The study used anonymised medical records for 11 million young Australians, including 6,80,000 who were exposed to Computerised Tomography (CT) scans between 1985 and 2005.
The Australian researchers, with colleagues at Oxford University and the International Agency for Research on Cancer in France, found that for every 1,400 CT scans before the age of 20 there was one extra case of cancer over the following 10 years.
This small increase in cancer risk must be weighed against the undoubted benefits from CT scans in diagnosing and monitoring many different health conditions, researchers said.
"Nevertheless, it is clear from our study that it is important for doctors to use CT scans only where they are necessary. By reducing the number of scans performed in a large population, there will be a small but corresponding reduction in the number of cancers in later years," Mathews said.
More From This Section
However, the radiation doses from CT scans are very small, and there has been uncertainty about whether such small doses would really cause cancer, and whether any small increase in risk could be measured reliably.
The new study was able to answer this question by linking anonymised Medicare records of CT exposures for the entire population of young Australians, aged 0-19 years between 1985 and 2005, to cancers diagnosed up to the end of 2007.
There were 866,580 CT scans given to 680,211 people in this cohort of 10.9 million young people. Of 60,674 cancers diagnosed, there were 3,150 among those who had received one or more CT scans, compared with 2542 cancers that would have been expected if there was no effect of CT scans on subsequent cancer risk.
Younger children had a greater proportional increase in risk of cancer following a CT scan, but more CT scans were carried out in the teenage years. The study was carried out just in children and young people, but there may be some risk from CT scans in adults as well, researchers said.
The study was published in the BMJ medical journal.