A cross-sectional assessment of US cancer diagnoses during the COVID-19 pandemic.

[BACKGROUND] Disruptions to cancer diagnoses were widely reported in the US during the early COVID-19 pandemic. Whether any cases remained unaccounted-for by the end of the pandemic has not been fully assessed.

[METHODS] We collected data on invasive cancer diagnoses occurring among individuals aged 20-89 years between January 2020 and December 2022 from the Surveillance, Epidemiology, and End Results database. Expected cancer case counts and incidence rates with 95 % credibility intervals (95 %CrIs) were estimated for 2020-2022 from pre-pandemic trends (2005-2019) using Bayesian Age-Period-Cohort models. We compared observed rates with expected rates, and estimated unaccounted-for cases. Additional site-, stage-, and subgroup-specific analyses were performed.

[RESULTS] Among 2260,704 cancer cases diagnosed in 2020-2022, the observed incidence rate was 595.5 per 100,000 persons (95 %CI, 594.7-596.2), which was 6.7 % lower than the expected rate of 638.1 (95 %CrI, 620.1-656.1) and corresponded to 160,475 fewer-than-expected cases (95 %CrI, 99,777-221,174). Annual observed rates were significantly lower than expected in 2020 (565.8 vs. 630.7), with recovery in 2021 and 2022, though not enough to overcome the existing case deficit. Incidence rates for persons aged ≥ 65 years, nonmetropolitan residents, and non-Hispanic White individuals, as well as site-specific rates for lung and kidney cancers and non-Hodgkin lymphoma, remained below expected levels beyond 2020. Early-stage colorectal cancer diagnoses were 14.2 % lower than expected over the period.

[CONCLUSION] While annual cancer incidence rates returned to expected levels by the end of the COVID-19 pandemic, substantial numbers of unaccounted-for cases remained, raising concerns for future increases in cancer morbidity and mortality.