The impact of substance use on health care utilization, treatment, and outcomes in patients with non-small cell lung cancer.

Background: Mortality from non-small cell lung cancer (NSCLC) has improved with screening and novel treatments. The substance use epidemic has threatened health outcomes in a variety of diseases, but little is known about how it is associated with NSCLC outcomes.

Methods: We performed a retrospective cohort study of 211 patients with NSCLC treated at a safety-net hospital. Sociodemographic data and clinical outcomes were extracted via review of electronic medical records. Patients were stratified based on substance use status. Comparative and multivariable analyses were performed to evaluate baseline characteristics and lung cancer outcomes including survival.

Results: Among 193 patients (91.5%) with information available on substance use, 24.9% reported substance use; specifically, alcohol, marijuana, and illicit substances. Patients with substance use were more likely to have increased health care utilization and poor social determinants of health, including safe housing, stable employment, and social support. There were no significant differences in treatment adherence. Only 6.3% of patients with substance use did not receive guideline concordant care (GCC) compared to 24.8% of patients without substance use; due to poor performance status, increased comorbidities, or loss to follow up. On univariable analysis, patients with substance use experienced inferior median overall survival (OS) if they had metastatic disease (0.40 1.03 years, P=0.01). However, in the multivariable analysis, substance use did not predict for survival. Independent predictors of mortality were sex (male HR, 1.67; 95% CI: 1.04-2.68; P=0.04), smoking status (current smoking HR, 2.63; 95% CI: 1.14-6.08; P=0.02), and stage (stage IV HR, 14.96; 95% CI: 6.28-35.63; P=0.008).

Conclusions: Substance use is associated with poor social determinants of health and increased health care utilization. On multivariable analysis, substance use was not independently associated with OS once guideline-concordant care was used. Future studies should focus on improving our understanding of these associations, delineating potential mechanisms, and developing evidence-based strategies to reduce health care utilization and overcome challenges related to poor social determinants of health.