Removing precancerous polyps found during a colonoscopy reduces by half a person’s odds of dying from colon cancer, according to a study that suggests the tests can be used to help separate patients by risk.

The research, reported Wednesday in the New England Journal of Medicine, followed 2,602 older patients who had precancerous polyps removed during a colonoscopy.

Over a median of 15.8 years, the people studied had less than half the number of colon cancer deaths than would have been expected in the general population that age, according to the results.

The finding suggests that an initial screening could be used to separate patients with polyps into high-risk and low-risk groups, said Michael Bretthauer, a gastroenterologist who co-wrote an editorial accompanying the study.

High-risk patients, those with precancerous polyps, benefited from getting repeated colonoscopies afterward, said Bretthauer, at Oslo University Hospital Rikshospitalet in Norway. When no precancerous polyps are found in a patient, “you can just forget about them, send them home.”

While the latest research doesn’t prove that colonoscopies saves lives in the general population, it supports colon cancer screening guidelines, said Ann Zauber, a biostatistician at Memorial Sloan-Kettering Cancer Center in New York and the report’s lead author.

The American Cancer Society recommends that men and women get a colonoscopy or another colon cancer screening test starting at age 50, according to its website. Colonoscopy should be done every 10 years, it recommends.

The 2,602 patients who got precancerous polyps removed were age 62 on average when they started the study between 1980 and 1990, in an era where colonoscopy was not generally used as a screening exam in the healthy population. The patients were then followed for as long as 23 years.

Through the end of 2003, only 12 of the patients had died from colon cancer, far lower than the 25.4 colon cancer deaths that would have been expected in otherwise similar patients in the general population, according to the study results.

In a second group of 773 patients who were found to have harmless colon polyps during their colonoscopy, only one patient died from colon cancer during the follow-up period.