Lung cancer screening reduces mortality but reaches only a fraction of patients. Low-dose computed tomography (LDCT) screening is recommended for high-risk populations and has been shown to reduce lung cancer mortality by 20% in landmark trials. The USPSTF 2021 guidelines expanded eligibility to adults aged 50 to 80 with at least 20 pack-years of smoking history and who currently smoke or quit within the past 15 years. However, these criteria were developed based on Western populations and may not translate to Asian patient demographics.
Lung cancer in China has distinct epidemiological characteristics. China bears the highest global burden of lung cancer, with approximately 820,000 new cases and 715,000 deaths annually. Critically, the demographic profile differs markedly from Western populations: a large proportion of Chinese lung cancer patients are never-smokers, and female patients represent a substantial share of the overall burden. These differences raise fundamental questions about whether USPSTF-based eligibility criteria are appropriate for Chinese asymptomatic patients.
Screening-ineligible patients remain understudied. Most research on lung cancer screening focuses on the eligible population by definition, leaving the clinical characteristics, disease outcomes, and survival trajectories of ineligible patients poorly characterized. Understanding what proportion of real-world lung cancer patients fall outside current screening criteria, and how those patients differ in terms of tumor biology and survival, is essential for evaluating and improving guideline design.
This study fills a major evidence gap. Prior studies have evaluated USPSTF eligibility in retrospective cohorts but rarely at the scale or multicenter diversity needed to draw reliable conclusions for a country as large and heterogeneous as China. This study uses a nationally representative cohort of over 100,000 patients across 26 hospitals to characterize the full scope of the eligibility gap and its clinical implications.
106,266 asymptomatic lung cancer patients from 26 Chinese hospitals. The study retrospectively enrolled patients diagnosed with primary lung cancer at 26 tertiary hospitals across China between January 2014 and December 2021. All patients were asymptomatic at the time of diagnosis, meaning they had no lung cancer-related symptoms and were identified through incidental imaging, physical examination, or health screening programs rather than symptom-driven clinical visits.
USPSTF 2021 criteria applied to define eligibility. Screening eligibility was defined per the 2021 USPSTF recommendation: age 50 to 80 years, at least 20 pack-years of cumulative smoking, and current smoking or cessation within the past 15 years. Patients meeting all three criteria were classified as eligible; those failing any criterion were classified as ineligible. This classification was applied uniformly across all 26 institutions.
3,711 patients excluded for missing data. Of the 106,266 enrolled patients, 3,711 were excluded due to incomplete smoking history data (pack-year information required for eligibility classification). The final analytical cohort included 102,555 patients with complete eligibility classification. Eligibility rates were calculated overall and stratified by year from 2014 to 2021 to assess temporal trends.
Survival analysis with propensity score matching. Overall survival was analyzed using Kaplan-Meier methods and compared between eligible and ineligible patients using log-rank tests. To control for confounding by age, sex, stage, histology, and treatment, propensity score matching (PSM) was performed at a 1:1 ratio. Cox proportional hazards regression provided adjusted hazard ratios for mortality. Analyses were stratified by TNM stage (I, II, III, IV) to characterize survival differences across disease stages.
Eligibility rate of 8.8% overall. Of the 102,555 patients with complete data, only 8,985 (8.8%) met all three USPSTF 2021 criteria for lung cancer screening. The vast majority, 93,570 patients (91.2%), were classified as screening-ineligible despite having confirmed lung cancer diagnoses. This figure represents the core finding of the study: current screening guidelines would have missed more than 9 in 10 Chinese lung cancer patients who were ultimately diagnosed asymptomatically.
Eligibility declined sharply from 2014 to 2021. Temporal analysis showed a pronounced downward trend in annual eligibility rates. In 2014, 21.6% of patients met USPSTF criteria. By 2021, this figure had fallen to 6.1%, representing an annual percentage change of -17.4%. This decline reflects the increasing share of never-smokers and younger patients being diagnosed over time, driven by expanded use of CT in health screening programs across China capturing a broader demographic of asymptomatic individuals.
Non-smoking was the dominant cause of ineligibility. Among ineligible patients, the leading reason for not meeting criteria was never having smoked (insufficient pack-years), accounting for the majority of exclusions. Females represented 66.1% of the ineligible patient group, reflecting the high proportion of never-smoking women with lung cancer in China. In contrast, the eligible group was predominantly male and had heavier lifetime tobacco exposure consistent with USPSTF criteria design.
Stage distribution favored earlier disease in ineligible patients. Ineligible patients were diagnosed at earlier stages on average compared to eligible patients. Stage Ia diagnoses were more common in the ineligible group (73.7% vs. 61.0% in the eligible group). This difference reflects the demographic composition of the ineligible group: younger never-smoking females are more likely to harbor early-stage adenocarcinomas that are frequently detected incidentally on CT during routine health screenings.
Screening-ineligible patients had significantly better survival. Kaplan-Meier analysis showed that ineligible patients had substantially longer overall survival than eligible patients across all follow-up periods. After propensity score matching to control for demographic and clinical differences, screening-ineligible patients had a 40% lower risk of mortality compared to eligible patients (adjusted hazard ratio = 0.60, 95% confidence interval 0.55 to 0.66, p less than 0.001). This survival advantage persisted after adjustment for age, sex, stage, histology, and treatment modality.
Stage-specific survival advantages confirmed. Stratified analysis by TNM stage revealed that the survival advantage in ineligible patients was present at multiple stages. At stage I, ineligible patients had an adjusted HR of 0.63, indicating 37% lower mortality risk. At stage III, the adjusted HR was 0.76, a 24% lower mortality risk. Stage II and stage IV results showed trends in the same direction, though with more modest effect sizes, likely reflecting the dominant influence of treatment options at advanced disease.
Adenocarcinoma more prevalent in ineligible patients. Histological analysis showed that adenocarcinoma was significantly more common in screening-ineligible patients (71.4%) compared to eligible patients (63.5%). Adenocarcinoma has better overall prognosis than squamous cell carcinoma or small cell lung cancer, and its higher prevalence among ineligible patients contributes to the observed survival advantage. However, the survival benefit persisted even after controlling for histology in multivariate models.
Earlier stage at diagnosis partly explains but does not fully account for better outcomes. While ineligible patients more frequently presented with stage Ia disease, multivariate Cox regression analysis showed that the survival advantage remained significant after adjusting for stage distribution. This means that factors beyond early detection, such as tumor biology, molecular characteristics of adenocarcinoma in never-smokers, and potentially different treatment responsiveness, also contribute to better outcomes in the ineligible group.
USPSTF criteria systematically exclude the majority of Chinese lung cancer patients. The finding that 91.2% of asymptomatic Chinese lung cancer patients fall outside USPSTF 2021 criteria represents a fundamental mismatch between guideline design and real-world Chinese epidemiology. This gap cannot be attributed to screening program failure; rather, it reflects that the biological and demographic drivers of lung cancer in China differ substantially from those in the US populations used to derive the USPSTF recommendations.
Never-smoking females require a different risk framework. The dominant ineligible subgroup is never-smoking females, a population with lung cancer risk driven by genetic susceptibility, indoor air pollution, cooking fume exposure, and other non-tobacco factors. These risk factors are not captured in any current USPSTF eligibility dimension. Effective screening of this population would require different eligibility criteria, potentially incorporating genetic biomarkers, imaging-based risk scores, or exposure assessments tailored to non-smoking risk factors.
Paradoxically, ineligible patients have better survival outcomes. The 40% lower mortality among ineligible patients challenges the framing that USPSTF eligibility identifies the highest-risk patients who benefit most from screening. Instead, the data suggest that the ineligible majority is experiencing better clinical outcomes, likely because they are younger, have different tumor biology, and present at earlier stages. This complicates the clinical narrative that ineligible patients simply do not need screening.
Health screening programs in China are capturing the ineligible majority. The declining trend in USPSTF eligibility from 21.6% to 6.1% over the study period coincides with the rapid expansion of health examination programs in China offering CT imaging to broader populations. These programs are detecting a growing proportion of never-smoking, younger, female patients with early-stage adenocarcinoma, a group that current USPSTF criteria define as not requiring screening but who in practice are being diagnosed and surviving well when found.
USPSTF criteria were not designed for Chinese epidemiology. The USPSTF 2021 recommendations were derived from the National Lung Screening Trial and NELSON trial, which enrolled predominantly white, male, heavy-smoking populations. The racial, demographic, and behavioral differences between these populations and Chinese lung cancer patients mean that direct application of USPSTF criteria to China will inevitably produce poor coverage. This study provides large-scale empirical evidence confirming the scale of that coverage gap.
Comparison to prior smaller studies. Previous studies in Asian populations have reported USPSTF eligibility rates ranging from 15% to 40% depending on cohort composition. The 8.8% eligibility rate in this study is lower than most prior estimates, which the authors attribute to the exclusively asymptomatic nature of the cohort, the broader demographic captured by Chinese health screening programs in recent years, and the use of the expanded USPSTF 2021 criteria rather than the older 2013 criteria which had stricter age and smoking thresholds.
Risk prediction models may offer a better alternative. Several risk prediction tools, including the PLCOm2012 model and the Liverpool Lung Project (LLP) risk score, incorporate non-smoking risk factors such as family history, occupational exposures, and prior respiratory disease. These approaches may achieve better sensitivity in Chinese populations by moving beyond tobacco-centric eligibility to a multifactorial risk assessment framework. The authors advocate for development and validation of China-specific risk models incorporating environmental and genetic factors.
Study limitations include retrospective design and referral bias. All participating hospitals are tertiary centers, meaning the cohort may overrepresent patients with access to high-quality healthcare and may not fully reflect rural or primary care-level disease patterns. Retrospective classification of smoking history relies on medical records, which may underestimate pack-years in some patients. Additionally, the study period coincides with major shifts in Chinese CT screening policy, making secular trends difficult to separate from true epidemiological changes.
A major eligibility gap demands guideline reform. This study demonstrates that USPSTF 2021 screening criteria, as currently formulated, would exclude 91.2% of asymptomatic Chinese lung cancer patients from recommended screening. This is not a marginal shortfall but a fundamental coverage failure that, if left unaddressed, will result in systematic underdiagnosis of a large and growing patient population identifiable through health screening programs already in widespread use in China.
Ineligible patients' better survival should inform, not dismiss, the problem. The finding that ineligible patients have 40% lower mortality than eligible patients does not imply they do not benefit from detection. It reflects that the ineligible population has different and often more favorable tumor biology, earlier stage at detection when found, and different demographic risk factors. The fact that these patients are surviving well when detected incidentally actually strengthens the argument for systematic screening of this group rather than leaving their detection to chance.
China needs population-specific screening criteria. The evidence supports developing Chinese-specific lung cancer screening eligibility criteria that move beyond the USPSTF smoking-based framework. Such criteria should incorporate the distinct risk factors of the Chinese population, including indoor air pollution, cooking fume exposure, genetic susceptibility markers, and family history of lung cancer. Prospective validation of alternative eligibility models in Chinese cohorts is an urgent research priority.
Health examination programs provide a practical platform. The rapid growth of routine CT-based health examination programs in China offers an existing infrastructure through which expanded or revised screening criteria could be implemented. Rather than treating guideline-ineligible patients as outside the scope of screening, Chinese public health policy should consider using these programs as a vehicle for systematic early detection in the broader asymptomatic population, particularly among never-smoking females who represent the fastest-growing demographic of lung cancer patients in China.