A retrospective epidemiological study of screening-derived solid cancer incidence proxy rates in South Korea before and after COVID-19 using Korean National Health Insurance Service data
TL;DR
Aggregate South Korean national screening data show pandemic-era temporal associations with heterogeneous changes in selected solid cancer program proxy rates, which support an indirect screening-disruption and ascertainment interpretation rather than causal proof of a direct COVID-19 effect on solid cancer incidence.
Abstract
Background The COVID-19 pandemic disrupted cancer screening and diagnostic pathways globally. Aggregate screening-derived indicators require cautious interpretation because they do not represent confirmed cancer registry incidence but might reflect program participation, referral patterns, and denominator composition. Objective To evaluate pre- and post-COVID-19 changes in selected solid cancer screening-derived incidence proxy rates in South Korea, and to estimate model-based proxy rates through 2028. Methods This retrospective ecological longitudinal study analyzed publicly available aggregate annual cancer-screening statistics from the Korean National Health Insurance Service (NHIS) and Korean Statistical Information Service (KOSIS) for six solid cancer programs (stomach, liver, colorectal, breast, cervical, and lung cancer). Annual proxy rates per 100,000 screened or eligible persons were calculated. Pre-COVID (2010–2019) and post-COVID (2020–2024) means were compared using Welch t-tests with Mann–Whitney U sensitivity analyses. Ordinary least-squares (OLS) forecasts were generated for 2025–2028. Interrupted time-series analysis was not feasible owing to an insufficient post-COVID observation window. Results Post-COVID changes in proxy rates were heterogeneous across selected programs. Breast cancer proxy rates increased by 17.1% (Welch p = 0.002) and liver cancer proxy rates by 83.6% (Welch p = 0.002); the liver cancer increase reflects denominator restriction to a high-risk population rather than a true epidemiological surge. Stomach cancer decreased by 23.3% (Welch p = 0.001), cervical cancer by 56.9% (Welch p = 0.005), and colorectal cancer by 25.8% (Welch p = 0.007). Lung cancer data were available only from 2019 to 2024 (program inception) and were interpreted descriptively. Conclusions Aggregate South Korean national screening data show pandemic-era temporal associations with heterogeneous changes in selected solid cancer program proxy rates. These findings support an indirect screening-disruption and ascertainment interpretation rather than causal proof of a direct COVID-19 effect on solid cancer incidence.