Lung Cancer Screening Misses Many People Who Go on to Develop the Disease

A national analysis of adults diagnosed with lung cancer has exposed a striking weakness in the United States’ strategy for finding the disease early: many patients who ultimately develop lung cancer do not qualify for screening under current rules, and most do not receive a low-dose scan before their diagnosis. The study, based on the 2024 National Health Interview Survey, found that only about half of adults with lung cancer met the eligibility criteria recommended by the US Preventive Services Task Force (USPSTF) at the time they were diagnosed. Even more unexpectedly, just one-third reported undergoing low-dose computed tomography, or LDCT, before their cancer was found. The findings suggest that the problem is not simply that people fail to use an available test. The rules themselves may overlook substantial numbers of people at risk, while practical and clinical barriers prevent many eligible individuals from being screened.

Lung cancer is the third most commonly diagnosed cancer in the United States and remains the country’s leading cause of cancer-related death. Its danger is closely tied to timing: tumors discovered after they have spread are much harder to treat successfully, whereas earlier-stage disease may be removed surgically or treated with curative intent. LDCT is designed to detect small abnormalities before symptoms appear. Unlike a conventional diagnostic CT scan, which may cover a broader clinical question and use a higher radiation dose, LDCT uses a carefully calibrated, lower amount of radiation to create images of the lungs. Evidence from the National Lung Screening Trial showed that screening high-risk people with LDCT reduced lung-cancer mortality by about 20 percent compared with chest radiography. Yet the test is useful only if people at elevated risk are identified and return for regular examinations, and if suspicious findings are rapidly evaluated without causing unnecessary invasive procedures.

Under the 2021 USPSTF recommendations, adults between 50 and 80 years old should receive annual LDCT if they currently smoke or quit within the previous 15 years and have accumulated at least 20 “pack-years” of smoking. A pack-year represents smoking one pack of cigarettes per day for one year; someone who smoked two packs daily for 10 years, for example, would have 20 pack-years. The 2021 update lowered the minimum age from 55 to 50 and reduced the smoking threshold from 30 to 20 pack-years, expanding the population eligible for screening. The criteria were intended to improve access and address disparities, but they remain largely dependent on age and cumulative smoking exposure. That creates a blind spot for people who develop lung cancer after lighter smoking histories, long periods since quitting, or no history of smoking at all.

Researchers from the University of California, Irvine, retrospectively examined survey responses from 115 adults aged 50 or older who reported a lung-cancer diagnosis within the previous 10 years. After applying the survey’s statistical weights, the group represented an estimated 638,702 US adults. The analysis used smoking histories to determine whether each person would have qualified for LDCT under the 2021 USPSTF rules at the time of diagnosis. Smoking information was incomplete for four participants, or 3.5 percent of the sample, meaning their eligibility could not be determined. Among the remaining 111 people, 58—52.3 percent in the unweighted sample and approximately 52.0 percent after weighting—met the screening criteria. When people who had never smoked were excluded, eligibility rose to 64.4 percent, or about 66.3 percent using weighted estimates. Even that higher figure means that roughly one in three patients with a smoking history would not have qualified under the current rules.

The study also revealed a sharp difference between having received a CT scan at some point and having received a true screening scan before cancer was diagnosed. Overall, 92 of 110 respondents, or 83.6 percent, said they had undergone LDCT at some time. But only 36 people, or 32.7 percent, reported receiving the scan before their lung-cancer diagnosis; the weighted estimate was 29.6 percent. This distinction matters because CT imaging can be ordered for many reasons, including investigating coughing, chest pain, infection, trauma, or another medical condition. It can also be performed to monitor a known cancer. Such scans may reveal a tumor, but they are not equivalent to systematic screening of people without symptoms. The gap between the two figures therefore suggests that a large share of the imaging reported by patients was diagnostic or follow-up care rather than preventive screening.

The analysis found that women had substantially lower odds of meeting eligibility criteria than men. In the adjusted statistical model, the odds ratio for female sex was 0.31, with a 95 percent confidence interval from 0.12 to 0.77. In practical terms, the result reflects the fact that women in the study generally had accumulated fewer pack-years and had stopped smoking longer ago than men. Women also made up a much larger share of patients who had never smoked: 76.2 percent of never-smokers with lung cancer were female, compared with 43.3 percent among participants with a history of smoking. The researchers observed a possible disparity for non-White participants as well, with an odds ratio of 0.35, although the result narrowly missed conventional statistical significance. When never-smokers were removed from the analysis, demographic factors were no longer significantly associated with eligibility, suggesting that smoking patterns—and the higher proportion of women among never-smokers—helped drive the apparent sex difference.

The findings highlight a biological and epidemiological challenge that fixed smoking thresholds cannot fully address. Cigarette smoking remains the dominant preventable cause of lung cancer, but it is not the only pathway to the disease. Some tumors arise in people who never smoked, including groups in which lung cancer among never-smokers appears to be increasing or is already disproportionately common. Asian women who have never smoked, for instance, have been identified in previous research as an emerging population with elevated risk. Other contributors can include secondhand smoke, occupational exposures, radon, air pollution, inherited susceptibility, and chronic lung inflammation. A rule based only on age and pack-years may therefore be efficient for identifying a major high-risk group while still missing individuals whose risk is clinically meaningful but distributed across several factors. The study’s authors argue that the results add to evidence that current criteria do not capture everyone who will eventually develop lung cancer.

Why eligible people fail to complete screening is less clear from the survey. No demographic characteristic examined by the researchers was significantly associated with receiving LDCT before diagnosis. The reasons may lie in a complicated chain of obstacles: clinicians may not identify eligible patients, patients may not understand that screening is recommended before symptoms appear, and health systems may lack the staff or infrastructure to arrange annual scans and follow-up. Cost, insurance coverage, transportation, time away from work, fear of radiation, stigma surrounding smoking, and anxiety about a possible cancer diagnosis can all reduce participation. Primary-care providers may also be uncertain about the detailed eligibility rules or may prioritize other urgent health needs. Screening is not a one-time event but a recurring process, so a program must maintain contact with patients and ensure that abnormal results are tracked. A single missed appointment can break that chain.

The researchers caution that their estimates should not be interpreted as a perfect measure of screening performance. The survey relied on participants’ memories of their smoking histories, the timing of scans, and the type of CT imaging they received. People may not distinguish an LDCT screening examination from a diagnostic CT performed after symptoms developed, which could make screening uptake appear higher than it truly was. Conversely, incomplete recall of smoking may cause some people to be classified incorrectly as ineligible. The study was cross-sectional and used information collected after diagnosis, so it cannot establish why screening did or did not occur. Because only one year of survey data was analyzed, some racial, ethnic, and language groups were represented by very small numbers, preventing reliable comparisons. Demographic and socioeconomic characteristics such as insurance, income, employment, and residence may also have changed between diagnosis and survey completion, limiting the ability to connect them to earlier screening decisions.

Risk-based screening models could offer one possible way forward. Instead of relying on age and pack-years alone, these models estimate an individual’s probability of developing lung cancer using multiple variables, potentially including smoking intensity, time since quitting, chronic obstructive pulmonary disease, family history, body characteristics, socioeconomic conditions, and other clinical factors. Such approaches might identify some people who fall outside the USPSTF thresholds, including those with lower smoking exposure or additional non-smoking risks. But they also introduce new complications: the necessary data may be missing from medical records, risk calculations can be difficult to explain, and adding more inputs may make programs harder to implement consistently. For now, the new analysis points to a two-part public-health emergency. Screening systems must do a better job reaching people who already qualify, while researchers and policymakers must determine whether eligibility rules should evolve beyond smoking and age. Until both problems are addressed, many lung cancers may continue to be discovered only after the window for the simplest and most effective treatment has begun to close.