A 17-year study has concluded that screening mammography — in which all women in certain age groups are routinely screened for breast cancer — does not reduce the incidence of advanced tumors, but does increase the diagnosis of lesions that would never have led to health problems.

In Denmark, screening was implemented in different regions at different times, so researchers there were able to compare groups of women who were screened with those who were not.

If screening were effective, a reduction in the incidence of advanced tumors would be expected — tumors would be caught and treated when small. Instead, the researchers found no difference in incidence between screened and unscreened groups. But in screened groups, they found substantial overdiagnosis — that is, detection of tumors that would not become cancers needing treatment.

The study, in Annals of Internal Medicine, found that, depending on age and other factors, between a fifth and a third of tumors detected by screening were overdiagnosed. The researchers concluded that screening does not prevent advanced cancers or lower breast cancer mortality.

“Some types of screening are a good idea — colorectal, for example,” said the lead author, Dr. Karsten Juhl Jorgensen, deputy director of the Nordic Cochrane Center. “But breast cancer has a biology that doesn’t lend itself to screening. Healthy women get a breast cancer diagnosis, and this has serious psychological consequences and well-known physical harms from unnecessary treatment. We’re really doing more harm than good.”