Connected Data: A New Approach to Information Sharing

WEBVTT
00:01:24.120 --> 00:01:33.719
Welcome to Guardians of Data, the show where we explore the world of information law and information governance, from privacy and AI to cybersecurity and freedom of information.
00:01:33.719 --> 00:01:35.239
I'm Ibrahim Hassan.
00:01:35.239 --> 00:01:42.599
In each episode, we speak with experts and practitioners to unpack the big issues shaping the IG profession.
00:01:42.599 --> 00:01:47.159
Today we're talking about one of the biggest questions facing public services.
00:01:47.159 --> 00:01:51.640
How do we make better use of the information we hold about children?
00:01:51.640 --> 00:02:06.599
The report, Connecting Data, Intelligent and Informed Delivery to Support Every Child to Succeed was published in July 2026 by a group consisting of academics, policymakers, public service leaders, and practitioners.
00:02:06.599 --> 00:02:13.319
It argues for a national devolved model of connected data to improve the lives of young people.
00:02:13.319 --> 00:02:22.039
The report says that the risks of sharing data have sometimes overshadowed the risks of not sharing data when children need coordinated support.
00:02:22.039 --> 00:02:31.639
It proposes a national devolved connected integrated care board using NHS infrastructure to connect information across services.
00:02:31.639 --> 00:02:40.199
The report contains 10 recommendations and draws on examples of connected data approaches already being used in the north of England.
00:02:40.199 --> 00:02:47.560
To help us discuss the report and the proposals, I'm joined by Professor Mark Monwilliams from the University of Leeds.
00:02:47.560 --> 00:02:57.480
Mark holds a chair in cognitive psychology at the University of Leeds and is Professor of Psychology at the Bradford Institute of Health Research.
00:02:57.480 --> 00:03:04.360
He's also a Turing Fellow at the Alan Turing Institute, the UK's National Data Analytics Center.
00:03:04.360 --> 00:03:08.599
Mark is one of the executive authors of the Connecting Data Report.
00:03:08.599 --> 00:03:21.000
He's also closely involved with the Connected Bradford Project, mentioned in the report, one of the UK's most interesting examples of bringing information together across organizational boundaries.
00:03:21.000 --> 00:03:29.319
With Mark's help, we answer some of the big questions, such as what does lawful and proportionate information sharing look like in practice?
00:03:29.319 --> 00:03:37.960
How do we make sure that connecting data actually improves lives rather than simply creating another layer of infrastructure?
00:03:37.960 --> 00:03:45.000
And how do we ensure that it's done in a way which is fair, lawful, transparent, and ethical?
00:03:45.000 --> 00:03:46.680
Let's jump in.
00:03:46.680 --> 00:03:49.560
Mark, welcome to Guardians of Data.
00:03:49.800 --> 00:03:50.120
Thank you.
00:03:50.120 --> 00:03:51.319
It's wonderful to be here.
00:03:51.560 --> 00:03:59.159
It's going to be a fascinating conversation, I am sure, because we get a lot of questions with regards to effective information sharing.
00:03:59.159 --> 00:04:08.199
And as you know, it's something that information governance people, as well as child protection experts and other public officials, have been grappling with for a while.
00:04:08.199 --> 00:04:13.479
So hopefully we will have lots of listeners really interested in this.
00:04:13.479 --> 00:04:20.759
Mark, you're one of the executive authors of the Connecting Data Report, which was published in July of this year.
00:04:20.759 --> 00:04:33.960
I'd like to start with the basic premise behind the report, which is that children are being left unsupported and exposed to avoidable harm because public services are failing to share information effectively.
00:04:33.960 --> 00:04:40.360
Having done so much work in this area over a number of years, what do you feel are the barriers to information sharing?
00:04:40.360 --> 00:04:52.199
Why are public officials and those who can make a difference to children's lives still thinking about the risks of sharing information as opposed to risks of not sharing information?
00:04:52.519 --> 00:04:56.120
So I think the barriers are cultural and political.
00:04:56.120 --> 00:04:57.960
They definitely aren't technical barriers.
00:04:57.960 --> 00:05:04.360
We know that because we've successfully connected information across our different public services.
00:05:04.360 --> 00:05:20.199
But there are complexities, and the complexities I think relate to the very siloed approach that we have to public service provision, where we have a number of different services, a number of different government departments, all responsible for a bit of a child or young person or family's lives.
00:05:20.199 --> 00:05:30.519
And they very much and correctly focus on the aspects of that child, young person, or family sort of lives that they have control over.
00:05:30.519 --> 00:05:36.759
But the difficulty is that children, young people, and families do not fall into nice, neat sort of silos.
00:05:36.759 --> 00:05:41.960
And this is why we really need to understand children, young people, and their families in a holistic manner.
00:05:41.960 --> 00:05:56.440
And the only way we can do that is by properly putting together all of the information that we hold on those children, young people, and families, and properly understanding the issues that they're facing in order that we can address those issues.
00:05:56.759 --> 00:05:57.159
Interesting.
00:05:57.159 --> 00:05:58.759
You mentioned the term silos.
00:05:58.759 --> 00:06:03.000
That's something that we hear often, particularly from information governance professionals.
00:06:03.000 --> 00:06:09.559
And we've been here before, though, haven't we, in terms of various reports into child deaths and neglect.
00:06:09.559 --> 00:06:15.639
I'm thinking about baby Peter, Victoria Climbier, and more recently, S Sara Sharif.
00:06:15.879 --> 00:06:26.919
So, as you've mentioned, pretty much every other serious case review concludes that serious harm would have been prevented if we had properly shared information.
00:06:26.919 --> 00:06:30.200
And we're seeing it again in the Southport inquiry.
00:06:30.200 --> 00:06:35.960
Every serious case review flags up the fact that our systems just aren't fit for a purpose.
00:06:35.960 --> 00:06:38.840
I don't think we need any more warning sort of signals.
00:06:38.840 --> 00:06:47.159
Really, the question is after decade after decade of identifying this as a fundamental problem, why are we not addressing the problem?
00:06:47.159 --> 00:06:50.279
And that's where I think we're at at this moment in time.
00:06:50.279 --> 00:06:57.960
We've all got to come together to say this is completely unacceptable in 2026 and work together to resolve the issue.
00:06:58.360 --> 00:06:59.320
Absolutely, I agree.
00:06:59.320 --> 00:07:03.399
What's great about your report, I found, was that it doesn't just highlight the issue.
00:07:03.399 --> 00:07:08.679
You have come up with a model of information sharing.
00:07:08.679 --> 00:07:12.120
But how is it different, this idea of connected data?
00:07:12.120 --> 00:07:28.679
How is it different to previous proposals of information sharing where individuals, where public officials, social workers may just pick up the phone, or the school will pick up the phone and pass on information to other agencies where they have concerns?
00:07:28.679 --> 00:07:32.600
This idea of connected data, can you just explain it a bit more, please?
00:07:32.840 --> 00:07:37.240
There are two fundamental problems to relying on people picking up a telephone.
00:07:37.240 --> 00:07:40.279
The first fundamental problem is it just doesn't work.
00:07:40.279 --> 00:07:46.519
We know that because decade after decade we've seen the consequences of that telephone not being picked up.
00:07:46.519 --> 00:07:52.759
Secondly, it's potentially problematic because you don't have control over what information is being shared.
00:07:52.759 --> 00:07:58.600
So we can't be certain that information is being shared appropriately, safely, and securely.
00:07:58.600 --> 00:08:10.039
And this is why we need a new model, a model that allows information to be stored safely and securely, and the appropriate information shared with the appropriate individuals.
00:08:10.039 --> 00:08:19.720
And the model that we've adopted really just sort of takes the idea that we already host very, very sensitive information on the public.
00:08:19.720 --> 00:08:23.080
We host that information within the NHS.
00:08:23.080 --> 00:08:34.840
So we're hosting that information, then clearly it is beholden on us to make sure that information is held securely, safely, and with all the appropriate governance processes around it.
00:08:34.840 --> 00:08:51.000
And all we then are doing, all our model then develops, is the idea that if we have all that really sensitive information, well, having educational records in amongst that information, having information about the built environment, simply allows us to provide better public services.
00:08:51.000 --> 00:08:56.200
So that's really at the heart of our model, the idea that we already hold very, very sensitive information.
00:08:56.200 --> 00:08:59.559
We already have secure and safe data environments.
00:08:59.559 --> 00:09:04.600
So why not expand that information so we can properly connect up our public services?
00:09:04.919 --> 00:09:10.679
That will certainly resonate with our listeners, the idea of properly joined up services.
00:09:10.679 --> 00:09:15.080
It's something that governments have been trying to tackle for many years.
00:09:15.080 --> 00:09:26.840
But here, are we talking about professionals being able to see information held by other organizations, data sets being linked for analysis, or something more sophisticated?
00:09:27.960 --> 00:09:31.080
So all of these are possibilities.
00:09:31.080 --> 00:09:36.600
Once you have the connected information, you can then start doing targeted commissioning.
00:09:36.600 --> 00:09:41.240
You can start understanding where we need to improve public sort of service delivery.
00:09:41.240 --> 00:09:54.360
We can suddenly start operating in a three-dimensional technicolor world rather than trying to guess at what might be happening on the basis of black and white two-dimensional snapshots of a particular area or a particular region.
00:09:54.360 --> 00:10:02.040
So we really need information in order to effectively run public services.
00:10:02.040 --> 00:10:07.160
But once you have that connected data, well, then you can do all kinds of sophisticated things in a very safe way.
00:10:07.160 --> 00:10:16.440
So we can take pseudonymized extracts of those data and have scientists really exploring the data to understand where we can improve public service delivery.
00:10:16.440 --> 00:10:26.600
But we can also, and this is the holy grail, start sharing relevant information across relevant professionals in a very safe and secure way.
00:10:26.600 --> 00:10:38.200
And that doesn't mean that a pediatrician suddenly sees all of a child's education records, but it does mean a pediatrician can know if a school have any potential safeguarding concerns.
00:10:38.200 --> 00:10:57.240
So this is the model we would really like to move towards, where we can safely and securely share relevant information across professionals who really need that information if they're to meet their statutory responsibilities, if they're to meet the legal obligations that they have to provide the best possible care to that child, young person, and the family.
00:10:57.480 --> 00:11:16.200
So it's not just about statistical analysis, looking at the sort of the bird's eye view, the overall picture of child services and services centered around children in a particular area, but it's also in a controlled environment, sharing and giving access to identify records where necessary.
00:11:16.200 --> 00:11:17.480
That's very interesting.
00:11:17.480 --> 00:11:29.320
What are we currently unable to see because of health, education, social care, and other services holding separate pieces of the big picture, which you mentioned in the report?
00:11:30.440 --> 00:11:38.920
We're unable to see the intersections and interactions between health and social care and education and the criminal justice system.
00:11:38.920 --> 00:11:46.360
And we know that in our really disadvantaged areas, it's those gaps through which far too many children, young people, and families are falling.
00:11:46.360 --> 00:11:56.279
How can we possibly provide integrated whole system care to a child or family when we completely fragment ourselves?
00:11:56.279 --> 00:12:05.399
We have to start really understanding what's playing out holistically in a child, young person's life in that family sort of context.
00:12:05.399 --> 00:12:16.519
And the only way we can do that is to actually understand what's happening in school, uh, what's happening in the accident and emergency referrals, what's happening in terms of interactions with sort of social care workers, and so on and so forth.
00:12:16.920 --> 00:12:21.879
All sounds very noble and that they are very important objectives.
00:12:21.879 --> 00:12:41.480
But of course, on the other side, as information governance professionals, as privacy professionals, our listeners will want to know what safeguards would you propose to put into place or have you put into place in projects that you've been involved in to ensure that the data doesn't or the system doesn't become a surveillance system.
00:12:41.480 --> 00:12:53.240
Of course, governments change, officials change, and we don't want to be in a situation where it becomes a way of keeping tabs on children as they grow up, monitoring adults.
00:12:53.240 --> 00:12:59.080
What are the safeguards that need to be put into place or that you would propose to put into place?
00:12:59.480 --> 00:13:01.800
So I'm absolutely passionate about this topic.
00:13:01.800 --> 00:13:11.240
And one of things that worries me greatly is just the huge amount of information that private companies and various nefarious actors already hold on our population.
00:13:11.240 --> 00:13:23.399
This actually gives us an opportunity to really debate all of the governance structures, all of the safety sort of mechanisms and secure mechanisms that we need in place whenever we hold any public data.
00:13:23.399 --> 00:13:27.320
Remember, these public data are already held on our various sort of systems.
00:13:27.320 --> 00:13:33.000
All we're talking about is properly connecting them in order that public services can be effective.
00:13:33.000 --> 00:13:40.519
And the biggest way that you can actually ensure that this is done for public good is involve the public in this.
00:13:40.519 --> 00:13:44.120
We shouldn't be doing this without taking the public on this journey.
00:13:44.120 --> 00:13:49.560
And one of the reasons we've been successful in Bradford is because the huge amount of community engagement we do.
00:13:49.560 --> 00:13:53.160
We genuinely want to improve outcomes for every child and young person.
00:13:53.160 --> 00:13:57.720
And that's something that is shared as a desire by our communities.
00:13:57.720 --> 00:13:59.480
So we work with our communities.
00:13:59.480 --> 00:14:10.759
We place our communities at the heart of these endeavors, and that ensures that everything that happens is being done for the public good, not for the benefit of a few individuals or a few organizations.
00:14:11.240 --> 00:14:13.879
That will be reassuring to our listeners.
00:14:13.879 --> 00:14:26.680
One of the podcasts that we have planned in the future is with Amnesty International because they've done a lot of work around the involvement of big tech in big government infrastructure and data projects.
00:14:26.680 --> 00:14:35.320
There's a lot of controversy at the moment with regards to the role that Palantir is playing in the NHS federated data platform.
00:14:35.320 --> 00:14:42.440
And there's been lots of concerns about their human rights record and what the final agenda is of these big tech companies.
00:14:42.440 --> 00:14:44.360
Are they hoovering up our data?
00:14:44.360 --> 00:14:45.960
What are they going to be using it for?
00:14:45.960 --> 00:14:53.560
So the idea that you are involving the public, you are consulting people who are going to be affected will be reassuring.
00:14:53.560 --> 00:15:05.240
And hopefully, those debates will come out as to which kind of companies, because inevitably, these systems and this connection, there will be an element of private sector involvement.
00:15:05.639 --> 00:15:07.080
I I completely agree.
00:15:07.080 --> 00:15:12.200
And again, I think this very process of starting this sort of debate is invaluable.
00:15:12.200 --> 00:15:28.600
Because the problem is we're just sleepwalking into a situation where we're giving our data away, or we think that uploading some data into cloud somehow is acceptable when we haven't really checked the extent to which we can retain the sovereignty of the data that we're uploading onto that cloud.
00:15:28.600 --> 00:15:51.720
And one of the things that we've been doing in Bradford is looking into investing in physical servers inside our NHS infrastructure so that we could be absolutely certain that the data that are held across our health services, but then the connected data sets, are as safe and secure and as robust to attacks as we can possibly create.
00:15:51.720 --> 00:16:07.080
And ensuring that some executive order written by some president in some country for dubious sort of reasons cannot suddenly lose our control of our public sort of other data.
00:16:07.080 --> 00:16:09.800
So, as I say, this is something we're absolutely passionate about.
00:16:09.800 --> 00:16:21.960
And we think that actually having this sort of conversation about connected data and what we can do to connect our data will really bring these issues out into the open and allow the public to really shape this debate.
00:16:22.279 --> 00:16:30.600
One of the examples of a project involving connected data, and which you've been involved in, of course, is the Bradford Connecting Data Project.
00:16:30.600 --> 00:16:44.680
Can you give us some examples of your work there where you have discovered something where you have identified certain needs which you would not have managed to do had the data not been connected?
00:16:45.480 --> 00:17:41.279
Yeah, so we've been very pleased to support a number of government initiatives and sort of priorities recently.
00:17:41.279 --> 00:17:55.039
So there's the independence prevalence review, which is really trying to understand why we have a special educational need sort of a crisis within the UK and why we have so much pressure on our autism and ADHD services within the NHS.
00:17:55.039 --> 00:18:02.240
And the reality is that these pressures are playing out across both health and education.
00:18:02.240 --> 00:18:09.839
So if you just look at the health data, it's very difficult to understand what are the levels of need within our population.
00:18:09.839 --> 00:18:19.680
If we just look at the education data, it's very difficult for us to understand how children's health barriers to education are playing out in the classroom.
00:18:19.680 --> 00:18:36.480
But by putting these sets of information together, we can suddenly start really understanding how we can improve our autism and ADHD assessment and support services in the NHS, whilst really improving the support that's provided in classrooms to children, young people.
00:18:36.480 --> 00:18:47.759
And what we can also see within the data is how that then will play all the way through to actually improving the rates of NEAT that we're seeing in this sort of country.
00:18:47.759 --> 00:18:53.359
So we can't have close to a million young people, not in employment, education, or training.
00:18:53.359 --> 00:18:58.160
And rightly, we're starting to look at what we can do for those young people.
00:18:58.160 --> 00:19:00.000
And this is part of the Milburn review.
00:19:00.000 --> 00:19:02.079
It's a big priority within government.
00:19:02.079 --> 00:19:06.240
But what we're able to show in the data is we need to start much, much earlier.
00:19:06.240 --> 00:19:17.440
And we need to be addressing that special educational needs sort of crisis if we really want to impact the number of young people who are having these adverse sort of outcomes at age 16 or 17.
00:19:17.440 --> 00:19:22.960
So these connected data sets are really allowing us to properly understand the root of the problems.
00:19:22.960 --> 00:19:29.600
And that means we can start tackling the root of the problems rather than continually reacting to the symptoms.
00:19:31.519 --> 00:19:32.799
That's a great example.
00:19:32.799 --> 00:19:40.799
What I find interesting is that you're starting with the child's experience rather than the organization's database.
00:19:40.799 --> 00:19:47.519
Normally, we tend to think of information as belonging to the GP or the school or the local authority.
00:19:47.519 --> 00:19:54.320
You're asking what that information collectively tells us about the child or children in a particular area.
00:19:54.320 --> 00:19:56.640
And I can see the benefits of that, particularly.
00:19:56.640 --> 00:19:59.360
I mean, you mentioned ADHD recently.
00:19:59.360 --> 00:20:06.000
There's been lots of uh stories in the press about the length of the waiting list for a ADHD diagnosis.
00:20:06.000 --> 00:20:10.320
So I can see the benefits there, but it's child-centered, isn't it?
00:20:10.720 --> 00:20:11.840
That's exactly right.
00:20:11.840 --> 00:20:18.640
So so I would argue strongly that that you know the NHS don't own the information about the sort of child.
00:20:18.640 --> 00:20:22.160
The school doesn't own the information about the child.
00:20:22.160 --> 00:20:25.600
Social care doesn't own the information about the child.
00:20:25.600 --> 00:20:33.360
What these public services have is information about their interactions with that child.
00:20:33.360 --> 00:20:43.440
And what we've got to start doing is moving away from doing what's best for an organization and start moving towards what's best for a child, what's best for a family.
00:20:43.440 --> 00:20:48.160
And that means really taking that family-centric, that child-centric view.
00:20:48.160 --> 00:20:50.400
And I think that's what the public wants.
00:20:50.400 --> 00:21:00.720
They don't want us operating in a way which is very, very, very convenient for an organization, but means that we fail to give a family the support that it requires.
00:21:00.720 --> 00:21:11.680
What we really have got to start with is the family, with the child, and start sort of saying, how do we make our public services fit that child, fit that young person, not the other way around?
00:21:13.440 --> 00:21:14.720
Absolutely, I agree.
00:21:14.720 --> 00:21:19.200
And as you mentioned before, I think we have to stop operating in silos.
00:21:19.200 --> 00:21:32.720
And to be fair, public officials have tried over the years, particularly in the area of child protection, to get out of those, that kind of mentality and to build systems as we've discussed before.
00:21:32.720 --> 00:21:41.759
All the benefits of connected data, Mark, uh rely very much on having the data, on having the complete picture about the child.
00:21:41.759 --> 00:21:49.519
You mentioned in the report data minorities, children whose records may be incomplete, inconsistent, or missing.
00:21:49.519 --> 00:21:52.160
What are the consequences of that?
00:21:53.440 --> 00:22:04.559
So the consequences are laid bare in so many serious case reviews that too often practitioners are trying to make informed decisions but on the basis of really sparse information.
00:22:04.559 --> 00:22:10.320
And the power of connecting data is you can suddenly start to synthesize a number of different viewpoints.
00:22:10.320 --> 00:22:16.720
So that allows you to make much, much better sort of decisions, but also allows you to identify where there are gaps in the records.
00:22:16.720 --> 00:22:28.400
And those gaps in the records may be because we haven't paid sufficient attention to really sort of supporting a child or young person, or it may just be limitations in our recording systems themselves.
00:22:28.400 --> 00:22:47.360
And this is another reason we're so passionate about starting to connect the data, because it allows us to start really challenging ourselves as to what we're collecting and are we collecting the information that we need in order to ensure that our services are provided to everybody, regardless of postcode or ethnicity or gender or social and economic position.
00:22:47.360 --> 00:22:48.880
And we saw that in COVID, right?
00:22:48.880 --> 00:22:52.559
That we saw that we're failing to record ethnicity.
00:22:52.559 --> 00:22:57.519
And this had huge practical implications for the way we were responding within the NHS.
00:22:57.519 --> 00:23:14.080
So gaining the process of connecting these data, having the debate about what information it is that we should be using to tailor our public services in a really effective sort of way, actually then allows us to start improving our very data collection processes themselves.
00:23:16.480 --> 00:23:27.360
So one of the benefits, as you've explained, of the connected data model is that we will end up hopefully improving our records management, improving the data.
00:23:27.360 --> 00:23:34.559
So that means better decisions in relation to children, which obviously is an important outcome and an important benefit.
00:23:34.559 --> 00:23:45.200
Now, the report says that the connected data model could improve outcomes for children, which we would understand, strengthen public services, but it also mentioned support economic growth.
00:23:45.200 --> 00:23:46.640
How will it do that?
00:23:46.960 --> 00:23:48.799
Well, through various sort of mechanisms.
00:23:48.799 --> 00:23:56.880
So the first thing that you get if you start better supporting children, young people, is a healthier, better educated workforce.
00:23:56.880 --> 00:24:03.360
And you need a healthy, well educated workforce if you to stand any sort of chance of growing the economy.
00:24:03.360 --> 00:24:07.200
So that's one very practical way that we can start seeing some of these sort of benefits that have come.
00:24:07.200 --> 00:24:08.000
Through.
00:24:08.000 --> 00:24:27.039
But it also opens up the possibility of creating new tools within the NHS or new tools within education that can really tailor provision of services, provision of educational approaches to individual children, sort of young people or adults.
00:24:27.039 --> 00:24:32.400
And that could be transformative in terms of turning around the fortunes of the UK.
00:24:32.400 --> 00:24:36.960
So there are real, real opportunities to make much better use of our data assets.
00:24:36.960 --> 00:24:41.120
And again, not for private companies per se.
00:24:41.120 --> 00:24:46.559
No issue at all if private sort of companies generate a profit because they're doing the right thing.
00:24:46.559 --> 00:24:56.400
But in and of itself, we can start making use of the assets that we possess within this sort of a country in order that everybody in the whole country benefits.
00:24:56.640 --> 00:24:59.120
And there are some great examples within the report.
00:24:59.120 --> 00:25:01.120
And we've already mentioned Connected Bradford.
00:25:01.120 --> 00:25:11.759
Now, in terms of the history of information sharing, just to recap, we have been here before, Mark, in terms of initiatives to encourage better information sharing.
00:25:11.759 --> 00:25:15.440
We had Contact Point abandoned in 2010.
00:25:15.440 --> 00:25:20.960
The ambition there was to make it easier for professionals to know who else was involved with the child.
00:25:20.960 --> 00:25:29.600
And then we had care data, where the potential benefits of data sharing were quite significant, but public confidence became a major issue.
00:25:29.600 --> 00:25:38.080
How are the proposals in the report regards to connected data different to previous information sharing initiatives?
00:25:39.120 --> 00:25:43.039
I think for a start, you can't fiddle around the edges.
00:25:43.039 --> 00:25:54.880
If we want proper connected public services and proper integrated professionals working together effectively, then we need to have a common understanding.
00:25:54.880 --> 00:25:56.720
We need a shared information base.
00:25:56.720 --> 00:26:04.960
So we really are not just sort of talking about a little initiative over here that tries to sort of patch up some of the issues that we're facing.
00:26:04.960 --> 00:26:13.200
We're really calling for a fundamental rethink of public service provision and how we can use connected data to support that public service provision.
00:26:13.200 --> 00:26:16.160
And there are a couple of barriers that we've identified.
00:26:16.160 --> 00:26:19.840
I think one of the first sort of barriers is people are very uncomfortable.
00:26:19.840 --> 00:26:24.880
And I share the discomfort of very sensitive health data leaving the NHS.
00:26:24.880 --> 00:26:38.320
And that's why our model uses the NHS safe, secure data environments as the spine where we pull in other data sets into that secure environment rather than trying to sort of share information out.
00:26:38.320 --> 00:26:41.440
And we think that's that's sort of transformative in and of itself.
00:26:41.440 --> 00:26:52.320
And the other aspect of this, and again, I think where where things have gone wrong in the sort of past, uh, is that rather than trying to do this at a national level, our model is a devolved model.
00:26:52.320 --> 00:27:04.480
And because it's a devolved model, it means that we can really start working with our populations, our communities in places to start co-producing these sort of solutions.
00:27:04.480 --> 00:27:08.640
And that gets around a lot of the interoperability issues, which are often problematic.
00:27:08.640 --> 00:27:25.360
It gets around a lot of the public trust issues because it allows that proper engagement with the community, with the people who live in that place, where everybody's coming together in order to build services that are really fit for purpose for those sort of communities.
00:27:25.360 --> 00:27:30.640
And that's tailored approach and that place-based approach, we think is very, very powerful.
00:27:30.640 --> 00:27:35.600
And of course, we now have a country that's moving in that sort of devolved direction.
00:27:35.600 --> 00:27:42.240
So we feel that the time is right now to bring all of these things together and hopefully avoid the problems of the past.
00:27:45.360 --> 00:28:09.840
And as you say, you're not proposing one giant national database, which previous initiatives have, and having it in the confines or within the protected space of the NHS's infrastructure, not impenetrable, but certainly better than having it in other areas and having it held by the private sector, will of course reassure people and build trust as well.
00:28:09.840 --> 00:28:17.759
Just talking about trust, Mark, how did you build trust in the connected data project in in Bradford?
00:28:19.120 --> 00:28:27.279
By holding conversations, by genuinely not wanting to do anything that didn't meet the desires of the communities that we're serving.
00:28:27.279 --> 00:28:29.039
At the end of the day, why are we doing any of this?
00:28:29.039 --> 00:28:36.000
We're doing this because we are passionate about improving the lives of the people who live in the Bradford district.
00:28:36.000 --> 00:28:47.200
So, of course, we want to work with the people in the Bradford district to ensure that what we're sort of doing meets their sort of concerns, that meets their desires for their for their families.
00:28:47.200 --> 00:28:56.480
So that engagement from the outset and that genuine desire to produce sort of something on behalf of our population really made the difference.
00:28:56.480 --> 00:29:00.000
And again, that's very, very difficult to do at a national level.
00:29:00.000 --> 00:29:05.759
But you really can do it when we're starting to talk at a local authority or an integrated care board sort of a level.
00:29:05.759 --> 00:29:11.120
So that's why we really feel that the devolved model is the only way this can be done successfully.
00:29:11.120 --> 00:29:14.400
Because if you don't do that, then how can you get that public engagement?
00:29:14.400 --> 00:29:17.519
And without that public engagement, how can you get the public sort of trust?
00:29:17.519 --> 00:29:24.320
So this is why we're so adamant that that devolved approach is the way to go forward from here.
00:29:26.080 --> 00:29:51.440
For me, that passion of wanting to make a difference in the community came across within the report because you've got quotes from people who are involved, not just professionals but individuals, the photographs of the children also, it gave that a strong impression that it was very much about the children, about making lives better and not just having another technically minded initiative.
00:29:51.440 --> 00:29:55.600
But some people will say one of the barriers is the legislation.
00:29:55.600 --> 00:30:02.080
You have called for more clearer guidance and more information sharing gateways.
00:30:02.080 --> 00:30:13.200
How did you, in terms of the Bradford Connected Data Project, ensure that the system is not just trusted but also lawful from a privacy, confidentiality, data protection perspective?
00:30:13.759 --> 00:30:20.080
So it's really, really important that we do everything ethically, safely, securely, and legally.
00:30:20.080 --> 00:30:39.680
And the approach that we took was to really take that child-centric approach and say, okay, what are the statutory responsibilities placed on our various organizations in order to safeguard children, provide them with appropriate health services, provide them with appropriate education services, provide them with appropriate social care services.
00:30:39.680 --> 00:30:54.559
And when we looked at all those statuary responsibilities, we realized that it was impossible for our various organizations to fulfill those responsibilities when they were unable to properly understand what was happening in that child's life.
00:30:54.559 --> 00:31:09.039
So we reframed the legal debate and started saying, okay, well, if these are all the statuary responsibilities and we need to connect the information in order for this organization to meet its responsibilities, is that is that an appropriate legal pathway?
00:31:09.039 --> 00:31:12.000
And it turns out that's an entirely appropriate legal pathway.
00:31:12.000 --> 00:31:24.240
And again, I think it's just the sort of difference between trying to say, uh, find your way around potential barriers and rather rethink this as a genuine effort to improve public services and start looking at the appropriate legislation.
00:31:24.240 --> 00:31:30.240
And I think when you start looking at the appropriate legislation, it becomes clear that actually, really, the impetus is on us to do this.
00:31:30.240 --> 00:31:41.200
And we should really be asking why we're not doing this, given the number of laws we start breaking if we're not able to operate in the way that the legislation is saying we should be operating.
00:31:42.320 --> 00:31:43.120
I agree.
00:31:43.120 --> 00:31:53.920
And sometimes the legislation is seen as a barrier, it's interpreted as a barrier, whereas, in fact, we've been saying for for many years that data protection actually enables information sharing.
00:31:53.920 --> 00:32:03.920
I mean, you mentioned the lawful basis, and for data protection professionals, they'll know about Article 5 of the GDPR, making sure that things are done fairly and lawfully.
00:32:03.920 --> 00:32:09.759
But when you look at Article 6, it talks about where it's necessary for compliance with a legal obligation.
00:32:09.759 --> 00:32:19.120
Obviously, lots of legal obligations in the area of child protection and public services, where it's necessary to protect the vital interests of the data subject or other people.
00:32:19.120 --> 00:32:23.840
And again, health, education, social work, they are vital services.
00:32:23.840 --> 00:32:35.279
And what you're trying to do is to ensure that those services are provided in the most efficient way, in the most holistic way for all members of the community.
00:32:36.559 --> 00:32:37.680
That's exactly right.
00:32:37.680 --> 00:32:42.880
And when you start doing the counterfactual and think, if we're not sort of doing this, where does that leave us?
00:32:42.880 --> 00:32:49.840
It leaves us, I think, in a very questionable legal sort of state as well, a very questionable ethical position.
00:32:49.840 --> 00:32:59.360
And again, it just takes us back to the point that when you start looking at things from the child's perspective, from family's perspective, then we really have systems that are not fit for a purpose.
00:32:59.360 --> 00:33:01.840
And that's got to change and it's got to change our pace.
00:33:04.240 --> 00:33:05.039
Absolutely.
00:33:05.039 --> 00:33:11.120
Just one last question in terms of the connected data project in in Bradford.
00:33:11.120 --> 00:33:13.680
Were there any findings that surprised you?
00:33:13.680 --> 00:33:18.320
Something that the connected data revealed that professionals hadn't expected?
00:33:20.000 --> 00:33:20.559
Hugely.
00:33:20.559 --> 00:33:23.279
I mean, every day brings up new surprises.
00:33:23.279 --> 00:33:33.680
And again, though, you know, the problem is that that we have such partial views of our different services, of our different sort of systems, that you know, the human sort of our mind starts making assumptions.
00:33:33.680 --> 00:33:38.960
And what you see when you start connecting the data is those assumptions are often wrong.
00:33:38.960 --> 00:33:41.600
So frequently we we're surprised.
00:33:41.600 --> 00:33:50.160
And that surprise often comes with a sort of pleasure of realizing that perhaps there are solutions to some problems that otherwise seem intractable.
00:33:50.160 --> 00:33:58.559
So one of our first examples of this is when we were looking at the large numbers of children in Bradford who were struggling to read.
00:33:58.559 --> 00:34:10.159
And of course, we're all sort of thinking about issues regarding phonetics, regarding uh exposure to rich language experiences in in early years and so on and so forth.
00:34:10.159 --> 00:34:16.639
And what the data showed us clearly is actually large numbers of children were struggling to read because they couldn't see.
00:34:16.639 --> 00:34:21.519
So in the health data, we could see that there were children who needed glasses who didn't have glasses.
00:34:21.519 --> 00:34:25.599
And then we could match that up with education records and suddenly realize, oh, guess what?
00:34:25.599 --> 00:34:29.039
Turns out that actually seeing stuff is quite important when you're learning to read.
00:34:29.039 --> 00:34:34.799
So it's these types of insights that that come as a surprise almost on a sort of daily basis.
00:34:34.799 --> 00:34:45.279
But it's these insights that we need if we actually to tackle problems like poor levels of literacy or high levels of NET or high levels of adolescent mental health.
00:34:46.960 --> 00:34:48.719
That's absolutely fascinating.
00:34:48.719 --> 00:34:56.559
So it's not just about child protection, it's about identifying the needs of the child in terms of education and other services.
00:34:56.559 --> 00:35:02.479
So you start off with the big picture, as in why are children not able to read in a particular area?
00:35:02.479 --> 00:35:13.359
They need classes, perhaps, but then that would filter down to the professionals so that the schools and the teachers can ask the right questions of the child and of the parents.
00:35:13.999 --> 00:35:16.479
And it also allows us to address the inequalities, right?
00:35:16.479 --> 00:35:25.279
Because often in our more affluent areas, families have enough social capital that they can kind of close the gaps that exist between the different services.
00:35:25.279 --> 00:35:29.599
When go to our disadvantaged areas, we can sort of see that that's really not the case.
00:35:29.599 --> 00:35:38.079
So, what this also allows us to do is just really start addressing some of those fundamental difficulties that are blighting the UK at the present moment.
00:35:38.079 --> 00:35:43.119
And that's also why uh we believe that this approach will help us grow the economy.
00:35:44.479 --> 00:35:45.119
Absolutely.
00:35:45.119 --> 00:35:47.359
And as you say, the data's there.
00:35:47.359 --> 00:35:58.479
It's just a question of connecting it and then obviously making sure that professionals who know how to interpret the data are able to get access to it and do their thing.
00:35:58.719 --> 00:35:59.679
Yep, absolutely.
00:35:59.679 --> 00:36:09.199
And that's the only way that we're going to properly be able to support every child and young person and make sure that every child and young person has the best opportunity to achieve and thrive.
00:36:09.199 --> 00:36:12.159
Uh, and clearly that's got to be the first obligation of a society.
00:36:12.159 --> 00:36:14.319
We're not looking after the next sort of generation.
00:36:14.319 --> 00:36:17.999
Then by definition, the future is extremely questionable.
00:36:18.319 --> 00:36:18.719
I agree.
00:36:18.719 --> 00:36:24.879
Now, Mark, the report that you're the executive author of has made 10 recommendations to government.
00:36:24.879 --> 00:36:27.439
What's been the response from government?
00:36:27.759 --> 00:36:30.079
We're delighted with the response from government.
00:36:30.079 --> 00:36:33.519
Uh, government are taking all of these recommendations extremely seriously.
00:36:33.519 --> 00:36:38.399
And we're really getting to the stage now where we're starting to talk about how do we operationalize these recommendations?
00:36:38.399 --> 00:36:40.319
What would this actually look like?
00:36:40.319 --> 00:36:46.959
And you know, this is difficult stuff because it is cutting across all of our various sort of asylos.
00:36:46.959 --> 00:36:53.919
So it does mean Department for Education, speaking to the Ministry of Justice, speaking to DWP, speaking to DHSC.
00:36:53.919 --> 00:37:01.039
And that's challenging, but it's definitely a challenge that governments seem to be willing to address.
00:37:01.039 --> 00:37:14.799
So we're delighted to be working with government and supporting government as it tries to see how it can make the fundamental changes that are desperately required within the country in order to give every child and young person the best possible start in life.
00:37:15.039 --> 00:37:15.439
Absolutely.
00:37:15.439 --> 00:37:33.359
And as recently, the government's announced or has enacted the Children's Wellbeing and Schools Act 2026, creates a statutory information sharing framework for safeguarding and promoting children's welfare, provision for information standards, a framework for consistent identifiers.
00:37:33.359 --> 00:37:40.399
We've also got the Every Child Achieving and Thriving Schools paper, proposes a new education data spine.
00:37:40.399 --> 00:37:43.279
Are you encouraged by those initiatives?
00:37:43.679 --> 00:37:44.799
Hugely encouraged.
00:37:44.799 --> 00:37:49.119
I think the School's White Paper is a really important part of the sort of picture.
00:37:49.119 --> 00:37:51.439
Really useful legislation.
00:37:51.439 --> 00:37:55.679
We've also got the NHS 10-year plan with its focus on prevention.
00:37:55.679 --> 00:38:01.359
And for me, prevention must mean children and young people, but we'll its focus on digital and sort of our data.
00:38:01.359 --> 00:38:05.279
So really encouraged by those those pieces of legislation.
00:38:05.279 --> 00:38:15.599
Also, really encouraged by the Independence Prevalence Review and the interim report that it put out, really encouraged by the Milburn report or its interim sort of a reversion.
00:38:15.599 --> 00:38:25.439
And I think that's really starting to show us that everybody's waking up to the issues that are playing out for practitioners, for policymakers, for the public.
00:38:25.439 --> 00:38:37.999
And hopefully really pushing us all to start that debate about how we can work together to improve public service delivery and how we can use connected data to really act as a facilitator towards that.
00:38:37.999 --> 00:38:45.599
And then the other thing, of course, which is really important right now is the move towards devolution and the model that we're proposing is a devolved model.
00:38:45.599 --> 00:38:57.919
So the increasing opportunity for mayors and combined authorities and integrated care boards to really seize the initiative within their regions is really, really exciting.
00:38:57.919 --> 00:38:59.759
So I think that the time's right.
00:38:59.759 --> 00:39:07.519
And it sort of feels like we're at that moment in time when we can start to get this over the finish line.
00:39:07.759 --> 00:39:10.479
We're certainly moving in the right direction, as you say.
00:39:10.479 --> 00:39:15.119
Let's imagine the government adopts all your recommendations, Mark.
00:39:15.119 --> 00:39:20.079
Five years from now, what would actually be different for a child and their family?
00:39:20.079 --> 00:39:22.639
What would be your perfect outcome?
00:39:22.959 --> 00:39:29.279
Well, hopefully in sort of five years' time, we've really learnt uh from five years of intensive activity.
00:39:29.279 --> 00:39:34.959
Because what we're talking about is really the creation of feedback loops that allow the sort of system to learn.
00:39:34.959 --> 00:39:39.599
And that learning has got to involve how we have our services operating together.
00:39:39.599 --> 00:39:48.879
So I'd love to see no further serious case review saying, oh dear, we failed to share information, and that's why we've had this tragic death of a child.
00:39:48.879 --> 00:39:59.439
I would love to see every child and young sort of a person having their neurodiverse needs sort of met, every child and young sort of a person entering employment, education, or training.
00:39:59.439 --> 00:40:01.679
And I think this is all within our grasp.
00:40:01.679 --> 00:40:18.159
If we really start focusing in on children, young people and their families and looking at how our public services can support those individuals and those sort of families, and how we can really have all of our services pushing in the same direction.
00:40:18.159 --> 00:40:22.239
I I think, you know, I think we're just in a complete different place as a country.
00:40:22.239 --> 00:40:25.039
And I think it's it's all highly achievable.
00:40:27.599 --> 00:40:28.239
I agree.
00:40:28.239 --> 00:40:30.959
The will seems to be there from government.
00:40:30.959 --> 00:40:33.919
Technical solutions, I am sure, can be built.
00:40:33.919 --> 00:40:44.799
It's just a question of ensuring, as you have done in Bradford, and as you are proposing, making sure that the systems are genuinely useful, lawful, and above all, trusted.
00:40:44.799 --> 00:40:48.479
So definitely that is something to look forward to.
00:40:48.479 --> 00:40:52.319
Mark, you've made a very strong case for connecting children's data.
00:40:52.319 --> 00:41:02.799
And I'm sure that our listeners, not just the information governance professionals, but the other professionals that uh listen to this podcast will want to know more, maybe even get involved.
00:41:02.799 --> 00:41:04.719
So, where can they do that?
00:41:05.439 --> 00:41:18.319
So, if you go to our Child of the North website, our Born and Bradford website, uh, then you'll see all of these reports, all of this information, uh, and you will see lots of opportunities to contact us and get involved and join the debate.
00:41:18.319 --> 00:41:19.999
So we love to work with people.
00:41:19.999 --> 00:41:30.239
Our whole approach is inclusivity, and we are always welcoming of anybody who wants to join our campaign to build a country that works for all children and young people.
00:41:30.559 --> 00:41:35.599
And if you'd kindly share those links with us, we'll ensure that we put them in the show's notes.
00:41:35.599 --> 00:41:42.319
And I gather that you have some events coming up where you'll be talking more about the projects, Mark?
00:41:43.039 --> 00:41:49.919
Yes, we're absolutely committed to disseminating our message and communicating.
00:41:49.919 --> 00:41:51.919
So we have lots of events that are coming up.
00:41:51.919 --> 00:42:01.599
If you look at our Yorkshire and Humber Applied Research Collaboration pages, uh, you'll see various talks and various meetings where people can join in the debate.
00:42:01.599 --> 00:42:02.559
Fantastic.
00:42:02.799 --> 00:42:05.919
Mark, it's been an absolutely fascinating conversation.
00:42:05.919 --> 00:42:06.959
We could go on.
00:42:06.959 --> 00:42:15.119
No doubt we will invite you back in five years or maybe earlier than that, hopefully, to talk about some more projects and how things are going.
00:42:15.599 --> 00:42:16.799
It's been a real pleasure.
00:42:16.799 --> 00:42:31.199
And as I say, we we're just really, really pleased that these debates are happening and really, really grateful for you two for pushing forward those those conversations and making sure that we're all talking about how we can build a country that works for all children and young people.
00:42:33.119 --> 00:42:33.919
Fantastic.
00:42:33.919 --> 00:42:35.839
Thank you very much for your time.
00:42:35.839 --> 00:42:39.519
That's all for this episode of Guardians of Data.
00:42:39.519 --> 00:42:47.839
A huge thank you to Professor Mark Mon Williams for sharing his insights and passions for a new approach to information sharing.
00:42:47.839 --> 00:43:00.799
At the heart of the Connected Data Report are some simple but powerful ideas that public services should be organized around the needs of the child and family rather than the boundaries between different organizations.
00:43:00.799 --> 00:43:12.719
And a new connected data model for information sharing would identify problems earlier, address inequalities, and give every child a better opportunity to achieve and thrive.
00:43:12.719 --> 00:43:19.199
And I think we can all support that ambition, but connecting data also creates responsibilities.
00:43:19.199 --> 00:43:25.839
The more information we bring together, the richer the picture we create of an individual and their family.
00:43:25.839 --> 00:43:32.959
That means we have to be absolutely clear about purpose, lawful access, security, and governance.
00:43:32.959 --> 00:43:45.039
We also have to think about what happens when governments, organizations, or priorities change and make sure that a system designed to support children doesn't gradually become a system for monitoring them.
00:43:45.039 --> 00:43:54.879
And there's also an important issue to consider with regards to the role of big tech companies in building and operating public sector data infrastructure.
00:43:54.879 --> 00:44:00.639
As these systems become more sophisticated, that's going to be an increasingly important debate.
00:44:00.639 --> 00:44:02.799
Who controls the infrastructure?
00:44:02.799 --> 00:44:04.479
Where is the data held?
00:44:04.479 --> 00:44:05.839
Who can access it?
00:44:05.839 --> 00:44:11.439
And what happens when commercial interests conflict with public sector interests?
00:44:11.439 --> 00:44:17.439
Mark himself highlighted the importance of retaining sovereignty and control over public data.
00:44:17.439 --> 00:44:27.999
For information governance professionals, there's an important role in ensuring that data protection and information governance are not seen simply as barriers to connected data.
00:44:27.999 --> 00:44:39.999
Our role is to help answer the important question, which is how do we connect data in a way that genuinely serves children and families, protects their rights, and retains public trust?
00:44:39.999 --> 00:44:47.999
Marx invited people to get involved in the work of the Connected Data Project, so do take a look at the links in the show's notes.
00:44:47.999 --> 00:44:53.839
If you want to learn more about the government's plans for children's data, have a listen to episode eight.
00:44:53.839 --> 00:45:00.719
If you enjoyed this episode of Guardians of Data, please share it with colleagues and others who work in information governance.
00:45:00.719 --> 00:45:05.839
It's a great way to help grow the IG community and support professional learning.
00:45:05.839 --> 00:45:10.399
Thanks for listening and join us next time on Guardians of Data.