Risk or Safety dice image

Photo courtesy of ©iStockphoto/Bratovanov


“How risky is he, out of 10?” “Is she high, medium or low risk?” My answer is always the same, “it depends…”
I’ve written previously on some essentials of risk assessment in children with sexual behaviour problems, but this came up again this week in conversation. It got me thinking…
Now, we all want to minimise risk, reduce the number of victims, etc. This is a given.
But I want to share some thoughts on why risk needs to be kept in proper perspective…

Keeping risk in perspective…


Professionals with a specialism in this area of work are regularly asked to make assessments of and recommendations about managing the degree of riskiness a child poses.
This is fair enough. After all, as well as the welfare of the child themselves, the safety of others is very important.
It’s not risk, per se, that is my issue; it’s the obsession with “measurement.”

The two faces of risk assessment…

There are two main facets of risk assessment, in my view:

  1. How likely is this young person to behave in a harmful sexual way in the future? – i.e. risk prediction
  2. What might the specific potential risk actually be? – i.e. what might they do, to whom and in what circumstances?

Our previous post on this subject looks at how we might do a better job of making accurate and helpful statements about potential risks.
Once we know this, we can plan what to do about the risk.
We can give useful guidance to fellow professionals: carers, parents, schools, etc. As well as to the young person themselves.

The risk prediction bandwagon & why we should jump off!…

Risk prediction-in terms of actuarial instruments-have no great value in working with children and young people with harmful sexual behaviour, in my view.
To my knowledge, there are no robust and validated accurate risk prediction measures for children and young people. Not all would see it this way. But I think the research-such as it is-adds up to this basic conclusion.
In contrast to North America, the U.K. and its European partners are generally less focussed on the statistical prediction of risk. This is a good thing.
Here’s why:

  • There are significant limits to our ability to predict risk accurately in this population
  • Research on the issue is varied and uncertain in its conclusions – some seem to suggest that we can be reasonably accurate, others claim we can’t
  • Population conclusions-i.e. what we can say about groups of young people from statistical research-don’t necessarily apply to individual young people
  • Having a firm and accurate prediction of likely risk does not take into account the positive impact of therapy or treatment to address the problem. And we know young people are more amenable to treatment than adults, in general.
  • The presence of strengths or protective factors can mitigate risk; most measures to date don’t account for this

For all of these reasons (and others) our focus should be on understanding the causes of harmful sexual behaviour and what to do about them.

Re-balancing risk prediction…

There are a number of factors that need to be borne in mind when considering assessment of harmful sexual behaviour in children and young people:

  • Young people are a work in progress – physically, neurologically, social and behaviourally
  • Adolescence is a time of great sexual flux as young people emerge from childhood and begin the transition into adulthood
  • Previous abusive experiences often come to the surface – (sometimes for the first time) during this period and need to be dealt with
  • A young person’s strengths – resilience and the presence of protective factors (family, friends, hobbies, etc.) must also be weighed up
  • Risk prediction as a statistical likeliness measure across a population tells us nothing about this young person, per se. – As such it is a minor factor in assessment terms and even more so in terms of treatment planning

Baby & bath water…

Don’t get me wrong: we need to understand and comment on the risks a young person poses. I just don’t think this can be done “out of ten” or with basic comment on “high, medium or low?”
Our assessments must address risk. But they must do so in a balanced and informative way by answering the three basic questions:

  1. Risk to whom?
  2. Risk of what?
  3. In what circumstances?

Then we can identify and put in place the right risk management measures.
By doing this, we help the child understand the impact of their behaviour, we equip those with caring roles to supervise wisely and we can usefully inform treatment planning.

Final thoughts…

Unlike the adult field, young people are a work in progress. We do well to target our principal efforts at addressing the developmentally causal factors (i.e. via therapy/treatment programmes), rather than on the recidivism prediction task.
Measuring the possible problem across populations has, in my view, a limited contribution to make in offering meaningful resolutions to individual young people in trouble.

What do you think?…

  • What are your thoughts on risk? Do we over state the importance of prediction? 

Please let me know what your thoughts are… Leave a comment below or click here.
You can subscribe to JonnyMatthew.com by typing your email address into the ‘Subscribe here’ boxes on the top right of this webpage. I send out blog posts like this twice each week. Your information is safe and you can unsubscribe anytime, easily.

Related posts…

Pass it on…

You can also “Like” this site on Facebook and “Follow” on Twitter

© Jonny Matthew 2014