Sex crimes are at the top of the justice agenda, which may mean revisiting preventative measures previously dismissed. Adam James reports.
The scientific establishment has, until now, been sceptical about lie-detector techniques. The British Psychological Society has not changed its view since 1986, when it called them unscientific. Last year, the National Academies of Science in the US condemned polygraphy for its "weak scientific underpinnings".
But in a book out this week, Don Grubin, a professor in Newcastle University's Sexual Behaviour Unit, puts the case for considering widescale use of polygraphy techniques to prevent sex offenders from reoffending.
In Sex Offenders in the Community: Managing and Reducing the Risks , Grubin details how lie detector techniques - such as measuring breathing rates, electrical conductivity of the skin, heart rate and blood pressure - were applied to 30 sex offenders on probation. They were asked if they had been in contact with children or on the lookout for victims. Grubin says he believes the techniques prevented at least three offenders from reoffending, and he argues that the time has come to take modern developments in polygraphy seriously when working with sex offenders.
Sex crimes (particularly those committed against children) and how to prevent released offenders from reoffending have risen to the top of the criminal justice agenda. Last month the government announced plans for the electronic tagging of sex offenders. Among the measures outlined in its sex offences bill were harsher prison sentences and a tightening of the powers of police and other agencies to track offenders after their release.
Policy-makers are, it seems, exploring all avenues in a bid to crack down on recidivism.
So far, the government's Sex Offender Treatment Programme has been heralded as the key therapeutic intervention in the prevention of repeat offending.
For more than a decade, SOTP psychologists have been applying cognitive behavioural therapy to males imprisoned for rape or for sex offences against children.
But some observers were concerned after the publication by the Home Office earlier this year of The Prison-based Sex Offender Treatment Programme - An Evaluation , which says that while the reconviction rate over two years for 650 treated SOTP sex offenders was 2.6 per cent, it was 2.8 per cent for untreated sex offenders; a negligible difference. The results became slightly more encouraging only when violent reconvictions were added to the equation. Then the reconviction rate for treated sex offenders was a statistically significant 4.6 per cent, compared with 8.1 per cent for untreated offenders.
What is certain, however, is that the SOTP faces the same difficulties that all interventions experience when they are evaluated, and electronic tagging and polygraphy will be no different.
At the heart of the problem is the fact that most convicted sex offenders are not reconvicted of a similar offence. This is known as "the low base rate problem". A review published in 2001 by the US government's Centre for Sex Offender Management reported rape reconviction rates within four to five years of between 11 and 28 per cent for rapists. Child sex abusers' reconviction for similar offences over four to five years varied enormously, from 6 to 43 per cent.
In the UK, some of the most recent work by Oxford University researchers commissioned by the Home Office looked at 192 male offenders four and six years after they had been released from long prison sentences. Less than 10 per cent were reconvicted for a subsequent serious sex offence; in comparison, burglar reconviction is about 30 to 40 per cent.
Researchers, such as Clive Hollin, professor of the division of forensic mental health at the University of Leicester, put reconviction at 10 per cent over five years for all sex offenders. He is among those who concede that with so many conflating factors, such as the heterogenous nature of offenders, changes of sex offence definitions and reliance on reconviction for measuring reoffending, results are far from conclusive. "We know incredibly little about how to prevent relapse, and there is very little evidence about what is effective in terms of prevention."
Anthony Beech, senior lecturer in forensic psychology at the University of Birmingham and one of the authors of The Prison-based Sex Offender Treatment Programme - An Evaluation , says SOTP programmes are effective, particularly in changing the attitude and motivation of sex offenders.
"(Committing sex offences) is all about motivation. Like with drink drivers, unless their motivation is changed, they will sit on their hands and make no change," he says.
Nevertheless, some are highly dubious about the Home Office pumping so much of its resources into cognitive behavioural therapy. These include David Wilson, professor of criminal justice at the University of Central England and co-author of Innocence Betrayed: Paedophilia, the Media and Society , published last year. He says cognitive behavioural therapy-based interventions are founded more on political expediency than on solid evidence. The therapy "absolutely speaks a language spoken by politicians.
It says: 'If you put people on this course we will reduce offending by a certain percentage.' That is just what politicians who say they want to be tough on crime and tough on the causes of crime want to hear," he says.
Wilson adds that the credibility of research on sex offending treatment by the Home Office is compromised by the fact that some of the key proponents of cognitive behavioural therapy for sex offenders are also responsible for evaluating its efficacy. Hollin says this is a "valid criticism" but "anyone evaluating their own research would get that criticism".
Wilson, a former governor of Grendon Prison, in Aylesbury, Buckinghamshire, believes that more emphasis should be placed on schemes aimed at befriending sex offenders once they are released from prison. These are based on a Canadian model in which the community works in collaboration with released sex offenders, who are given the chance to become a "core member" of a "circle of support" of volunteers. The Home Office funds three such projects around the country.
The offender makes a covenant with one volunteer on how they work together, such as regular phone conversations or shopping trips together. The core premise is that the community should use its resources to include an offender, rather than exclude him or drive him underground (as many fear would be the consequence if a scheme such as the News of the World 's Sarah's Law were introduced, allowing a community to know where local offenders live).
There is little hard research on the effectiveness of befriending, but reports from Canada suggest that for high-risk sex offenders who have been in one of its 30 circles of support over a two-year period, the reoffending rate is 10 per cent, compared with a usual level of 21.5 per cent.
But, with so many methods of intervention claiming effectiveness, is it possible to know which works best?
Cognitive behavioural therapy practitioners emphasise that the evidence at present favours it as the leading rehabilitation intervention. "It is important to be driven by evidence and not by preference for an approach," Hollin says. "While I don't think it is the only approach that should be used, the reason why it is so popular is because of the steady accumulation of evidence on reducing reoffending."
Polygraphy may well be next in line for similar empirical scrutiny.
Sex Offenders in the Community: Managing and Reducing the Risks , edited by Amanda Matravers (Cambridge Criminal Justice Series) is published by Willan Publishing on November 7, £30.00.