Public Audit and Post-legislative Scrutiny Committee

Meeting date: Thursday, November 19, 2020

Official Report 577KB pdf

---

Data Collection and Planning for Outcomes (Key Audit Themes)

Agenda item 2 is on our key audit themes. Today’s theme is data collection and planning for outcomes. I welcome our witnesses: Stephen Boyle, Auditor General for Scotland, who joins us in person; Alex Hutchison, director of the Data for Children Collaborative with UNICEF at the Data Lab, who also joins us in person; Linda Hutton, research team manager at Citizens Advice Scotland, who joins us remotely; Dr Emma Miller, senior research fellow at the University of Strathclyde’s school of social work and social policy, who is representing the Personal Outcomes Network and who also joins us remotely; Claire Sweeney, director of place and wellbeing with Public Health Scotland, who joins us remotely; and Jennifer Wallace, head of policy at Carnegie UK, who joins us remotely.

I say to Stephen Boyle and Alex Hutchison that, when you speak, your microphone will be activated automatically, so there is no need to touch the buttons. If you would like to respond to a question, please raise your hand. To the witnesses joining us remotely, please raise your hand—there is a screen in front of me that I can see you on—or, preferably, type R into the chat function if you wish to respond to or ask a question. I will respond to you directly, or the clerks will alert me to the fact that you wish to come in. If at any point we lose the connection with you, I will come back to you at a later point.

I will structure the session around four themes, which are set out in the committee paper. Although the meeting will be structured on those four themes, I am happy for us to have a fluid conversation and discussion. The rigidity will be around the four themes, but not within what we discuss on those themes.

As our paper explains, this session is part of a series of evidence sessions that the committee is holding on key audit themes. The committee’s work is directly linked to the reports of the Auditor General, and a number of those reports have raised concerns about incomplete or poor-quality data and a lack of planning for outcomes. Examples include reports on early learning and childcare, self-directed support and children and young people’s mental health services. The focus of these sessions, including today’s, is to explore actions and steps that could be taken to bring about improvements in those areas, particularly given that the Covid-19 pandemic has highlighted the importance of accessible, quality data.

As the key themes have come up repeatedly in Audit Scotland reports, I will ask the Auditor General to start the session with some brief opening remarks before we kick off the actual discussion.

Good morning, everybody. Public services exist to support and improve people’s lives and wellbeing. Tackling priorities such as inequalities will take many years and decades to achieve. An outcomes-based approach helps to plan for the longer term the changes that we want to see and, critically, to demonstrate to all of us as taxpayers what has been achieved from public spending. Making well-informed long-term decisions relies on having high-quality data. It is central to deciding what spending is needed, how money can most effectively be allocated and how performance will be measured and evaluated over time.

However, it is not easy. Joining up data across organisations and sectors requires trust and strong and collaborative working. On top of that, maintaining a longer-term perspective when public services and finances are under short-term pressures such as those that we are currently experiencing is ever more difficult, but it is vital that we do so. Covid-19 is likely to have long-term effects on people’s lives, whether they are financial, physical or mental. Having high-quality data to measure those effects is crucial to making robust future decisions to help people to recover and to improve their outcomes.

Thank you, Auditor General.

To kick off theme 1, which is on the purpose and benefits of collecting data and planning for outcomes, I am particularly keen to hear from organisations that are not directly linked to the work of Government but are obviously involved in trying to move the Government in the right direction. I am interested to hear about the importance of data collection, the depth of that data collection and how the data then helps to shape your priorities and working. I am keen to speak to some of our external witnesses about that first to get a greater understanding.

Alex Hutchison, as you are right in front of me, you have drawn the short straw.

I cannot avoid eye contact with you, convener.

The Data for Children Collaborative with UNICEF has linkages with UNICEF, the Scottish Government and the University of Edinburgh. We look at a number of different types of data and how we can use those in new ways to improve outcomes for children. That can be existing routine data that is collected in the public sector. We look at Scottish children’s outcomes and have close linkages with Research Data Scotland as well as looking at what exists today in the data world.

We know that there are swathes of data out there. We are talking about data collection, but it could just be about access to data that already exists. The data is being collected, although we know that the quality of data is an issue. However, we need to look more widely than just at routine existing data. We need to consider what novel data sets exist out there. That could be anything from satellite imagery to telephone data to data from surveys that are run by loads of different organisations. How do we make that accessible and make people aware that it exists, that there are ways of tapping into it and that it can make a difference to people’s lives if it is used properly, in the right way and safely?

Do you mean not just Government-collected data but data collection across the board?

Yes—absolutely.

Within that, is there an issue with how the data is presented?

Yes.

Sometimes, data might exist but be presented so badly that it is hard to use.

It is hard to get a handle on what data exists across the whole landscape of the public, private and third sectors. What data exists out there? Who owns it? How do you access it? How do you apply for access to it? What is even in it? What are the data variables and the catalogues behind it? It is hard to get a grip of all that, because the landscape is massive. We need to be more systematised in how we present information to people so that they feel empowered to go and use it for benefit.

Excellent—thank you.

Linda Hutton, every MSP gets data from your organisation and uses it for local campaigns and to make pitches inside Parliament, so I will come to you for your reflections.

When it comes to the data collected by citizens advice bureaux, one thing that the Covid pandemic has taught us is the increasing importance of the information that comes through the door.

However, in terms of access, we need to take one step back from that. CAS works hard to maintain quality and have a cycle of continuous improvement, not only in the way that our data is collected but in the way that it is analysed and understood. From my experience previous to working for CAS, I know that many organisations out there already have a fantastic wealth of useful data, but they are missing someone in-house who not only knows how the data is collected and how it all relates to each other, but who can analyse the data and make it meaningful. Above all else, data tells stories. However, you need a particular skill set to make the data tell those stories.

It is my belief that more organisations than perhaps we believe already have data. They are not ready for other people to access it, because they do not fully understand how valuable the information is that they have and what stories that data tells.

For me, from my learning from CAS, before we get to the access point, we have to take a bit of a step back and look to see how organisations understand their data and what they do with it.

Claire Sweeney, you have obviously seen the issue from both sides, having previously been with Audit Scotland, when you no doubt wrote about the need for more quality data. You now have a key role in Public Health Scotland’s continuous data analysis, particularly in the Covid situation. I am sure that you have some interesting perspectives from both sides.

Yes—I have a huge range of issues to talk about. I will start from Public Health Scotland’s perspective and acknowledge that there is much more work to be done. We are a relatively new organisation that is made up of some legacy organisations, so we are reviewing all our programmes of work. One of the messages across our organisation with our partners has been that we want to have a relentless focus on impact and, for me, a bit of humility about those other organisations that have already been touched on and which have a history of doing this very well.

For example, I spoke to Penumbra this week. The questions that it asks the people with whom it works include, “What gives you a sense of hope?” and, “How do you live a good life or what do you need to live a good life?” Traditionally, we have not been good at asking such questions, but it is happening out there. For me, part of the answer is to have a bit of humility about working with a wide range of partners to bring all that information and intelligence together to make sure that services are far more focused on making a difference. That is part of what we want to contribute. We bring a certain amount of skills and experience, but we definitely do not have all the answers, so partnership for us is absolutely key.

How many external requests do you get for data from organisations or campaigners to help them shape their work, given that Public Health Scotland obviously has loads of data and people want to use it for campaigning purposes? Do you get many external requests?

We get a lot of external requests from the public and from organisations, particularly third sector organisations. There are two parts to that. One is all the more visible stuff—the big data sets that we publish routinely. You will be well aware of all the cancer information that has come out recently around Covid and so on.

For me, the much more interesting part of the job is all the partnership working that we are doing. For example, we work with lots of organisations on what makes a sustainable community and what makes for good and healthy places to live. Our data is part of that, but not all of it.

Another example is that we work closely with all the big charities that deal with tobacco, drugs and alcohol issues. We work with them in partnership to bring all that intelligence together. Rather than a lobbying message, the data is used to work out what can make things better for people in Scotland. A wide range of partnership working goes on. It is not always national in its nature and it is not perhaps as visible as it could be—we want to do more on that—but lots of partnership working is going on to share data.

I suppose that there is a wider question about how much the data helps to shape policy making, but we will come back to that.

I will hand over to Jennifer Wallace from Carnegie UK.

It is a pleasure to be with you this morning to talk about this really important issue. I hope that you are all familiar with Carnegie’s work but, for those of you who are not, I describe us as being a critical friend of the Scottish Government on the journey around the national performance framework since 2010, when we ran a round table on measuring performance in social progress. We have been on that journey, and we are also an organisation that covers the United Kingdom and Ireland and has an international lens, so I hope to bring some of that to today’s conversation.

I agree with Claire Sweeney’s points about openness and transparency and about the ability to use different types of data and bring them together. For me, Scotland is missing a structure for how we link our aspirations and outcomes to what are essentially lag indicators in the NPF and to the lead indicators on which we have a wealth of data, which can be administrative data or data from surveys. As Linda Hutton said, some of our most timely data is collected by service providers and people at the front line. At the moment, in Scotland we have no system for understanding that. We tend to talk about data as if it is one big thing rather than a data chain of understanding how those things can all be connected to build up a picture of what is happening in Scotland.

Another point is that we need to ensure that the data that we collect is good enough to address the inequalities in our society. It is not always capable of doing that, particularly when we are talking about some of our minority communities, where surveys can struggle to pick up those numbers.

Thank you. I will hand over to Bill Bowman.

09:45  

Stephen Boyle spoke about there needing to be trust and strong, collaborative working. Does blockchain feature in the future of data?

I think that what you are suggesting is correct. A number of the contributions have talked already about the need for that collaborative leadership and partnership working. Effective data collection through to the monitoring of milestones and, ultimately, outcomes will not be achieved by one organisation alone.

I will test the boundaries of my understanding of blockchain. We can see the implications of new technology. As I understand it, blockchain will fundamentally change my profession—the auditing profession—and that of accountants and create a more open, accessible and visible set of transactions and data. That and other technologies will inevitably play a huge role in how we deliver, track and monitor data from data collection through to outcomes, but I must confess that, as you can probably tell, we are still thinking about and grappling with what those new technologies might mean for our own work and data collection in the round.

Does anybody else want to chip in with a quick definition of blockchain?

I can give you my version. Having tried to get my head around it and having spent time reading articles and listening to conferences, the way I try to explain it in my terms is that it is another piece of technology—one of many—that gives us security and creates trust in the transactional relationships of data. It is not the be all and end all or a panacea. It is a technology that happens to have an unbreakable code and algorithm around it that nobody could hack into and, therefore, just keeps certifying as data moves along that chain—I think that Jennifer Wallace mentioned a data chain—and makes sure the data that is moving from A to B to C cannot be manipulated or handled incorrectly. A lot of people refer to blockchain, but we are really talking about whether we can handle data safely and what technologies are out there to help us to do that.

If you are content with that response, Mr Bowman, I will move on to Dr Miller.

I will pick up on a couple of points that others have made, but first I want to clarify what the Personal Outcomes Network is. We are a multiagency network working across health, social care, housing and children’s services. Our members come from across Scotland. We tend to meet quarterly to try to focus on issues such as how to get good enough data, recording outcomes for individuals within services, and also, to a certain extent, using that data.

Reflecting on what Linda Hutton was saying, I agree that a lot of organisations in Scotland are currently working hard to collect good personal outcomes data, but there are significant issues around analysis and use of that data. We have some useful learning, though. We had a project a few years ago called meaningful and measurable. Penumbra, which has already been mentioned, was one of nine practice partners in that project. We made quite significant progress that year in improving understanding of how to collect, record and use outcomes data.

One of the best and strongest features of that project was, in fact, partnership working. There was something about getting different organisations in the same room every so often to exchange some of the challenges that they were facing with this whole agenda, as well as exchanging tips and tricks for how they had managed to progress. The issue of working in partnership is fundamental to progressing this whole agenda.

Thank you, Dr Miller. I believe that Claire Sweeney wants to come back in.

I asked to come in when I had not heard the end of the blockchain question, so I am glad that I avoided that.

For me, the key is the openness of data. That is the point that I was going to make. I have been with Public Health Scotland since June, and, even in that time, I have seen quite a significant stepping-up of data being accessible and of the speed with which it is released, with people trying to get it out as quickly as possible. There is much more to do around that, but we are starting to reap the rewards of some of the new technology that we can use around data.

To come back to the relationship point that has just been mentioned, for me what we are engaged in is not an oppositional thing. We are trying to get people in a space where we are all trying to improve outcomes for people in Scotland. There is no hiding. There is no criticism of each other, other than in a constructive way. We need to work together and bring all of our efforts together to best effect. That is the way we are trying to move things forward.

This is a really important point about trust in and the transparency of the data—particularly in the era that we are living in, with the level of scepticism and endless references to fake news—and about the quality of data that is being collected and how that is evidencing progress or otherwise. That equally supports the scrutiny that this committee and the Parliament will do and accountability. That stems from good quality data right at the outset that all service users and participants can track and monitor, so that there is always one version of the truth. The times that we are living in emphasise the importance of high-quality data, starting right at the collection point.

On that point, Auditor General, are you suggesting that the main barrier to the use of data is not the collection of it but how it is accessed—that is, the platform from which it is accessed and how people get to it?

I think that both those elements are involved. It is about what happens at the outset of implementing policy development and policy changes. Over a number of years, Audit Scotland has said that it is important to design in high-quality data collection methodologies right at the start of that process, alongside the development of and implementation of the policy matters.

On access, this is something that CAS is working towards at present. We have very clever people constructing a—[Inaudible.]—for us that will allow us to share a reasonable amount of data via a dashboard, so that anyone, from someone in the street through to—[Inaudible.]—or MSPs, can access information directly.

One thing to mention on that is that our case management systems are confidence based, to put it very politely. Our data is also very complex. While we are more than happy to be open and to share data, we still have to be quite cautious about what we put into the public domain because some of it could be quite easily misunderstood. While we are working towards open access or much easier access for people, there will be organisations like ourselves that still have to be careful about what is put into the public domain. It goes back to what the Auditor General was saying about the fake news issue. We have some figures that, in the wrong hands, could be wildly misinterpreted. Because we are concerned about the quality of our data and what our data tells us, we are very cautious about what is released to the public domain.

Thank you, Linda. How easy is it for you as an organisation to access public sector data or Government data to help inform your own campaigning work?

We have never had any issue accessing that data. The things that we access most often would be things such as the Scottish household survey. That is made available through what I think is called the data bank. We can access that quite easily. If we are looking for specific pieces of information, I would feel quite confident that I could go to the relevant people within Scottish Government and come to some kind of agreement about what we could access and where. I would not see any problem with that at all. Reciprocally, we are always happy, if we can, to feed into anything, essentially. There are quite a few projects and pieces of work, some by the Scottish Government, that we have fed into in recent times.

I have a few additional points. On the data gap issue, we are part of the consortium that produces Understanding Scottish Places. We have experienced significant difficulty in getting data at the level of Scotland’s towns. As we are moving forward with our localisation agenda and with the commitment to 20-minute neighbourhoods and so on, that data will become even more important. A reflection on our experience is that it is very difficult to get data at the lower-level, super-output area level to build up that picture. You can get population data from estimates from the census and you can get economic data but, once you start going into social outcomes, it is incredibly difficult to get that data in a comparative way. I just highlight that as a specific gap.

I want to make a point based on our international work. Data does not so much give you one version of the truth as help you to have a shared conversation on why things are happening. One of the difficulties that we have as a society—and that a lot of countries have at the moment—involves the fact that we are flinging data at each other in an unconstructive way. For example, if we all agree to the indicators in the NPF, or we understand why one type of data may be preferable to another type of data for the conversation, we can have a constructive debate on why certain indicators are moving in certain directions. At the moment, that conversation is difficult for us to have because of issues around data availability and, frankly, data literacy as well.

There is a suite of work that needs to be done to be able to get to the point where we can have that shared conversation, and it is one that I think Scotland has started well on. I used to be the convener of Evaluation Support Scotland and in that role I worked with a lot of third sector organisations. There was a sense that there is a lot of rhetoric in Scotland about outcomes but, when it comes down to what is written in a contract for a third sector organisation, the process is often not outcomes focused—that is not the detail that people are being asked to collect. They are being asked to collect output data and they are being asked to collect input data—how many people came to their courses and those kinds of things.

I go back to the argument about needing a golden thread. We need some sense of what we expect all different levels in Scotland to be working to in order to build up a picture that means that we are all working to outcomes rather than different parts of the system using different performance measurements.

Graham Simpson would like to ask a supplementary question. I see that Claire Sweeney and Emma Miller want to come in. I will go to Mr Simpson first, and I will ask Dr Miller and Claire Sweeney to keep their responses brief, because I am keen that we move on to theme 2. We have probably rolled into that anyway, but we will do so formally.

I will be really brief. Linda Hutton said that data can be misinterpreted. I guess that it can, but there is also a danger, and we have had this debate in Parliament recently, that bad news in data can be sanitised if you hang on to it for long enough, clean it up and present it in the way that you want to present it. There is a danger in not publishing the data as well. What do people think about that?

That is a broad, open question. I am going to go to Claire Sweeney and Dr Miller, and then we can come back to this in the next theme, because it is part of our future discussions.

I will keep it short and say that that is clearly wrong. That should not be happening and, as far as I am aware, it is not happening with any of the information that we have. We are bound by strict information governance and statistical quality controls. I am happy to let the committee know more about that, but that should not be happening for our work.

On the point about accountability, one of the issues that we face is that people are just not held accountable. Jennifer Wallace mentioned that. For example, what are our leaders across the system held to account for? Traditionally, it has been the things that we are very good at measuring and have data around, and that really does need to shift if things are to change.

To make a point that refers back to the meaningful and measurable project and, in fact, the work of the Personal Outcomes Network in general, we tend to make a distinction between information for improvement and information for performance purposes. That also refers back to a point that Linda Hutton made earlier about the use of data for improvement. In the meaningful and measurable project, we found that organisations were much more able to be honest and to exchange information about some of their struggles and perhaps some of their data that suggested that things were not going so well if that took place in the context of improvement. When there is a tendency to focus outwards for performance purposes, that is where that honesty or the level of frankness, perhaps, starts to change a little bit, because of a fear of blame or judgment and so on. That suggests that there is something about the culture of information use that we need to consider as well.

10:00  

Thank you, Dr Miller. We will move on to theme 2, which is on barriers to collecting good-quality data and challenges to moving to an outcomes-based approach. I ask Alex Neil, who is joining us remotely, to kick off this theme for us.

Data is extremely important to good governance, whether it is the private sector, the third sector or the public sector. Both as a minister and as an MSP, down the years I have found very frustrating the fact that the public sector as a whole does not have an information technology strategy that allows us to collect the data that we need more efficiently, more cost effectively and more effectively.

A very good example of that is the national health service. There are 22 or 23 boards involved in running the national health service in Scotland. The national health service spends about half a billion pounds a year on IT systems in Scotland, yet we are bedevilled by the fact that a lot of the systems do not talk to each other and, therefore, cannot collate the data that we need and which should be available for decision makers.

The situation has been exacerbated with the integration of health and social care because the social care IT systems are not compatible with the health service systems. Very often, the social care IT systems are not even compatible with other internal local government systems.

To my mind, that is the single biggest barrier to getting the kind of data that we need regularly and reliably. I would welcome the witnesses’ comments on the total lack of a comprehensive IT strategy within the public sector. We are probably spending well over £1 billion a year in Scotland on IT systems across the public sector, and we are not getting value for money because they do not talk to each other.

Who fancies kicking off on Mr Neil’s challenge? Alex Hutchison is volunteering—he has the disadvantage of being in my eye line.

Systems not talking to each other is a problem, and it is a problem that is really hard to fix. I say that with a Royal Bank of Scotland background. I used to work at RBS, where we had to integrate RBS and NatWest systems. That is mini model of the public sector: there were a multitude of different organisations and legacy systems, and the aim was to tie them all together.

The learning from that situation is the importance of being clear about what a data strategy is versus what an IT strategy is, and separating the two. Getting the technology in place and the systems to talk to each other is a piping exercise; it is a matter of getting infrastructure built so that A equals A in both systems. The data—the core variable—can mean such different things in such different contexts that people have to be really aware of what they are using it for.

Let us think about some of the data variables that the public sector might be looking to link up and have talk to each other. The information that somebody has given in a form about early years funding, selecting from a dropdown list, might mean one thing in the context of a child’s performance, but it might mean a completely different thing to somebody in finance who wants to understand spend and how much is being spent on the funding models.

It is a question of breaking down the data to make sure that there is understanding of why we are asking for the variables. The data literacy part is in bringing out the skill set and upskilling people so that they know why they are filling out a particular form, why they should take the time to do that and why they should care—because it will have an impact and influence 10 stages down the line.

The situation is absolutely a problem. I am thinking about an experience I had at Audit Scotland when we were working with a very big partnership in Scotland on health and social care. The partners had done a fantastic job, with others’ help, to link up all their data on health and social care. We got very excited and thought, “This is revolutionary, this will have all the answers, it will be great”. However, the message from the partners was that the information was overwhelming and they did not know where to start.

There is so much information and data in primary care that the technical issues in joining it up are particularly problematic. We know that that is a problem, but it is not the only problem. The additional problems are the “So what?” questions. What do you do with the information? How do you interpret it? How do you make sure that it leads to demonstrable changes?

There is a package of things that are incredibly complicated. We are trying to help address many of them through the contribution that Public Health Scotland and others can make. It is important to acknowledge that joining up data is an issue but that it is not the only problem we face.

I go back to what Jennifer Wallace and Alex Hutchison said earlier. The lack of a shared language and understanding of what data means is a huge problem to the concept of joining things up. Because people tend not to measure things consistently—what is A to someone will be X to someone else—we are still far away from joining up. For me, it is about taking a step back and looking at the data that we hold, how we record it and how we have it speak to the needs that we are trying to identify.

On the question of consistency, for example, in CAS we keep an eye on the Scottish census and the categories that are used within it. Broadly, our demographics will be based on the categories used in the census. However, if you picked up two or three data sets from anywhere, you would probably find a lack of consistency among them over things like age categories and the way that a housing status is identified. Until we have consistency over the very basic items within data sets, it will be difficult to have joined-up thinking.

One other barrier to having a joined-up service can be funder requests. As you know, citizens advice bureaux are charities. We rely on the UK and Scottish Governments as well as the local authorities for funding. Sometimes when projects are put in place, people who make the decisions on what needs to be measured are not data people—to put it politely. They do not necessarily understand that what they are asking for does not meet the need.

To fix that, I would like to see a more open dialogue. When a funder comes to third sector organisations such as ours and says, “We need you to measure X, Y and Z”, I would like them to be open to us saying, “We do not have X but we have W and W is almost X and W will give you this.” We need to be able to challenge what has been asked for in a constructive way so that we can create data that both meets the needs of our funders and is accessible and understandable to a much wider body of people.

I see Jennifer Wallace and the Auditor General want to come in, so I will go to them first before bringing in Colin Beattie.

I am thinking back to the time that I spent working with local government when I was working at the Scottish Consumer Council—quite a while back now. One of the difficulties that we had was people talking about the importance of what I would now class as service exceptionalism. They would say that their service and their user group was different and that they had to measure it in a different way. It is that sense of needing to do things our own way—needing to build from the bottom up—that sometimes cuts across the collective understanding.

There is a way around that, but it has to be a very large, co-owned and co-designed conversation on what we are measuring collectively and why we are measuring it, to make sure that we get consistency. If people are not involved in the process, they will not understand why the need for consistency is so much more important than the exceptionalism that they might have within their own service. There have been so many attempts to achieve harmonisation over the years, and it is always incredibly difficult to do.

The other point that I would make in talking about IT systems and the public sector is that so many of our services are delivered by the third sector or private sector. There is another layer of complexity within the discussion, and there are significant issues about data transfer that need to be overcome to ensure we are getting a full picture of what is happening in Scotland.

I sense Mr Neil’s frustration around the issue, given the recurring theme in the committee’s work.

Alex Hutchison makes an important point. The data strategy applies as well as the IT strategy, and both things matter equally. What is interesting, particularly from a public body’s perspective, is the wide and disparate nature of public organisations that exist in the country. We are a country of only 5.5 million people, and it feels intuitive that IT and data could be made to work more straightforwardly than they currently do.

One thing that we should be asking ourselves about is the lack of progress that has been made in some of the shared services in the country. There are so many different structures and infrastructure that exist. Why has there not been enough progress in making connections between the public bodies? Ultimately, progress will come down to leadership and the incentives that exist in different public bodies to make the step change.

One thing that has been consistent in the Auditor General’s reports, in the almost 10 years now that I have been on the committee, is data: the accuracy of data and how it is used to determine or validate policy.

On the one hand, we get data about the numbers, which helps to determine where resources will be put. On the other hand—and this is where I see the data falling short—we do not see information about the outcomes. Is the money being spent in a way that is commensurate with the outcomes that are being achieved? In other words, is the intended outcome being achieved?

Consistently, we are not seeing that measured, but where scarce public funds are being invested, it is important that we have that information. I do not know how the cultural shift can take place and how we can move into that particular field, because I do not see it happening in many cases across the public sector and elsewhere. I would be grateful for the panel’s view on that.

That is a really important point, and things have not been good enough. I can see that through the Audit Scotland report, if nothing else.

In our organisation, we are in the process of planning the new business, and we are putting outcomes at the heart of that, being ruthless about trying to capture information on what difference the work is making and reporting on it. You will see more on that in due course.

We must also recognise that it is so much more complicated than it seems. For example, we have a tool that will not just produce data on things that have happened but which increasingly tries to predict things that might happen relating to policy. The tool basically takes in a lot of information and builds models. People who are planning policy can use it to see what the impact is likely to be on inequalities if they do X, Y and Z.

We need more of those tools to help support the development of policy and decision making about various interventions across Scotland. That is the case not just in healthcare but across the whole system. The tool can be used particularly in relation to factors in the economy. We would like to see that work developed so it is a matter not just of reporting on the data but of knowing what is happening in the future. It is really important that we do more of that work.

In CAS, we are very lucky that we have such a broad range of data. We use it to feed into our policy work, our advocacy work and our campaigning work.

One key thing in CAS that I think makes a difference when it comes to the quality and importance of the data is the fact that our senior leadership team understands its importance. They are happy to engage with the staff within the service who understand the data, who interpret the data and who create the IT systems to hold the data. They listen to their knowledge and expertise to continuously improve and develop the way that data is collected and interpreted.

I wonder whether that happens in other organisations, because data often does not become important until something needs to be reported. The annual report comes around and somebody says, “Okay, we need data to explain this, this and this”. All of a sudden, they realise that they do not have the data that they need or that the data is not of an adequate quality to meet the need. Essentially, I am saying that data is not a part-time job. If we want good data, it is something that has to be dealt with daily. It has to be an on-going process and not just something that is thought about when someone needs a report.

10:15  

If I understand the member’s comments correctly, we are talking about outcome budgeting, which I would probably more readily refer to as wellbeing budgeting.

There are developments in relation to shifting budgeting towards the wellbeing outcomes that we seek for the population. Scotland is most definitely on a journey in relation to that, probably following New Zealand and its approach on wellbeing budgeting. There has definitely been a step forward, but it is an additive approach. There has not been structural change in how we think about budgeting and how we think about spending money.

I know that the committee has considered reports on early years in particular. One thing that we are struck by as an organisation is the fact that we do not know how much we currently spend on our youngest children. If we do not know what we spend, we cannot begin to answer the question of whether we are spending it in an effective way.

We recognise that Scotland is on a journey to a wellbeing budget, and we recognise that it has a journey to go, but at the moment there is no pathway between the traditional budgeting and the wellbeing budgeting.

We are running a project with Children in Scotland, Cattanach and Katherine Trebeck to look at how to develop a wellbeing budget for the early years, thinking particularly about chunking the budget up for outcomes per population group. If we cannot take the whole budget as one goal, we may be able to take a population group at a time and shift it towards outcomes. The report on that will be published early next year, and we are very much looking forward to sharing our findings with you at that time.

I will respond possibly in part to the question of mapping out intended outcomes with a particular policy, at national level or at a more local level, and programme outcomes. We have made some progress in Scotland in the last few years on outcome mapping, which can involve diverse partners coming together to identify what the intended outcomes are. I have a couple of points to add to that.

The approach comes from contribution analysis, and there is a real understanding that we are looking at the contribution of different partners towards outcomes, not attribution. It is less of a direct cause-and-effect approach than a question of understanding how different agencies and, indeed, individuals can contribute towards outcomes.

An important part of that work is not just of tracking the intended outcomes but being mindful of the unintended outcomes. What additional benefits are emerging that we did not anticipate in the first place, as well as the less beneficial outcomes? It is a question not just of focusing on the outcomes that we think we will see but of keeping a broader focus.

My answer aligns to Emma Miller’s point about what we mean by outcomes, how measurable they are and how easy they are to measure.

There might be a number we could put on something. My ultimate success measure of what I am doing is how many children’s lives around the globe I am improving. It will take me a lot of years to be able to come back with a number, so we have broken down our success measures into the things that will make that happen. What are the inputs and outputs and the outcomes that will have an impact?

If we are able to measure the meaningful steps and changes that we are making in certain areas, we will know that if we keep nudging those parts on—if we collaborate with a broader set of expertise, if we bring in new data sets or if we bring in new academic partners to look at different ways of modelling—we will make a difference.

My recommendation about how we measure outcomes and impact, from a data and public sector point of view, is to make sure that we are looking at a step before. It will be difficult to measure and know whether a particular investment has had a particular outcome in 10 years’ time, and it will be difficult to be patient enough to see it. What are the things that will nudge it forward? What are the positive impacts from the investments?

Thank you. Willie Coffey is next; he is joining us remotely.

Good morning. I apologise for being slightly late in joining the meeting.

This is a really interesting discussion that we are having about data. A problem that I have always encountered as an MSP is trying to get data that actually relates to the constituency that I represent in Parliament. Often, over the years, data comes out about this, that and the other thing. When I ask the follow-up question, “How does that affect my constituency?”, I can never get an answer because the data always relates to the local authority or to NHS Ayrshire and Arran. It used to relate to the different police boards and units, but it never, ever relates to my own constituency or component parts of my constituency. I was interested to hear some of Linda Hutton’s comments earlier about never being able to get data based on a town and so on.

The fundamental problem is around collecting the data. If you tagged where it came from when you started collecting it, that would enable you to build up a picture. Then when folk like us ask how it affects our constituencies, we would be able to get an answer. It is not just the organisations here that face that problem. We have the same problem at Scottish Government level. How many times have we heard the message, “That data is not held centrally,” or something like that? The fundamental issue is that, when we collect the data, we should also think about tagging it with where it came from—for example, what community or postcode it came from—so that we can build that picture up. Do the witnesses share those views and what should we do to try to improve it in the future?

While I wait for someone to flag that they want to come in, just to amplify Willie Coffey’s point, so often, we get that response back because there are different bodies involved in collecting different parts of the data. That is a real gap and a real problem and, fundamentally, it needs to be addressed. I see that Alex Hutchison wants to come in and then Jennifer Wallace and Claire Sweeney, so there is some activation of interest.

It is a really good question and it is a challenge. The data will exist; the data is there. They know where the data came from, but the issue is the complexity of who has collected the data, who owns the data, to what level of granularity the data is about a data subject and an individual and, therefore, they have aggregated and aggregated.

The local authority keeps it in one system, in one version and one set of data, and then it submits it to central Government, so it is only submitting a certain version. There are so many hand-ins and hand-offs and, rightly, there is protection around certain streams of data or certain pieces of information, but almost to the point where we are so cautious and so worried about the general data protection regulation that we keep thinking of it as a preventer rather than a protector. We should be using GDPR as a tool and an enabler to understand how to handle data responsibly. We are on a journey of learning how to do that properly. That goes back to data literacy and making sure that local authorities feel that they have a good grasp of the security of their citizens—

[Inaudible.]—GDPR, though, because Mr Coffey will have had that problem long before GDPR came in, and the answers that we get would have come long before GDPR came in. I think that the Government can hide behind GDPR, but is it GDPR that is the problem?

It is about people protecting data. Whether it is GDPR or not, it is about people thinking, “I cannot give that out. I have to be careful. I will get in trouble,” which is the right attitude: they have to care about owning their data. However, the way that we are collecting it, housing it, reformatting it, aggregating it and moving it around means that the data loses a lot of its integrity as it is flowing around systems.

I do not doubt that that is true and I do not doubt that GDPR is a challenge, but there is something about not wanting bad-news data out there either. We have to be alive to that fact. We will now hear from Jennifer Wallace and then Claire Sweeney.

Can I come back in, convener?

Yes. Jennifer Wallace and Claire Sweeney want to respond to your first question, and then I will get you back in. Is that okay?

Yes, thank you.

The cultural aversion to data sharing existed before GDPR. For me, it goes back to a very human response; when people do not necessarily understand what it is they own and what the proportionate risks are of sharing that, it leads them to be overly cautious about levels of data. I am not talking about administrative case data about children in schools; I am talking about population-level data for place. You get the same reaction from people regardless of whether there really is a risk of exposing people’s personal data, so there is a cultural aversion.

If I go back to our experience on Understanding Scottish Places, which I referred to earlier, to produce that we needed to have a consortium of very committed people for at least a couple of years working on it, with full-time staff working on it to corral data into local areas. What we reflected on—and others will have reflected on this as well from their experience—is that administrative data in particular is collected based on administrative boundaries, but that is not how people live. People live in towns and cities and places that they themselves relate to.

If we want to talk about having a data conversation with the public, it needs to be based on places that they understand, allowing them to enjoy playing with the data and thinking about how it relates to their local area, rather than it being set up at a level that has very little resonance to their own lives. If you see data for NHS Greater Glasgow and Clyde, that statistic does not tell you very much about what is happening in your community. We understand so much more now about how people live in their local communities and how that sense of understanding place is crucial to their ability to work together to improve their places. I just wanted to provide a little more information about our experience when we did that and how time consuming it was, which it ought not to have been.

The opportunity that we have now, particularly because of the things that we have seen through Covid, with an increased focus on place and communities, is to start to move this forward at pace. We will certainly be involved in a bit of work, with our partners, around thinking about what makes sustainable communities, 20-minute neighbourhoods and so on. There is an opportunity to start to move forward much more fully on that. Part of it is to do with thinking about why we collect this stuff in the first place. Are we clear that the information is there to help to support improvements for the lives of people in Scotland? Because we are not clear about that in all instances from the start, it means that we often cannot tell. That is the issue we are dealing with. It is complicated and it is very long-standing.

It is not a GDPR issue. GDPR is a recent thing. It is about understanding the boundaries and the communities that we think it is appropriate to tell the public something about. You will not hear me saying this very much, but our Westminster colleagues are far better at this than we are. Westminster always reports all its data pertaining to the constituencies that make up that Parliament, but we do not. I could ask any question of the Scottish Government or of any of the agencies that are sitting in front of us today about any particular aspect of anything in Kilmarnock and Irvine Valley and nobody will be able to give me an answer, because we do not recognise that as a boundary that we think we should report on. It is about understanding the community boundaries that are relevant, as Jennifer Wallace mentioned.

10:30  

For me, it is about understanding data and all the different community boundaries that we have in Scotland and being able to report data on that basis when people like you and I come along, inevitably, and ask what is happening in our community. It is about that, so it is good to hear that although we have that problem and people recognise it, we are looking for ways to solve it so that perhaps in the next session of Parliament, we can begin to tell people in the communities that we live in what is actually happening in their community.

Thank you. I will go to Alex Neil next and then Linda Hudson wants to come in.

I share Willie Coffey’s frustration and the frustration of others on the panel this morning because, quite frankly, the head of statistics in the Scottish Government should be able to sort out the problem of getting more stuff reported on a constituency basis.

Can I look at the wider question, where as a society we could try to organise data in such a way that we get maximum use of it without necessarily having to continually put more resources in and maybe not get the value for money that we always anticipated? The Carnegie Trust and CAS have highlighted their frustrations. I know the committee will discuss this in more detail later, but I think that we could take a bit of a lead in this. The issue is not beyond the wit of man or woman. Data is collected at an individual level. We are all individuals. We all live in households—it may be a household of one, but we all live in households. Household data can be collected at household level. Households go to make up postcode areas. Postcode areas make up health board areas, parliamentary constituencies, council wards, council areas, communities, community council areas and all the rest of it.

If we start with the basics of the units of data and what we are trying to collect, there is a need for some kind of overarching framework at Scottish Government level—certainly, to cover the public sector in Scotland—agreed with the public sector agencies, on how at least we could all collect data relating to each of these units, households, people, postcode areas and all the rest of it. If you collect the data by postcode area, it should be easy to then put it into constituency-level data, for example.

Quite frankly, this is not about GDPR or about something that is impossible or expensive to do. It is about political will and the will of the civil service to do it. I think that the Scottish Government is the ideal body in Scotland to do it. We are a small society of 5.4 million people. It should be easy to do. Every other country seems to do it a lot better than us.

If we had an agreed framework, which people collecting data throughout Scotland could then use, knowing that everybody was working to that framework, we could get into a much better position over time. I do not underestimate the difficulty but, at the same time, I sometimes think that we build a mountain of a molehill. It is not a molehill, but it is certainly less than a mountain.

To go back to Mr Coffey’s point about reporting at constituency level, we do report regularly at both Holyrood and Westminster constituency level. We are lucky that we have a data set large enough to be able to do that. It is only where the bureaux are not represented that we cannot provide that. We can come across difficulties; anyone would come across difficulties reporting at constituency level. It is down to the specificity of the question. The more detailed information you want, the more likely the sample is to be too small to report on. For example, we could say that within Mr Coffey’s constituency, X number of people sought advice on benefits. Potentially, we could not say how many people in a particular area of that constituency sought advice about the young carer grant, because it might take it down to just a handful of people. When it comes to GDPR, we cannot stray into that territory, because we risk identifying individuals. I assure Mr Coffey that if he wants to come to me at any point in the near future for a report on his constituency, I am more than happy to provide such a report.

Thank you, Linda. One of the reasons why we were keen to have you on the panel is that Citizens Advice Scotland is a good example of how data can be broken down by constituency and local area. Our wish is that we could have that approach much more widely across the board and across the public sector in general. I think that that is a general demand from across the Parliament.

I ask Graham Simpson to kick off on theme 3, which is how to improve data collection and planning for outcomes.

Before I do that, I note that the information that I have had over the years from my local citizens advice office in East Kilbride has been very good. It provides local information. Citizens Advice Scotland is an excellent and very helpful organisation. It appears to be able to share useful statistics. Maybe there is a message there for others.

I want to explore how we get better, and I will refer to a report by the Auditor General for Scotland on city deals. I think that it was by your predecessor, Auditor General. I will not ask you to comment on city deals, because you might not know about them, but the report makes the general point that they are collaborations between the Scottish and UK Governments and councils. They are spending an awful lot of money, but one of the criticisms is that they do not appear to have set out at any point what they want to achieve and how it should be measured.

In a number of projects in my area, millions of pounds have been spent on road projects with no obvious benefit to anyone, just for the sake of it. There will be other examples around the country. When we are spending lots of money, do we need to establish at the start—before we spend the money—what we want to achieve and how we are going to measure it? I am not asking you to comment on city deals specifically, unless you want to do that, but do we need to get things right at the start, before we spend the money?

Yes—absolutely. I think that we can help to support organisations to do that across a range of service areas. Thinking particularly about the city region work, I note that we are working with a couple of partnerships around exactly that issue. How do we get the public health voice in? How do we have a clear focus on outcomes, if not from the start, then as soon as possible? It should be absolutely built in from the beginning.

On the previous discussion about levels of data, I wanted to come in briefly and mention that, for some areas, we have an enormous amount of detailed data. In the health system, we can look at things by individual and then by postcode. We can cut it in so many different ways. However, that is not true for most parts of the public sector.

Thinking in particular about the budgeting side and large expenditure, there are a couple of things that we could do collectively to improve. One of them takes me back to the point about openness and transparency. I have been working for non-governmental organisations in Scotland for 20 years and I have never seen a pre-expenditure assessment. It is difficult for me to know how the Government goes about doing that and what we can contribute, as outsiders, if we are unable to look at the evidence and the data that is put together in order to propose spend. For me, there is a point about openness and transparency at the heart of this.

Secondly, Scotland has a public finance manual, which I am sure you are all familiar with. It was of course updated for the national performance framework, but in my view it was not updated substantially and significantly. It was more about people saying, “Here’s something that you might need to think about”, rather than being about a theory of change and how we go about putting inputs into the system through the financing of projects in order to achieve the outcomes that we seek.

In that regard, I note in particular that the public finance manual still points to the Treasury green book. The Treasury green book has had some additions about social value, as we would expect, and about wellbeing measurement and the role of subjective wellbeing in that, but it is very much driven from the perspective of the UK Government approach. It is not driven from the perspective of the Scottish Government outcomes approach. If we were able to have a manual that really nailed the question of how we understand the theory of change for outcomes, I think that we would see improvements in the whole system.

The committee will be familiar with Audit Scotland’s reporting on how money has been spent, how well it has been spent and the connections between the NPF and its own financial reporting. Jennifer Wallace is quite right. Even if we look at the Scottish Government’s consolidated accounts, they tend to report fundamentally that the Scottish Government has broken even on the measures on which it is held to account. We do not yet see a clear enough connection between the consolidated accounts and “Scotland’s Wellbeing”, which is the report that sets out how the country has performed against the national performance framework.

My predecessor called repeatedly for better use to be made of the consolidated accounts so that the performance report sets out what has been achieved in outcomes for the £40 billion-plus of public expenditure. There is a way to go on that front.

I agree with Mr Simpson’s point. When people are designing complex policy, the outcomes—as we have talked about a number of times this morning—should routinely be captured in the design phase and, thereafter, in the milestones and reporting. Mr Simpson mentioned city deals, but the point applies equally to the comparable report on the expansion of early years and childcare provision.

Auditor General, you and your colleagues have written multiple reports that talk about leadership, different areas working together, the collation of data, the need for collaboration and the need for the right skills. Do you want to amplify that and say a little more on what you mean by that in practice and what change you believe is needed across the Government?

I am happy to do that, convener. We have already touched on a couple of points to do with culture. How is the implementation of projects measured? What are the incentives for leaders across the country and how is their performance assessed? With the national performance framework, we have made great progress as a country in starting to move away from inputs and, to an extent, outputs to the aspiration around outcome measures, but there is still a way to go.

It now feels like a time, almost, to take stock as to whether the infrastructure around our measuring and reporting is sufficient to make the kind of step change that I think we all believe in—of course, the Parliament signed up to the national performance framework. Whether it is the public finance manual or the role of the accountable officer, it all matters and it all counts. It is important that the Parliament and the various boards and committees around the country are equipped with the right tools to measure outcomes. At the moment, that does not feature prominently enough in the conversations that the various public bodies around the country have. What are the outcomes that the organisations are supposed to be delivering?

It is really complicated. I think that we have seen that through this morning’s conversation. However, I come back to another point that we have touched on a number of times. In a country of our size, we should feel intuitively that we can take the steps that will lead to transparency and appropriate comment on value for money and what has been achieved for the money that we invest in public services.

If we go back to what Alex Neil rightly said about political will, what is the blockage? As Colin Beattie said, issues have consistently been raised for over 10 years—and that is just the time that he has been on the committee—about the need for proper, good quality collection, dissemination and sharing of data and the need for adequate skills and leadership. What is the blockage? Is it simply to do with the will, or is there a layer underneath that that also needs to be resolved?

10:45  

It is incredibly complicated. It is undoubtedly cultural as well. If I recall correctly, the committee asked my predecessor similar questions about the extent to which some brave decisions will need to be taken and a consensus built around them. The key themes report talks about police numbers being a good example. To an extent, the measure of success is often held to be about the number of police officers that we have in the country, rather than shifting to be about the outcomes that the police service is delivering. The same applies to some of the performance measures on which the NHS is held to account.

For the Government and others, including members of Parliament, there needs to be a consensus on the functions and delivery of those organisations. It is about a will and leadership from across Parliament, and also from the civil service and public leaders.

When you say that it is cultural, what do you mean by that? Is there a bad culture just now in data collection?

Culture is what it is, convener, but what we are seeing—we will maybe come on to this—is short-term pressures, to an extent, rubbing up against the ability to step back and say, “What can we deliver in the medium and longer term?” This also goes back to how we assess performance and how people are held to account, whether they are public leaders or politicians. As I asked in my statement, are short-term pressures getting in the way of making some of the longer-term changes that we need?

On Graham Simpson’s point, is it about people wanting to avoid or delay bad news stories? If something is hot, people could take three or four months to collect the data and get the data analysis out there, by which time the story has moved on, or the world has moved on. Is there some truth in that?

I would have to think about specific examples. I can understand why there would be an incentive, perhaps, for people to behave in that way but, as others have said, there is an ethics in some of this. Actually, the timing of the reporting is taken out of the hands of those who could have that incentive. It is important that, at the outset, in designing and commenting on a project, we know when it is going to report. It takes it out of the hands of individuals who could seek to manage the message if it is clear when a project will report at milestones throughout its duration.

I will make some similar points. To me, part of the problem has been that we have had too much focus on the short-term, reactionary stuff and not enough on the longer term aims and sticking with it. Some of the change that we are discussing is generational, which is hard when we are trying to demonstrate year-on-year impact.

The approach that we are taking is to encourage teams to think about the short-term impact that can be made, the medium-term impact and the long-term impact. It is not the case that we will be unable to demonstrate any change over a generation but then the switch will go on and it will be fixed. There are steps that we can take in the interim. There needs to be recognition that some things will not be achieved until much further down the track, but that does not mean that we are not doing the hard work now that needs to be in place. That long-term planning is key.

For me, there is something about it being really hard and therefore, perhaps, not as much of a pressing priority as other things have been. With the pressure around Covid, the impact on reporting and on all our lives and services has been very significant. As an organisation, we are trying to deal with the Covid response but not lose sight of the much longer-term and, we could argue, more important impact that Covid is going to have; the impact on inequalities; and our role in helping to reduce that over time.

Finally, there are a load of legacy issues. It is really hard to switch that stuff off. I think that there needs to be a bit of bravery around saying, “That was fit for purpose before, but we’re in a very different place now. We need to be rigorous in challenging ourselves. Do we need to stop doing things or do them differently?”

As the Auditor General said, some outcomes take a long time to be achieved. When there are short-term projects with short-term funding, it can be incredibly difficult to report on the outcomes that people think that we should be able to report on.

I am thinking of an example. Before I came to CAS, I did some work on addictions. It is unrealistic to expect someone with a long-term addiction to recover fully within a year. With a short-term programme, someone might say, “We want an outcome to be that this person is no longer using drugs after 12 months”, but that is not going to happen. It is important to have outcomes that measure the short-term, smaller gains rather than the medium and long-term outcomes.

As others have said, there is an issue with performance culture, the focus of that and the way that it can distract from the potential benefits of a focus on outcomes. The personal outcomes network focuses on changing practice at the front line. It is important to be clear about what outcome-focused practice means. It involves a shift away from focusing on people’s needs and deficits to focus on their strengths, and it means changing the conversation to flow with that. It takes a long time to change that culture in an organisation.

Scotland is doing well on outcome-focused practice, although there is some way to go. We need a performance culture that supports meaningful conversations rather than constraining them by putting in place measures that do not support them. There are also equalities issues attached to that. Meaningful conversations that are tailored and adapted to the individual allow more meaningful engagement, for example with people with addictions or traumatic backgrounds, who need that more detailed approach to engagement. It is important to facilitate that rather than having performance indicators that work against it.

On the point about constituency boundary data, Public Health Scotland has a brilliant database that shows us all the Covid data. We are coming to that topic next. We can drill in and look at very small communities to see what is happening. The point that I have made throughout this discussion is that it will not tell me what the picture is in the Kilmarnock and Irvine Valley constituency; we are rarely told that. We can see communities at a smaller level, but we can never ask what the picture is within our own constituency boundaries. That is common across the board. If Citizens Advice Scotland has information on a constituency basis, that is brilliant. However, in general, nobody provides that, and we should aspire to having that in the future.

I think that all parliamentarians would whole-heartedly agree with that.

In our final minutes, we will move on to talk about the impact of Covid-19, which is theme 4. That has been added because it is clear that the way in which we collect and disseminate data, the way in which data has evolved in fast decision making and policy making, and how we scrutinise policy making have moved into sharp focus with Covid-19. I will give a moment to reflect on that.

I suppose that that is an issue for Claire Sweeney first, because Public Health Scotland is a key part of the approach. How much are quality data and the speed of data helping to drive quick decision making and policy making? How can parliamentarians help to use that data to hold decision makers to account and—I hope—to constructively challenge?

Obviously, the Covid pandemic has been incredibly difficult. There have been heart-breaking impacts on families across Scotland, but the immediate threat to health and life is not the only concern. We are equally concerned—indeed, some of us are more concerned—about the longer-term impact on people who live in Scotland.

Covid’s indirect impacts on health and wellbeing, health and care services—members have seen the cancer statistics that we recently published—the economy and wider society are significant, and they will be long term and enduring. As the committee well knows, things were already difficult in Scotland in terms of inequalities. Public Health Scotland was established to help to try to move that issue on, and we are actively involved in that work.

Members will not be surprised to hear that Covid has meant not the easiest start to establishing a new body. We have shifted a large proportion of our resources to immediate Covid work. Members probably want me to highlight to the committee ways in which we are doing that.

We are helping organisations to make decisions about how to respond. That is at the Scottish Government level, but it involves all our public bodies and communities. It includes things such as the national data that have already been mentioned and which members will be familiar with, and advice and guidance to communities, not just on Covid and trying to contain the spread of the virus, but on things as broad and wide as encouraging people to do more physical activity, the impact of people being stuck in their own homes, and trying to minimise the health impact from that. A huge, wide and varied amount of work is going on in relation to guidance and data, for example.

We are also helping to support the limitations and contain the spread of the virus. As the committee knows, we support the test and protect approach. We help local and national bodies through our public health scientists and through our expertise to advise people on how to respond to local outbreaks. Members will not be surprised to hear that that has been a very significant part of our work.

We are working with all partners to think about recovery and renewal. We are thinking about, for example, how to get the health system back up on its feet and how we can help partners to find evidence-based solutions to build resilience into communities locally, and we are helping Scotland to prepare for any future outbreaks, and keeping a close eye on certain other issues that are prevalent across the world.

I could talk at great length about that, but I hope that that gives the committee a good feel for the range of work that we are involved in at the moment.

We outside Government appreciate all the work that has gone on to try to create accessible and timely information on the pandemic so that we, as citizens, can understand what is happening. I thank you for all of that.

My comments are about the next stage rather than what has happened over the past nine months. One issue in planning for recovery is that we have a history of privileging our economic projections rather than our—[Inaudible.]—or other social outcomes. We need to think long and hard about that because, from what has happened in the pandemic, we understand far better the relationship between our economic outcomes and our health outcomes. We expect—I am sure that Public Health Scotland colleagues will agree with this—that the economic recovery might happen faster than the social recovery. That lag may take a lot longer. The scarring effects that we know happen from unemployment, for example, and the long-term effects of mental health issues go on for many years. We need to be careful in our construction of the recovery and in how we think about evidence that we give equal weight to the evidence that we have and the predictions that we make on health outcomes and other social outcomes, and recognise that they are much longer term. If we can find ways of having a conversation about that collectively and collaboratively, that would be incredibly helpful.

I am conscious of the time, so I will take final reflections.

On the Covid question and the reference to good-quality data driving decision making, the key for the public is the trust that they can have in Government decisions if good-quality data is used. Tying together transparency and openness as a continual message in communications and decisions is really important for people.

Does anyone else who is joining us remotely want to make any final comments? They do not.

Before I close the session, I invite the Auditor General to reflect on the issues that have been discussed and to draw out some of the key points.

11:00  

Thanks, convener. The conversation has been full, open and interesting, and many interesting points flowed from the committee’s report on the key themes. I will touch briefly on a number of things.

Perhaps we take for granted the importance of data skills. I refer to the point that Linda Hutton rightly made. We need to ensure that all of us are being equipped with the right skills. We have talked about the transparency of data, the use of data across the system, and how much that matters from the design of projects through to the milestones and the reporting of the outcomes that flow thereafter. That bridges into the point, which the convener has mentioned a couple of times, about clarity on when data and information will be reported and their availability, and information being owned by everybody, not necessarily just those who are in possession of the information at the time, so that it is clear to all the users when it comes through.

We have heard about the complexity of the landscape and the number of organisations sometimes getting in the way of things; the importance of partnership working and collaborative leadership from the Scottish Government and public bodies; and the importance of the role that our third sector colleagues play. Intuitively, it seems that a country the size of ours can do that work and can make a difference. As Dr Emma Miller mentioned earlier, we have undoubtedly made progress as a country, but it feels like there is more opportunity to make progress on how we report our public information against outcomes.

I acknowledge the terrible impact that Covid has had on individuals’ lives and the economy—no doubt it will have such an impact for many years to come—but some of the risks that exist in the short term can get in the way of the long term. We have seen examples of that in the past. We know that decisions are being taken at pace. As Claire Sweeney rightly said, we cannot lose sight of the pace at which we are taking decisions, but we must remember the longer-term implications and sustain the progress that we have already seen.

Thank you very much, Auditor General.

I thank the Auditor General, Alex Hutchison, Linda Hutton, Dr Emma Miller, Claire Sweeney and Jennifer Wallace for joining us. That was a fruitful discussion, and I genuinely thank all of you. We look forward to following up the discussion in the weeks and months ahead.

I now close the public part of the meeting, as we are moving into private session.

11:03 Meeting continued in private until 11:23.