‘Risk of Harms’, summary critique

The MoJ’s ‘Risk of Harms’ Report is invalid and should not be relied upon

The Ministry of Justice published the tripartite ‘Assessing risk of harm to children and parents in private law children cases’ in June 2020.

However, despite carrying the MoJ’s insignia, this was not a product of the MoJ at all. Although this independent project was funded by the MoJ, it was established so that it could not be beholden to any standards of practice or quality controls.

Inadequacy, malpractice and deviation

The assembled “expert” panel of 12 were almost all lawyers, with no technical expertise, only one subject-matter expert (who was not a report author), and were completely devoid of any scientific or statistical expertise. This violated the President of the Family Division’s guidance on ‘experts’ in family matters [30]

The report they produced has no valid empirical support; failed to make any ‘assessment of risk’ (probability, proximity or impact); and ignored all of the objective data available to the MoJ from court records – instead generating its own unreliable subjective data. The authors classified any person accused of abuse as a ‘perpetrator’, regardless of the findings of the court, and inappropriately commissioned a Literature Review to synthesize qualitative data. Such an investigative structure would inevitably generate ‘misinformation’.

Importantly, the report ignored the legal requirement that the paramount consideration of Family Court is the children’s welfare interests. The panel were all but entirely focussed on parents who allege domestic abuse (predominantly women) and barely acknowledged, let alone assessed, the welfare outcomes of the children.

From this uncontrolled mix of inadequacy, malpractice and deviation, the Report identified and problematised a ‘pro-contact culture’ in Family Court.

However, maintaining safe contact is a positive legal obligation of the Family Court, as established by common law both in the UK and at the European Court of Human Rights, and enumerated in Articles 7-9 of the United Nations Convention on the Rights of the Child.

Free from the requirements of the Civil Service Code, the RoH Panel engaged ‘pseudo science’ and semantic chicanery to undermine children’s rights, and create bias in the Family Court. The report should never have been accepted.

Invalid methods

The appropriate method for this type of analysis is to draw a randomised sample of court files sufficient to derive statistical significance, and to conduct a structured query of all parties independently. However, the Panel instead undertook two processes to form its evidence base:

  1. A Literature Review; and
  2. An uncontrolled internet survey, followed by a series of ‘roundtables’.
Literature Review:

The author of the Literature Review, another lawyer and colleague of the panel’s authors, was appointed in a non-competitive tender without any selection criteria[31]. She had published disparaging views about the efficacy of the scientific method and recent partisan views on the subject matter which should properly have precluded her from consideration.[32]

The Panel wrongly commissioned a ‘literature review’ from her to synthesise a predominantly qualitative body of evidence. A literature review is a non-systematic form of ‘Rapid Evidence Assessment’, and the civil service’s ‘Magenta Book’[33] recommends against this instrument:

“Systematic reviews often take months and therefore, Rapid Evidence Assessments can be used if less rigour can be tolerated. These methods often focus on quantitative data, but other frameworks can be used to systematically integrate qualitative information.”

The guidance[34] specifically states that Literature Reviews are “Not Systematic”, “their findings should therefore be treated with a great deal of caution” and

“Literature reviews are prone to selection and publication bias; that is, they only tend to review evidence that is readily available, and they can be over reliant on sources that disproportionately report studies with positive outcomes.”

Predictably, the Literature Review contained no reference to any study that ran contrary to the author’s conclusion; it misstated and ignored the body of evidence available in respect to parental alienation; it contained errors of fact (such as misleadingly stating the prediction of an ONS study on coercive control,  but the study actually debunked the prediction); and it twice referenced and appended the ‘Duluth model’, which has no scientific support and the Office of the Statistics Regulator has found is ‘likely to mislead’ due to its anti-male sexism.

Survey and roundtables:

The panel’s internet-based qualitative survey process was bereft of any statistical controls, and is therefore unable to be used for conclusions. The subsequent ‘round tables’ were selected from the same sample and so inherited the same errors.

  • The data was entirely subjective and anecdotal;
  • Allegations were assumed to be true and took no account of false, vexatious or exaggerated claims;
  • There was no verification whether the responders actually attended court, or check of their accounts against court or other records, or even any means to prevent multiple responses from the same person;
  • The responders were most commonly an aggrieved party who had already lost their case once their claims were tested in court, and so were unusually unreliable even for subjective data; 
  • There was no assessment of (or curiosity about) the children’s wellbeing that resulted from the court’s decisions;
  • There was no control group (ie any result could be independent of court intervention);
  • The sample was unrepresentative at 80% female responders (private law is 50:50 male : female parties), and so biased and inherently unreliable.
  • It is clear the responses from the lawyers were different to the responses from the mothers’, but no question appears to be raised as to why the lawyers saw things so differently.

Semantic Chicanery:

The footnotes of the Report reveal that it uses the term ‘perpetrator’ to describe anyone accused of domestic abuse, even if they were entirely exonerated by the court. Barrister Sarah Phillimore commented “I have never known a case where a person found to have perpetrated serious abuse was simply left to get on with it and unsupervised contact ordered.” Yet the panel fail to acknowledge that the allegations of abuse were subjected to the court’s determinations before orders for contact were made.

Therefore a key finding of the report, that:

“some form of direct contact between children and perpetrators of domestic abuse was ordered in the great majority of all private law cases”

which suggested the Family Court is failing to safeguard victims and children, actually meant that:

“the great majority of domestic abuse allegations in private law cases were determined to be either totally unfounded or too trivial to prevent direct contact”

The sleight of hand by the Panel to include all allegations as if they were true, coupled with it’s studied incuriosity about the welfare outcomes for the children, means its report was both misleading and potentially harmful to children’s welfare.

No professional standards apply:

As an ‘independent’ experts report, neither the Civil Service Code nor the Magenta Book apply. Alternatively, if this Report were an advertisement, it would violate the rules of the Advertising Standards Agency due to the unrepresentative sampling, the absence of statistical controls, the use of subjective data where objective data is available (“even if the marketer’s intention was to make a subjective claim”), the presentation of subjective claims as objective claims, and misleading publication under the MoJ’s insignia without adequate warning.[35] 

Unfortunately, the Report was commissioned so that it would fall into a regulatory ‘black hole’ between government publication and marketing material, and so the panel’s conduct and product cannot be held to any professional standards.