Sex offending: genes more important than family environment

9 April 2015

Brothers and fathers of men convicted of sexual offences are up to five times more likely to be convicted of sexual offences than men in the general population, a new study shows. Genetic factors were found to make a substantial contribution to this increased risk with the shared family environment having a relatively small influence.

The study, by researchers from Oxford University (UK) and the Karolinska Institute (Sweden), used statistical methods to analyse data on all 21,566 men convicted of sexual offences in Sweden between 1973 and 2009.

The researchers looked at the share of sexual offences perpetrated by fathers and brothers of convicted male sex offenders and compared this to the proportion among Swedish men from the general population with similar age and family profiles. They found that around 2.5% of brothers or fathers of convicted sex offenders are themselves convicted of sexual offences. This compares to convicted sex offenders making up about 0.5% of men in the general population.

Having a brother convicted of a sexual offence led to five times the risk of a man being convicted of a sexual offence compared to a man whose brother had not been convicted of this type of crime. Meanwhile, being a father of a man convicted of a sexual offence led to almost four times the risk of being convicted of a sexual offence.

Statistical modelling suggests that around 40% of the difference in risk seen between brothers and fathers of convicted sexual offenders and brothers and fathers of those without such a conviction is due to genetic factors. Around 58% of the difference in risk appears linked to ‘non-shared’ environmental factors (such as perinatal complications, head injuries, childhood sexual victimisation, and peer group attitudes) that affect an individual but are not shared with other family members. Evidence on the difference in risk between maternal and paternal half-brothers (presumably reared in separate family environments) supports the idea that genetics makes a substantial contribution to increased risk.

A report of the research is published under open access in the International Journal of Epidemiology.

‘Importantly, this does not imply that sons or brothers of sex offenders inevitably become offenders too. But although sex crime convictions are relatively few overall, our study shows that the family risk increase is substantial. Preventive treatment for families at risk could possibly reduce the number of future victims,’ said Niklas Långström, Professor of Psychiatric Epidemiology at Karolinska Institute and the study’s lead author.

‘We are definitely not saying that we have ‘found a gene for sexual offending’ or anything of the kind. What we have found is high quality evidence from a large population study that genetic factors have a substantial influence on an increased risk of being convicted of sexual offences,’ said Professor Seena Fazel of Oxford University’s Department of Psychiatry, an author of the paper. ‘It tells us something about why if we take two sets of brothers, whose backgrounds might look identical, one set has a higher risk of sexual offending than the other – a large proportion of this difference is likely to be due to genetic factors.

‘At the moment genetic factors are typically ignored when it comes to making risk assessments of those at high risk of sexual offending. We know that current tools are not very effective at predicting who might commit sexual offences and it could be that taking family risk into account would lead to more accurate predictions. Many of the families we are talking about may already be known to social services for other reasons, and if we can predict those at high risk of offending with greater accuracy then it may be possible to shape these interventions and target education and preventative therapies where they could do the most good,’ said Professor Fazel.

Any conviction for a sexual offence under the Swedish Penal Code over the 37 year period was counted as a sexual conviction. This included convictions for the rape of an adult (6,131 offenders), child molestation (4,465 offenders) as well as crimes such as possession of child pornography, indecent exposure, and sexual harassment. Rates of sexual offending in Sweden are largely similar to those reported in the UK, Europe, the USA and Canada. Whilst direct comparisons should be handled cautiously, the type of offences that lead to such convictions in Sweden, as well as its recording methods and reporting levels, are thought to make it broadly comparable to other European countries.

Professor Fazel said: ‘Our research was epidemiological but a logical next step would be to devise a large-scale study that included biological samples, and could include a wider set of outcomes and intermediaries. In this way it would be possible to look for patterns in genes that might contribute to a higher risk of problem sexual behaviours alongside environmental factors.’

For more information contact the Oxford University News Office on +44 (0)1865 283877 or email

Notes to editors:

A report of the research, entitled ‘Sexual offending runs in families: A 37-year nationwide study’, by Niklas Långström, Kelly M. Babchishin, Seena Fazel, Paul Lichtenstein & Thomas Frisell, is published in the International Journal of Epidemiology.