Characterizing Opioid Prescribing Trends of Medical Oncologists From 2013 to 2019: Analysis From the Centers for Medicare & Medicaid Services Medicare Part D Prescribers Database.

PURPOSE: Opioid prescribing trends in medical oncology are poorly defined past 2017, the year after the CDC updated opioid prescription guidelines in noncancer settings. We aim to characterize pain management by medical oncologists by analyzing opioid and gabapentin prescribing trends from 2013 to 2019, identify physician-related factors associated with prescribing patterns, and assess whether CDC guidelines for nononcologic settings changed prescribing patterns. METHODS: The Centers for Medicare & Medicaid Services (CMS) Medicare Part D Prescribers-by Provider, CMS Medicare Part D Prescribers-by Provider and Drug, and CMS Medicare Physician National Downloadable files from 2013 to 2019 were merged by National Provider Identification. The database included physicians' sex, years of practice, regions, and practice settings. Multivariable binary logistic regression identified significant predictors of total opioid, long-acting opioid, and gabapentin prescriptions. RESULTS: Binary logistic regression modeling revealed no significant difference in mean daily total opioid prescriptions from 2013 to 2017. Daily opioid prescriptions by medical oncologists decreased significantly after 2017 (P < .001). Increased opioid prescribing was associated with physician male sex (P < .001), practicing over 10 years (P < .001), and practice in nonurban areas (P < .001). Opioid prescribing was greatest in the South and Midwest United States (P < .001). The same patterns were observed with total long-acting opioid prescriptions, whereas gabapentin prescribing increased from 2013 to 2019 (P < .001). CONCLUSION: Opioid prescriptions by medical oncologists decreased significantly from 2013 to 2019, but this decrease was most substantial from 2017 to 2019. These results may imply that the 2016 CDC guidelines influenced medical oncologists, particularly more junior physicians in urban settings, to manage chronic cancer pain with alternative therapies.

Disparities in lung cancer short- and long-term outcomes after surgery: Analysis from the national cancer database.

INTRODUCTION: Social determinants of health are particularly important in lung cancer epidemiology. Previous studies have primarily associated social determinants with long-term outcomes, such as survival, but fail to include short-term outcomes after surgery. The National Cancer Database (NCDB) was used to draw associations between social factors of patients with lung cancer and short-term post-surgical outcomes, while comparing them to prognostic factors, including stage at diagnosis and survival. METHODS: The 2004-17 NCDB was queried for patients with primary epithelial tumor, squamous cell carcinoma, or adenocarcinoma of the lung treated with curative intent. Linear, binary logistic, Kaplan-Meier, and Cox proportional hazards regression models were utilized. RESULTS: On logistic regression modeling, male gender, low income, lacking insurance, and facility in the central United States were associated with poor short-term outcomes (<0.05). Increased age, White race, and Black race were associated with increased length of hospital stay and mortality, but negatively correlated with readmission rates (<0.05). Medicare and Medicaid were associated with increased length of stay and mortality, respectively (<0.05). Similar patterns were observed for higher stage at diagnosis (<0.05). Hazard ratios were elevated with increased age, male gender, White race, lacking insurance, Medicaid, and facility in the central United States (<0.05). CONCLUSION: Many social factors previously associated with poor prognosis after lung cancer diagnosis are also associated with poor short-term outcomes after surgery. This study implies that healthcare providers treating lung cancer should proceed with care while aware that patients with the discussed social factors are predisposed to complicated recoveries.