Back to 1984

This article by Allan Norman first appeared in Professional Social Work, a journal of the British Association of Social Workers, in October 2016 edition
This version is fully hyperlinked, but the final edited version appears in slightly different form.

For social workers, there has long been tension between the ethics of confidentiality and the belief – driven by countless enquiries into social work tragedies – that good practice involves more and better information sharing.

Social work places an emphasis on human rights and as such we need to understand how information-sharing initiatives, including those within our own profession, can potentially breach such rights.

In 2014 Supreme Court judge Lord Reed warned of this danger when he said: “The United Kingdom has never had a secret police or internal intelligence agency comparable to those that have existed in some other European countries, the East German Stasi being a well-known example. There has however been growing concern in recent times about surveillance and the collection and use of personal data by the state.”

Lord Reed went on to say: “...in many other European countries... for reasons of history, there has been a more vigilant attitude towards state surveillance. That concern and vigilance are reflected in the jurisprudence of the European Court of Human Rights in relation to the collection, storage and use by the state of personal data."[1]

Lord Reed was warning us that Britain's lack of history of the kind of authoritarian state control that crushed individual rights in other European countries could make us less vigilant to its creep here.

The judge made his comments in the wake of two cases in which people were denied employment after information about previous convictions was disclosed to their potential employer by state agencies.

In each case legal challenges were mounted – and won – on the grounds that the right to a private life enshrined under Article 8 of the European Convention on Human Rights had been violated.

Lord Reed’s warning – which followed an appeal by the Home Office to overturn the court’s decision – effectively did two things. Firstly, it explained how principles of human rights limit the jurisdiction of the state to share information and monitor the population. Secondly, it indicated why this is necessary, which is to serve as a bulwark protecting the citizen against the kind of over-authoritarian states that emerged out of the tumultuous events of the mid 20th century. Sadly Lord Reed went on to set out a list of examples where the UK has fallen short of human rights standards in this area.[2],

Fast forward to this summer and the Supreme Court declared another legislative information-sharing scheme to be in breach of human rights[3] – the so-called ‘Named Persons’ scheme in Scotland[4]. This judgement is, I believe, of truly momentous significance for social work practice across the UK.

The Scottish scheme not only appoints a Named Person for every child in Scotland to be a single point of contact for parents and facilitate support services, but also to receive and act upon concerns about a child's wellbeing. It is this sharing of information that could fall foul of data protection and human rights law, something I warned about back in 2013 in a legal opinion still available on the Scottish Parliament website[5].

To understand why, there are a couple of key concepts we need to explore to understand why. The first is known technically as a "statutory gateway". Information sharing is lawful where it is necessary "for the exercise of any functions conferred on any person by or under any enactment". So, the idea goes, if we pass new legislation which creates new functions that allow or require information-sharing, then that new legislation is itself a "statutory gateway" to lawful information sharing. Only, it isn't. The Scottish scheme aimed to be such a "statutory gateway". Only, it wasn't. The idea of being able to point to a specific piece of legislation as the authority for what we do seems to be reassuringly simple. Unfortunately, such legislation doesn't absolve professionals from responsibility for assessing the need to share information, nor from the human rights principles that constrain the right to do so.

Is anyone going to give balanced advice about when we should not be sharing, as well as when we should? The Supreme Court judgment was heavily influenced by an assessment that the guidance given to professionals was simply too woolly to meet this criterion, and that woolly guidance undermined the lawfulness of the whole scheme[6].

The other key idea is wellbeing. The Scottish scheme attempted to set the threshold for information-sharing around concerns about a child's wellbeing. Two points need to be made about this. It sets the threshold too low. And it is yet another woolly concept. What is very plain is that as a legal threshold, concerns about wellbeing would arise much sooner than concerns about the risk of significant harm. In Scotland, the campaign against Named Persons made hay with an example given in guidance about children not having a say in the decoration of their room or their TV viewing choices[7]. While the author of such ridiculous guidance clearly made themselves a hostage to fortune, they also neatly highlighted that wellbeing is a Humpty Dumpty concept[8] – it means no more nor less than what you want it to, and the real question is who is to be master.

But if you think that can be readily dismissed as an isolated absurdity, it is worth taking a closer look at the new paragraph 5.2 of the Standards of Conduct, Performance and Ethics issued by the HCPC to social workers in England earlier this year[9]. It permits professionals to breach confidentiality and share information in a service users "best interests" – like wellbeing, a threshold lower than significant harm, and just as vague. My advice is clear. Do not rely on your regulator's ethical standards as justification for information sharing. Paragraph 5.2 of our regulatory code is unlawful for the same reason as the Scottish scheme. The regulator should know better. Unfortunately, this amply demonstrate how difficult it is to get authoritative guidance on when and why you should not share information or concerns.

Since the mere existence of a statutory gateway, and of wellbeing concerns, does not make information-sharing lawful, the wider implications for social work practice may now be apparent.

Can you share information within a multi-agency safeguarding hub without consent? Not until you have conducted a proper evaluation of the risk, need and alternatives. Moreover, the Supreme Court made clear that where the subject of the information does not even know that it is happening, that creates its own unlawfulness[10] – so the fact that information sharing is within a "sealed hub" is no defence.

Can you seek information in the context of a child-in-need assessment? There is a framework for mandatory co-operation and information-sharing in child protection matters. But how often do you exchange information outside of that framework because you have all proceeded on the assumption of a right and duty to share?

And what about mandatory reporting? Of course, much depends on the exact terms of any legislation, since incompatibility is likely to arise where thresholds are set too low, or where there are insufficient procedural safeguards, such as where sharing takes place without independent scrutiny. But by now it should be clear that just because information-sharing schemes are in legislation, that does not make them human rights compatible. Both the scheme to share information about criminal records considered by the Supreme Court in 2014, and that to share wellbeing concerns considered by the Supreme Court this year, were contained in legislation, but incompatible with human rights.

Back in 2007, writing about confidentiality and information sharing under the Every Child Matters agenda[11], Eileen Munro wrote that "the introduction of a screening programme for a social problem needs to be measured against the same scientific criteria as screening for medical problems, such as screening for cervical cancer. There are three key criteria against which to judge a screening programme: predictive accuracy, treatability, and the level of damaging effects."

These wise words about evidence-based practice also provide a clear framework for reflecting upon the ethical basis for information-sharing. Predictive accuracy refers to the need to address both false negatives (we failed to detect a risk of harm when there was one) and false positives (we predicted a risk of harm when there was none). Simply put, the statutory schemes that extend information-sharing provisions tend to be addressed towards eliminating false negatives, while the human rights protections serve as a reminder of the ethical risks of false positives.

Munro's second criterion of treatability sounds a bit medical, but it links to familiar social work discourses. If we do not have the resources to provide the supportive input that will benefit families that are struggling, what is the point in recognising and recording that their children's "best interests" might not be being served? Auditors audit what they can measure, and recorded information routinely forms a central plank of this; but treatability is concerned rather with whether we have made people's lives better, and our information recording is rarely the central plank of this.

And finally, the damaging effects. Munro is inviting us critically to reflect that it is not only a failure to spot the risk of harm that has damaging effects. Unwarranted intrusion, intervention without consent, breaches of confidentiality also have damaging effects.

Many parents do indeed "get" the risk of harm of over-intervention that the Supreme Court cautions we in the UK have not experienced, as our European neighbours have. Who will seek information in confidence, if professionals routinely breach their confidence? Who will be comfortable at the idea that their minor misdemeanours as a child (as in the 2014 case), their childhood experiences of harm, mental health difficulties, or educational needs are even now being weighed up in some multi-agency forum to which they are not party? If professionals cannot be trusted with information, service users will choose not to share it, and that will cause harm. A vicious circle will develop, where ever less trusting relationships lead to ever fewer consensual interventions. Ultimately, this may not look so very different from the kind of totalitarian State that our European neighbours already better understand.

Which is why the Supreme Court in its judgment a few weeks ago warned of the risks: "the first thing that a totalitarian regime tries to do is to get at the children, to distance them from the subversive, varied influences of their families, and indoctrinate them in their rulers’ view of the world. Within limits, families must be left to bring up their children in their own way."[12]