lung cancer patients aged 66 years or more who underwent lung resection between 2008 and 2017 were identified from the Surveillance, Epidemiology, and End Results-Medicare database. Persistent opioid use was defined as filling an opioid prescription 3 to 6 months after lung resection. Adjusted analyses were performed to evaluate surgical approach and persistent opioid use. RESULTS We identified 19,673 patients: 7479 (38%) underwent open surgery, 10,388 (52.8%) VATS, and 1806 (9.2%) robotic surgery. Persistent opioid use was 38% in the entire cohort, including 27% of opioid naïve patients, and highest after open surgery (42.5%), followed by VATS (35.3%) and robotic (33.1%, P < .001). In multivariable analyses, robotic (odds ratio 0.84; 95% CI, 0.72-0.98; P = .028) and VATS (odds ratio 0.87; 95% CI, 0.79-0.95; P = .003) approaches were both associated with decreased persistent opioid use compared with open surgery in opioid naïve patients. At 12 months, patients resected using a robotic approach had the lowest oral morphine equivalent per month compared with VATS (133 vs 160, P < .001) and open surgery (133 vs 200, P < .001). Among chronic opioid patients, surgical approach was not associated with postoperative opioid use. CONCLUSIONS Persistent opioid use after lung resection is common. Both robotic and VATS approaches were associated with decreased persistent opioid use compared with open surgery among opioid naïve patients. Whether a robotic approach yields additional long-term advantages over VATS warrants further investigation.