Masked divergence in cancer mortality following COVID-19 disruption

While overall cancer mortality in the United States has declined steadily for decades, systemic healthcare disruptions during the COVID-19 pandemic-including delays in screening, diagnosis, and treatment-may have created heterogeneous impacts across different malignancy types that are not visible in aggregate data.

To determine whether site-specific cancer mortality during the COVID-19 pandemic (2020-2023) deviated from expected trajectories based on established pre-pandemic trends.

This population-based analysis utilized U.S. National Vital Statistics System data from 2003 through 2023. Age-standardized mortality rates (AASRs) were calculated for specific cancer sites and modeled using log-linear regression over the baseline period (2003-2019). These models were used to project counterfactual expected mortality for the pandemic period (2020-2023). Excess mortality was defined as the percent difference between observed and projected AASRs, with significance determined via bootstrap simulation and false discovery rate correction.

Overall cancer mortality remained closely aligned with pre-pandemic declining projections. However, site-specific analysis revealed significant divergence. Positive excess mortality was observed for Hodgkin lymphoma (mean +9.5%; 95% CI, 2.7-17.2%) and prostate cancer (mean +7.9%; 95% CI, 2.8-13.2%). Additionally, mortality attributed to ill-defined malignant sites (+15.2%) and unspecified sites (+8.5%) increased significantly. Conversely, liver and intrahepatic bile duct cancers demonstrated mortality below expected levels (mean -15.5%; 95% CI, -19.5% to -11.3%). Most positive deviations peaked in 2022, suggesting a delayed effect of disrupted diagnostic and clinical pathways.

Aggregate cancer mortality trends during the pandemic masked significant, disease-specific divergences. The observed excess mortality in certain highly curable or screen-detected malignancies may reflect differential vulnerability to disruptions in screening, diagnosis, treatment access, or cause-of-death attribution. These findings highlight the need for site-specific surveillance to identify long-term consequences of systemic strain and to inform targeted recovery efforts in oncologic care.