Risk-Based Strategies Superior to USPSTF Criteria for Lung Cancer Screening

THURSDAY, July 2, 2026 (HealthDay News) — Risk-based strategies are superior to the U.S. Preventive Services Task Force (USPSTF) criteria for optimizing efficiency and minimizing variation of lung cancer screening across racial and ethnic groups, according to a study published online June 30 in the Annals of Internal Medicine.

Xiaoshuang Feng, Ph.D., from the International Agency for Research on Cancer in Lyon, France, and colleagues examined the performance of risk prediction models to define screening eligibility across four U.S. racial and ethnic groups in a cohort study involving 641,830 participants aged 50 to 80 years with a smoking history from 12 U.S. cohorts. For 16 lung cancer prediction models, calibration and discrimination were quantified.

The researchers found that across the 16 models, general patterns included substantial underestimation of lung cancer risk in non-Hispanic Black participants (11 models: expected to observed ratio <0.75); lower discrimination in Asian participants versus all other groups (13 models); and lower discrimination in non-Hispanic Black versus non-Hispanic White participants (15 models). All risk-based strategies achieved better average estimated screening efficiency and reduced racial and ethnic differences in efficiency compared with the 2021 criteria from the USPSTF (USPSTF-2021) when a same-sized screening-eligible population as USPSTF-2021 was enforced. The best performance was seen for the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial Model 2012 and Life Years Gained From Screening-Computed Tomography model. None of the strategies could optimize eligibility, sensitivity, and efficiency at the same time as reducing racial and ethnic differences.

“Our results lend support for the use of lung cancer risk models for screening eligibility, and for the inclusion of race and ethnicity as a predictor in risk models,” the authors write.

Abstract/Full Text (subscription or payment may be required)