Lung cancer · JAMA Netw Open · 2025
Summary: women’s higher share among added lung-cancer patients isn’t chance. Its results: could be chance.
How often chance alone gives this gap in women’s share if none existed
The paper says
Today’s US guideline says who should get lung-cancer scans. A looser rule, 20 years of smoking, adds people. The summary: women are 55% of added people with lung cancer versus 48% under the guideline; chance alone would make that gap under 1 time in 1,000.
Its own data show
The paper’s results: 70 women of 128 added people with cancer (55%) versus 370 of 747 under the guideline (50%). Chance alone makes that gap more than 1 time in 4; redoing the calculation gives the same. The summary’s 55% vs 48% compares everyone added, cancer or not (1,103), with everyone the guideline covers (3,840).
Shares wobble by chance, more in small groups; a gap usually counts as convincing only if chance alone makes it under 1 time in 20. Both comparisons show 55% women; the cancer one has a smaller gap (vs 50%, not 48%) among 128 people, not 1,103. So women are clearly overrepresented among everyone added, not provably among those with cancer.