The researchers, reporting in Proceedings of the Royal Society B, analysed data from around 24,000 parents and their children who underwent paternity testing in Germany and Austria. They specifically looked at microsatellites - repetitive stretches of DNA that only mutate during cell replication - to identify new mutations in offspring that had arisen in either the egg of the mother or the sperm of the father.
The study found that in fathers aged 15 to 19.9 years old, the rate of mutations was already 6.7 times greater than the rate in mothers of the same age. And, while an age-related increase in mutation rate was observed among men, the magnitude of the change from teenagers to forty-something men was subsequently less than might have been assumed.
In a claim that most news sources ran with, the authors also say that their findings show that teenage fathers have an elevated rate of mutations over men in their twenties, by around 30 percent. They suggest that this could explain why some studies have found a higher rate of birth defects and genetic disease among the children of teenage fathers compared with older fathers.
In an email to BioNews, lead author Dr Peter Forster from the University of Cambridge said: 'Based on the implicit assumption that teenage fathers have a negligible mutation rate, researchers concluded that there must be a separate explanation for an increased birth defects risk in teenage fathers' children.
'Our DNA study shows this premise is wrong: the teenage fathers have a sixfold higher mutation rate than was previously thought. It is therefore not necessary to invoke a non-genetic explanation for increased birth defects in children of teenage fathers.'
However, in an article on the Guardian website, Professor Allan Pacey from the University of Sheffield said that the paper lacked statistical support for the claim that the rate of mutations was elevated among teenage fathers.
Speaking to BioNews, Professor Pacey described this conclusion as 'utter tosh'.
He explained that the data point for the rate of mutations in 15-19.9 year-old fathers falls within the 95 percent confidence interval of the authors' statistical analysis: 'So it isn't different and I think that they've been really bad by claiming that it's different.'
'I think that journalists have just swallowed it hook, line and sinker,' Professor Pacey said.