What the premium checkup package reliably buys is radiation. Price and dose correlate at 0.812

Korean comprehensive checkup packages are sold in tiers. The tiers have names — intensive, premium, overnight — and prices that rise with them. What has not generally been published is what the extra money buys, measured in the one unit that can be compared across hospitals.

What was measured#

A study in Environmental Health and Toxicology took the comprehensive examination programmes advertised on the websites of ten university hospitals in Seoul — 190 programmes in total — and had radiologists review the radiation-based studies each one included. Only the effective doses of the basic studies were counted; optional add-ons were excluded, so these figures are floors rather than ceilings.

The annual effective dose limit for a member of the public from artificial sources is 1 mSv. These are single examinations, delivered to people attending because they are well.

The correlation that names what the tiers are#

A correlation of 0.812 between price and dose is strong enough to say what the tier system substantially is. The single most critical factor determining which band a programme fell into was whether it included positron emission tomography combined with computed tomography.

The variation between hospitals is its own finding. Programmes with the same stated purpose differed in dose by as much as 2.1 times depending on where they were bought. A patient choosing between two hospitals' cancer packages is, without being told, choosing between two very different exposures.

What a dose figure translates into#

Millisieverts are an abstraction to almost everybody who receives them. A 2025 risk model published in JAMA Internal Medicine converts the abstraction into cases. Using organ-specific dose distributions from a multicentre CT dose registry, scaled to the approximately 93 million CT examinations performed on 61.5 million patients in the United States in 2023, it projected about 103,000 radiation-induced cancers over those patients' lifetimes, with a 90 percent uncertainty limit of 96,400 to 109,500.

• Lung cancer was the commonest projected outcome at 22,400 cases, followed by colon at 8,700, leukaemia at 7,900 and bladder at 7,100.
• Abdomen and pelvis CT in adults accounted for 37,500 of the 103,000 projected cancers, from 30 million of the 93 million examinations.
• Risk per examination was higher in children and adolescents, but higher use in adults meant adults accounted for 91 percent of the projected cancers.
• Across sensitivity analyses the projection ranged from 80,000 to 127,000.

The authors concluded that if current practice persists, CT-associated cancer could eventually account for 5 percent of all new cancer diagnoses annually.

Reading these two studies together, carefully#

These are separate bodies of evidence and they must not be merged into a Korean casualty figure. The JAMA Internal Medicine projection is of American CT use, and it is a model built on the linear no-threshold assumption — that risk scales with dose all the way down — which is the standard basis for radiation protection and is not a directly measured result at these dose levels. Nobody counted 103,000 cancers. They were projected.

What the two studies jointly establish is the shape of the problem rather than its size in Korea. CT delivers a quantifiable dose; the dose carries a modelled risk that regulators treat as real; Korean private checkup packages deliver that dose to asymptomatic people in quantities that rise with price; and no study cited here has measured the outcomes of the Korean population receiving them.

That last absence is the finding worth stating plainly. Korea has the national claims infrastructure to link checkup imaging exposure to later cancer incidence, and no published analysis has done it. The dose is documented. The consequence is not.