Korea built the world's first national lung cancer screening programme. Five years in, uptake doubled and the income gap did not close

Lung cancer screening with low-dose computed tomography has strong randomised evidence behind it in high-risk smokers, and almost nowhere has been willing to run it at national scale. Korea started in 2019, and the first five years of data are now published.

What was built, and who it is for#

The Korean National Lung Cancer Screening Program invites current smokers aged 54 to 74 with a smoking history of at least 30 pack-years. Screening is by low-dose CT, and the programme attaches physician counselling on the result and on stopping smoking. A study in Cancer Research and Treatment analysed National Health Insurance Service and National Cancer Screening Information System records for 2019 to 2023.

Doubling uptake in five years, in a group defined by an ongoing smoking habit, is a substantial operational achievement. The authors attribute it to systematically inviting eligible people rather than waiting for them to present.

Why fewer positives is a sign of maturity#

Positive screening rates fell from 9.1 percent in 2019 to 7.0 percent in 2023. That reads like a programme finding less, and it is not.

A first screen photographs a lifetime of accumulated nodules, most of them harmless and none of them previously seen. A second screen has the first one to compare against, so a stable nodule is recognised rather than flagged. The fall came mainly from fewer Lung-RADS category 3 findings — the indeterminate middle — among people being screened for a second or third time.

Variation between institutions in category 4 findings also fell significantly over the period, which is the other signal of a programme settling: the same scan increasingly gets the same reading regardless of where it is done.

The gap that did not close#

The same pattern appears across Korean screening programmes rather than being peculiar to this one. Liver cancer surveillance participation among people with disabilities rose from 25.7 percent in 2006 to 49.6 percent in 2015 while participation among people without disabilities rose from 24.9 to 54.5 percent — the two lines crossed and then diverged.

What five years cannot yet show#

This study reports uptake, counselling rates and the distribution of scan findings. It does not report whether Korean lung cancer mortality has changed, and it could not: a screening programme's effect on deaths takes considerably longer than five years to become visible, and the randomised evidence that justifies the programme was generated elsewhere.

Nor does it quantify the harms — the biopsies and operations that follow findings that turn out to be benign, or the radiation dose of repeated CT in a population being scanned annually. Those are the counterweights to any screening programme, and this analysis is not where they are counted.

What it establishes is that a national low-dose CT programme can be run at scale, that participation can be doubled by inviting people systematically, and that doing so does not by itself reach the people who smoke most.