Not all sex offenders are high risk

There is a need to change the public and professional conversation around sex offenders. The present debate centres around the perception that all sex offenders are high risk. This affects public policy as well as how people respond to someone with a sex offending history.

Certainly there are some high risk, multiple offenders who deserve the highest level of scrutiny that we can offer. In Canada, we need only think of Paul Bernado to bring to mind those we are less likely wandering the streets. In addition, we certainly don't want an approach that fails to recognize how devastating sexual assaults are on victims. Those who have been assaulted, regardless of how young or old, should have significant supports through the health care and criminal justice systems.

Paul Bernardo

Where things go off the rails, is when we view all sex offenders as belonging to one risk group. This would have us spending large amounts of money and resources for all offenders when really they should be spent in accordance with risk. This is a tough conversation to have as we rightly find sexual assault so offensive.

In Canada, we have a group of world recognized researchers who have helped to develop a series of actuarial tools that assist in determining the relative risk that an individual poses. These tools are well researched and have good predictive ability, albeit certainly not perfect. There are no perfect tools in any field for predicting calamity. We know that we face a risk by getting in the car. Based upon certain driving behaviours, your risk of an accident will go up. But even with high risk behaviours, such as speeding and driving while impaired, we cannot predict with certainty that you will have an accident. When you go into hospital for surgery, they can give you odds about how successful the surgery is likely to be, but they cannot guarantee it.

As a society, we seem to want greater certainty with sex offenders than we want with other forms of risk. Perhaps understandable given the life changing nature of the crime upon the victim. But we cannot jail everyone who has committed a crime we find abhorrent.

The Canadian research group has long been looking at the relative risk of offenders and thus, the level of service need that should be applied to an offender.  They have now published new research that concludes that high risk sex offenders may not be high risk forever. This is important work as it helps to sort out those who stay high risk from those whose risk diminishes over time.

Their work looked at a group of offenders over a 20 year follow up period. Thus, this is what is known as longitudinal research. It also looked at a large number of offenders - 7.740 who had been incarcerated. Their conclusions matter:

• The risk was highest during the first few years after the offender is released from jail. This suggests that we more intense supports and monitoring in the period following release really does matter;
• The longer an offender remained sex offence free, the more the risk lowered. The researchers found this to be particularly true for the high risk offenders. 

Thus, we should be careful to ensure that there is careful consideration of the re-offence patterns of offenders. The longer that there is a lack of re-offeding, the lower the risk. Of course, low risk does not mean risk free. Thus, there remains no sense in taking someone with an offence history and placing them in charge of children. 

So what can we do? Here are some thoughts:

• properly assess offenders for level of risk using actuarial tools. Then assign offenders to the appropriate level of supervision. For example, it makes no sense to put a low risk offender in groups with high risk offenders. This tends to increase the level of risk of the low risk offender;
• Monitor individuals over time to ensure that they are monitored relative to risk;
• Ensure that agencies that are responsible for working with children screen for those who are inappropriate  to work with children. An example of when this was not done (and there are many examples) us found here
• Listen carefully when children talk about behaviours of adults that could be concerning;
• Listen when other adults feel that a person in a caring position appears to be behaving in ways that don't feel right; and
• follow up.

Research reference:

Hanson, R.K., Harris, A.J., Helmus, L. & Thornton, D. (2014). High risk sex offenders may not be high risk forever. Journal of Interpersonal Violence. In Press.

Child protection's inconvenient truth: Is Daniel Pelka an example?

Media in the United Kingdom are swarming around the death of Daniel Pelka. He was four years old at his death at the hands of his mother, Magdelena Liczak and step father, Mariusz Krezolek, . They have now been convicted of his murder. He was starved and treated cruelly for at least 6 months prior to dying.

Daniel Pelka

The media reports are slowly leaking into Canada, although they are not creating the sensation that even a casual reading of UK papers suggests is going on there.

There is a serious case review in the matter, but its publication has been delayed until next month. Apparently the SCR panel learned information in the trial that it must now consider. That, in itself, is a fascinating statement suggesting that those working on the case had information gaps. At this point, it is not clear the child protection had a significant, or even any role in the family. The case is, none the less, highly reminiscent of many cases such as Jeffrey Baldwin, Victoria Climbie, Khyra Ishaq, Logan Marr and so on. Here in Canada, we have another tragic case subject to a public review that has just finished up. This is the case of Phoenix Sinclair. It will be several months before we hear what the commissioner, Justice Ted Hughes has to say.

In all of these cases there are valuable lessons to be had on child protection practice. These cases highlight the ways in which case management and interventions can go wrong. They also help us to see the sorts of common errors that occur. It is those errors, in particular, that need avoidance.

Cases such as these raise the delicate balance between protecting children and preserving families. The challenge with these cases is that they inevitably result in several things:

1. A scapegoat will be sought. In Daniel's case, the politicians are already doing this;
2. In some fashion, a politician will also state that the protection of children is paramount and that they intend to get to the bottom of this so it will not happen again;
3. New procedures will be introduced, manualized and bureaucratized; 
4. New training will be recommended; 
5. Social workers will become wary and fearful of making a mistake; and
6. Another child will die.

I am reminded of the article by anthropologist  China Scherz in the United States who looked at how social workers try to put into place good judgement while also trying to manage that balance between protection and family. She looked at the introduction of actuarial tools to help better identify risk. Her article is well worth reading. What one concludes is that there is a real risk that we create what looks to be science (e.g. actuarial tools, structured decision making processes, formal assessment processes) which mask the reality of child protection in science. It is a false mask.

While I am a supporter of these various types of tools, there is still the reality that a front line worker is going to have to make decisions from that information - leave the child in the home? close the file? remove the child? offer services? Scherz's article shows that this still is a very individual decision that a worker must make, and that there can be a wide variance in the ways that workers see a case.

There is no way to remove the worker and the personal decision making from the work. Workers bring training, experience and, yes, beliefs into the day to day interaction with families. No matter how much structure and science you try to wrap around that, this truth remains.

Which leads to the inconvenient truth in child protection - mistakes will be made, children will be subject to further abuse and maltreatment that might have been avoided if a different decision was made, and a small number of children involved in child protection will die - but most will not.

To expect perfection in child protection is unrealistic. Let these tragic cases continue to serve as educational opportunities so that the common mistakes can be highlighted and the highly unusual errors can also be brought to light.

Reference:

Scherz, C. (2011). Protecting children, preserving families: Moral conflict and actuarial science in a problem of contemporary governance. PoLAR: Political and Legal Anthropology Review, 34 (1), 33-50.