7. Dr Audrey MacDougall

DR AUDREY MacDOUGALL (affirmed).

Questions From Counsel to the Inquiry

Lady Hallett: Ms Arlidge.

Ms Arlidge: Thank you very much, my Lady.

Dr MacDougall, you have – thank you for your assistance in providing your evidence to this Inquiry. We see a witness statement from you, reference INQ000346964. It’s just been brought up on the screen. I hope it’s familiar with you. I believe on page 23, it will be redacted in this version, but you have signed that statement.

Dr Audrey MacDougall: That’s correct.

Counsel Inquiry: And you’re happy with the contents of –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – the statement, that they’re truthful –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – to the best of your knowledge and belief.

There are going to be a series of areas I take you through during the course of your evidence. I will try myself to be as slow as I can be for the stenographer, who is doing sterling work in the background. I would ask you also to try and keep your answers nice and slow. I realise we’re at the end of the day, and I can only apologise for that.

Dr Audrey MacDougall: Certainly, that’s fine.

Counsel Inquiry: You are – starting from the beginning, you wear a number of hats in Scottish Government in terms of analytical, statistical, research hats. Is that a fair way of putting it as a broad spectrum to start off with?

Dr Audrey MacDougall: I think to try and just encapsulate the period covered by the Inquiry, at the start of that period I wore two hats, if you like, I was the Scottish Government’s chief social researcher, which is a professional role, looking after social researchers within the government and looking after the promotion of social research within the government. I was also head of what was called the communities analysis division, which was a division made up of different types of analysts, researchers, statisticians, economists, who provided analysis covering a range of areas, poverty, social security, housing, a range of different areas. So when Covid started, those were the two roles that I fulfilled.

I then moved to establish the Covid modelling and analysis hub, to specifically look at doing the same job, if you like, but solely focused on Covid issues, so looking at providing a wide range of evidence and analysis but related to Covid issues.

However, I didn’t give up my communities job for about a year, so there was about a year when I held three posts.

Counsel Inquiry: And shortly after you commenced working, setting up the modelling and analytical hub – I’m sure I’ll get that wrong multiple times – MAH, for the –

Dr Audrey MacDougall: MAH.

Counsel Inquiry: MAH. Shortly after you joined that and set it up, you recognised that the scope of the work that was involved no doubt was enormous and you asked Mr Roger Halliday, who has just given evidence, to join you in that hub?

Dr Audrey MacDougall: That’s correct. We felt that we had a very good complementary set of skills that myself as a social researcher and Mr Halliday as a statistician, we knew each other well, we had worked together before, so we felt that that would be a very good combination. And then in practical terms the hours that were being worked by the hub and the need for that senior oversight, you know, for very many hours every day meant that it was impossible for one person to do the job.

Counsel Inquiry: And in due course, early on but in due course, you were also very heavily involved in setting up and the development of the four harms strategy and providing evidence and analysis in that regard?

Dr Audrey MacDougall: That’s correct.

Counsel Inquiry: To go to the very sort of genesis of the MAH, my Lady has already heard evidence – sorry, comments yesterday from Jamie Dawson about the set-up of things like SGoRR, obviously that was pre-pandemic?

Dr Audrey MacDougall: Yeah.

Counsel Inquiry: It was a system that was already in place?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Is it correct to say that SGoRR, having been activated in light of the pandemic and the need for responding, requested your analytical approach or your evidence and your assistance in the very early stages –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – early March to say “We need eyes on things from a statistical and modelling analysis” –

Dr Audrey MacDougall: Yes, that would be correct, yes.

Counsel Inquiry: So you were requested to provide that sort of analysis from about 4 March 2020?

Dr Audrey MacDougall: So that was when I became involved, but there were other people who had been providing analysis before then under the auspices, say, of the health analysis colleagues, or there may have been other colleagues in other parts of the office who were starting to get involved as well. The whole organisation was turning itself towards looking at Covid, so I wouldn’t want to give the impression that there was nothing happening before then, there certainly was activity going on. But when SGoRR invited me to become involved, it was really recognising that this was a step change to any crisis that had been dealt with before and would benefit from having a central co-ordinating and specialist division. That would then draw in work that was happening in other parts of the office as well.

Counsel Inquiry: So you were therefore to able to provide a sort of central focal point –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – and leadership role in respect of multiple areas –

Dr Audrey MacDougall: That’s right.

Counsel Inquiry: – of information and modelling –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – coming to your attention and effectively drawing together threads from different departments –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – who were presumably all carrying out their own individual assessments from different, in different, transport looking at one aspect of things –

Dr Audrey MacDougall: That’s correct, absolutely, yes.

Counsel Inquiry: When you took leadership of the MAH, how many people were sort of at your disposal, as it were?

Dr Audrey MacDougall: Well, I think we probably started in week one with about five people, and one of my first tasks was to actually staff the division correctly, so that was the – indeed my first week or so spent in actually making sure that we had sufficient staff of the right grade and of the right quality and the right skill, and that was undertaken by drawing out the relevant staff from other parts of the Scottish Government.

From my role as head of profession and working with the other heads of profession in Scottish Government, I had a reasonable oversight of the talent that we had available to us, so I went shopping and asked for particular people to come and work with me, and gradually built up, you know, a reasonable sized division.

Counsel Inquiry: You set out at paragraph 15 of your statement that you agreed effectively a programme of work with your colleague – with the key –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – Scottish Government directors?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Is it the case effectively you had something of a blank sheet of paper and you had – there was a discussion that went on between the various directors and yourself saying “This is the sort of thing we can assist with, these are the things that are on our sort of radar”?

Dr Audrey MacDougall: That’s exactly it, I mean, this is the initial programme and, as you will see going through the statement, it changes over time in terms of the actual work we do, but the work always generated either from commissions from ministers or from other colleagues, particularly policy colleagues, or the work could be generated by ourselves where we proactively felt that there was some gaps in the evidence base or gaps in our knowledge, and we initiated work to fill those gaps. So it could go either way, it could be a commission or it could be something that we put forward.

Counsel Inquiry: Just looking at those various – again, like my predecessor in standing up here a moment ago, please don’t get too complicated in these answers – but looking at the initial programme, we’ve got things – we’ve got developing the population base –

Dr Audrey MacDougall: Mm-hm.

Counsel Inquiry: Could you just very briefly explain what a population base –

Dr Audrey MacDougall: Sure.

Counsel Inquiry: Probably saying to you briefly explain something that is an incredibly complicated issue is a bit difficult, but population base models: is that looking at who is a member of the society – the population that you’re looking to model?

Dr Audrey MacDougall: So I’ll just say a little bit which hopefully might help in terms of modelling, and I should preface this by saying I am not a modelling expert, I had some modelling experts in my staff, but broadly speaking there were three types of modelling carried out during Covid. There was epidemiological modelling, which was the modelling that gave us the R number, the growth rate, the infection rate, and that’s what SPI-M-O was concerned with, and that was at the core of understanding what was happening with the pandemic.

Then there was what was called operational modelling, and by its very nature – you can get the idea – operational modelling was taking that epidemiological modelling, using that to say: well, what does that tell us about cases, what does that tell us about potential hospitalisations, ICU, so you could use it for operations, how many hospital beds might I need.

Then you have policy based modelling, and that’s what this refers to, is taking that epidemiological modelling but then using it to try and say: well, what would happen if we did X or we did Y, if we implemented different types of scenarios, and that might be either things that we’re doing, ie things that the government had chosen to do, or it might be just different scenarios. Well, what if we think, instead of 80% of the population getting infected, it would be 60%, what would that mean? So … does that help?

Counsel Inquiry: It does.

Just going through some of the other points there, so leading on responding to commissions from SGoRR, again you say you’re not a modeller, but this is where SGoRR, headed by Andrew Morris, comes and says to you “Please can you give us a model about the likely transmission rate in” –

Dr Audrey MacDougall: So – indeed, so we could be asked: could you model what might happen if the transmission – if, you know, 60% of people became infected, 70%, or could you model what might happen if R was 2 or R was 4 or R was 6, you could model on that basis. Commissions from SGoRR that could also consist of work that wasn’t modelled, it might be, for example: what do we know? I mean, you know, what’s the latest on the science and transmission then? What is it telling us? Or it might be: what do you think people in the street are saying, you know, what’s happening? So it could be a range of different types of evidence.

Counsel Inquiry: I’m asked to remind you – remind myself that I misspoke by saying SGoRR was headed by Andrew Morris. It’s SCAG that was headed by Andrew Morris, I apologise.

But nevertheless, the same sort of approach, so SGoRR, because SCAG was subsequently set up and –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – were also asking for data and modelling to be carried out and to, on that same sort of basis; is that right?

Dr Audrey MacDougall: On the same sort of basis, but with SCAG, when it was established, initially Roger was the – was a member to represent our division, and then when Roger moved on, I became the member of SCAG to represent our division.

So it went – again, there was a kind of a both-way relationship with SCAG. We would present the work we had done to SCAG – to look for commentary, critique, discussion – or SCAG could ask us to produce the particular pieces of work, and we would go and do that.

Counsel Inquiry: Then updating, controlling and sharing the central assumptions and parameters that everyone should be working to; is that a sort of proced – that’s not modelling assumptions necessarily, that’s much more – is that more in the sort of “these are the key performance indicator styles” approach?

Dr Audrey MacDougall: Could be modelling assumptions but, it would be just generally: here’s what we’re assuming about the pandemic at the moment, based on the current state of what we know in terms of the science, in terms of what SAGE is telling us, in terms of what SPI-M-O is, and ensuring that everybody across the office had that same broad understanding of what was happening.

Counsel Inquiry: Was the MAH asked to provide specific advice on specific policy questions from the outset or from early March, or was it more sort of general “Show us where things are going”?

Dr Audrey MacDougall: So it started off obviously with “Show us where things are going”, but as soon as government moved to a position of wanting to introduce NPIs or wanting to, you know, make changes of any sort, then we were asked if we could model through the impact of certain of those changes that were going to be made. So that became a regular – a regular occurrence throughout the whole of the pandemic.

Counsel Inquiry: From – if I might take some sort of specific examples, were you asked, for instance, to model issues about discharge of patients into care homes and the effect of that?

Dr Audrey MacDougall: No, we did not model discharge into care homes.

Counsel Inquiry: Were you asked, in terms of care – appreciating that there was subsequent issues in terms of subgroups about children and education – before those subgroups came about, were you asked about modelling into or providing evidence in terms of things like education closures and the like?

Dr Audrey MacDougall: So we were asked to look at what would be the impact, let’s say, on R and then the subsequent case numbers of the closure of schools, that would be a valid modelling –

Counsel Inquiry: When was that sort of – I appreciate it’s not meant to be a memory –

Dr Audrey MacDougall: Oh, goodness. It would have been done at various points in time, because schools were opened, closed, and then partially opened, you know, so there was – it would have taken place at different times, it would’ve done that kind of modelling.

And that would have been based – just to clarify, as well, the basis on which it would have been done – it would have been based on assumptions that would have come from SAGE.

Counsel Inquiry: I apologise, Dr MacDougall, I’m being asked if you could just try and slow your responses slightly.

Dr Audrey MacDougall: I apologise.

Counsel Inquiry: I apologise. I know it’s very difficult.

At the very beginning, 4 March, 10 March, 12 March, when SGoRR come and ask you to have greater input?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Of course you’ve already given evidence that there was embedded analysis in each different department, for instance.

Dr Audrey MacDougall: Yes.

Counsel Inquiry: The – I think it’s uncontroversial that knowledge of the existence of the threat of Covid was growing.

Dr Audrey MacDougall: Yes.

Counsel Inquiry: And the early COBR meetings in which people were, you know, recognising that there was a need to ascertain whether there was sufficient resilience in the system, what’s it going to look like in Scotland, have we got enough beds, have we got –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Was that sort of analysis being done within the departments before you set up MAH, or were you having to start effectively from not necessarily zero but a very basic level of blank sheets to start?

Dr Audrey MacDougall: So because I wasn’t involved before 4 March, I can’t comment on exactly what was happening, but I don’t think one could say that nothing was happening.

Counsel Inquiry: Could effort – could MAH have been set up earlier?

Dr Audrey MacDougall: It’s a question that as analysts we will always say we want to be in the room from the beginning, but there’s always a trade-off between people trying to just get it – try and formulate the question before we get involved, trying to work out what the scale is of the issue before we get involved. So it’s always a little bit of a trade-off as to at what point, say, should something like MAH be put in place.

We are, for the future, for future crises, following on from debriefing from Covid, we have written some guidelines about what might happen in the future.

Counsel Inquiry: And what do those guidelines say, briefly?

Dr Audrey MacDougall: And I think, yes, I would look for perhaps an earlier activation of this kind of – this kind of division.

Counsel Inquiry: Because to some extent it’s sort of self – it’s a self-fulfilling prophecy or self-evident that the earlier you get involved, to some extent, the more data that you’re able to get your hands on, the more data you’re able to analyse, the more able you are to think about the questions that need to be asked at an early stage, so you don’t have so much of a blank sheet of paper when you start?

Dr Audrey MacDougall: I think that would be true, but I would like to qualify that by saying although my division wasn’t there, there were other people doing work.

Counsel Inquiry: When you – just on that point, then, when you set up the division, were the people that were working in the various departments, were they people that were bringing into the division or did they remain sort of embedded in their respective directorates?

Dr Audrey MacDougall: No, I brought people into the division, they moved away from their own directorates, so they weren’t trying to do two jobs at once.

Counsel Inquiry: So – but you were utilising their expertise and their particular –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – or the knowledge of the work that they’d been doing prior to MAH being –

Dr Audrey MacDougall: Yes, yes.

Counsel Inquiry: Following on, again just in terms of data sharing and how things were set up, if we look at paragraph 19 of your statement – it should come up on the screen.

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Thank you.

You note there that you need – I suppose this is the sort of the other end of the telescope. You’re getting information and analysing it, but it has to go somewhere –

Dr Audrey MacDougall: Yes, that’s correct.

Counsel Inquiry: – and it has to be shared with the right department?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: So you recognised the importance of effectively not working in a silo, and using the other end of the telescope to give that information to the relevant people and the relevant decision-makers?

Dr Audrey MacDougall: That’s right, yes.

Counsel Inquiry: Now, you say in this statement that initially it was for senior analysts in Scottish Government, but by the end of May it was for all interested parties, it was gradually expanded to include latest data, evidence and research alongside modelling. So it grew –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – as it – from no doubt 12 March when you were setting it all up –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – there was less information –

Dr Audrey MacDougall: That’s correct.

Counsel Inquiry: – and you were giving it to fewer people.

Dr Audrey MacDougall: Yes.

Counsel Inquiry: You had a weekly call.

Dr Audrey MacDougall: Yes.

Counsel Inquiry: When did the weekly calls start, the original, the initial weekly calls?

Dr Audrey MacDougall: The initial weekly calls with a small group of analysts will have started a week after I started in the role.

Counsel Inquiry: And by senior analysts, do you mean other modellers, other … what do you mean?

Dr Audrey MacDougall: I’ll explain to you. So each, you know – as you have no doubt had a lot of information about the structure of the Scottish Government, so the Scottish Government is made up of directorates, if you think of a DG family, each DG family would have at least one division in it that was made up of analysts, and that division would be headed up by a senior civil servant who was an analyst, so my equivalent in – and there would be one of those divisions in education, in justice, in health, in … with specialists in those topic areas as well as particularly specialists in particular skills and methodologies and so on. So initially I was dealing with those SCS analysts.

Pre-Covid, and indeed as a routine, we had a leadership group called the analytical leadership group where all the senior analysts meet about once every six weeks to discuss areas of mutual interest and cross-cutting issues across the government. So this was a ramp-up, if you like, of that.

Counsel Inquiry: So, to start off with, it was analyst to analyst, as it were?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: And then those analysts were expected to feed up the –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – analysis that you were providing –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – to the hub to their respective director general areas?

Dr Audrey MacDougall: Indeed.

Counsel Inquiry: So at that point is it fair to say that you didn’t have so much direct contact with the director general or the directorate decision-makers in terms of sharing your modelling?

Dr Audrey MacDougall: So in terms of sharing modelling, at this point I was engaged primarily with the senior policymakers in the health DG, and was – and I have to say occasional, because I can’t necessarily say I was at every meeting – but was invited to gold meetings and various other meetings where ministers were present to present on modelling and – I think Mr Halliday said the same – we took it in turns to present information at those types of meetings.

Counsel Inquiry: You say – staying with the same paragraph of your statement, you say it became a key – the weekly Covid –

Dr Audrey MacDougall: Mm-hmm.

Counsel Inquiry: – became a key communication tool for disseminating the modelling the evidence and the analysis. It was regularly attended by over 100 people, including the Chief Medical Officer, CSA, NCD and senior colleagues.

To be clear, is that the “by the end of May” time, or is that later on? So you say earlier in your statement “By the end of May it was for all interested parties”, but did it continue to expand into people like the CMO attending and –

Dr Audrey MacDougall: Yeah, it was opened up to all interested parties, so quite literally an invite was sent out to anybody in the government who was working on Covid who would find it useful to have a weekly update, and attendance built up as more people got involved, more people got engaged, and as the range of our evidence expanded as well, and that naturally expanded as more evidence became available and more research became available.

Counsel Inquiry: So when – again I don’t need a specific date, but approximately when did the CMO start attending those weekly calls?

Dr Audrey MacDougall: Oh, goodness. I’m not sure. I just can’t hazard a guess on that, honestly.

Counsel Inquiry: And as they expanded and you had more people joining and an open invitation, presumably the – with other analysts, they understand the principles about modelling, they understand more readily how the evidence is being assessed, did it – did your calls have to engage at a level of trying to explain what it all meant and explain the principles –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – behind modelling as it went through?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: So how were you able to ensure that those decision-makers and those senior members who were attending were able to, you know, understand what is modelling and the like?

Dr Audrey MacDougall: I think we tried to present the information in a way that was readily comprehendible, but we also – the call itself wasn’t just my team presenting information, there was also a period of time set aside for questions, and it was made very, very clear that any question could be asked, so it could be what people might think of as a very basic question, that was absolutely fine, or it could be a more complex question, and we had a mix of both, and we used that as feedback. So if we were getting feedback that seemed to indicate that people didn’t understand one element of our presentation, then we would change that for the following week.

Counsel Inquiry: I’m relieved to hear that no question is too stupid from me at least.

Can we think about – and appreciating you’re not a modeller, but you’ve already touched on slightly in terms of how modelling, what modelling is and how it works. It’s a term that is used sort of interchangeably, isn’t it, to not necessarily reflect those three areas that you were talking about, but if we take them all in stages in slightly more detail, but very slightly more detail, than you spoke to a moment ago: epidemiological modelling, that was done in both in-house in Scottish Government, wasn’t it, and also by applying – by using other external modelling groups, so SPI-M, as you’ve already spoken to, that fed into SAGE?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: You were working with other universities?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: And Scottish Government had their own –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – modelling approach in which, is it right that they were using the Imperial College of London’s modelling?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Like base –

Dr Audrey MacDougall: Yeah.

Counsel Inquiry: – sort of structure, as it were, and then applying Scottish specific data to it?

Dr Audrey MacDougall: That’s correct. So from a very early stage Imperial College published their modelling code and so it was available to use. We obviously spoke to Imperial College and we made adjustments for Scotland so we could make adjustments for the age profile of the population, for example. We then ran, if you like, our own model. We wished to build our capacity in-house and run our model in-house, so we ran that. We also, via our CMO, we asked our CMO to request that SPI-M started to model Scotland separately, because to begin with there were only UK models and we felt that the UK models weren’t an adequate representation of what we were starting to see happening in Scotland. At that point in time SPI-M-O were starting to run regional models for England so they then agreed that, yes, separate modelling for Scotland would be appropriate.

Counsel Inquiry: Just pausing there for a second, when was “at that time”?

Dr Audrey MacDougall: Oh, it was in March, I mean, it was very early on, so we’re a couple of weeks into it when we decide that we really do need to have separate Scottish modelling.

We – a combination then of our own conversations by our modelling team spoke to some of the modelling groups in the universities to say: would you like to model Scotland? And a number of the groups agreed that, yes, they would model Scotland. So that enabled us to have a number of groups modelling Scotland specifically, including ourselves, and then we could gradually bring those models together to form SPI-M or consensus.

You’ve probably heard about the consensus approach being really important, because any one model by itself could be misleading or could be less accurate, if you like, by bringing a number of models together and creating a consensus you get a quality assurance of what’s being done.

Counsel Inquiry: Because someone could just simply put the wrong assumption in and it would result in an aberrant outcome?

Dr Audrey MacDougall: Absolutely, yes.

Counsel Inquiry: So there’s a bit of cross-marking and peer review, effectively?

Dr Audrey MacDougall: So it’s a peer – indeed, that’s exactly what SPI-M-O would be, a peer review process.

Counsel Inquiry: And that sort of epidemiological modelling allows for sort of short term analysis, “Look, this is what we think is going to happen, within a certainty level of degrees, over the next couple of weeks, the next month or so”, and then longer term modelling with –

Dr Audrey MacDougall: It’s uncertain.

Counsel Inquiry: It becomes more and more uncertain?

Dr Audrey MacDougall: It becomes more and more uncertain, particularly it depends on what assumptions you might want to make about whether you intervene to change the existing situation. Obviously if you assume that what’s happening today will continue to happen and I am making no interventions, you’ll project forward and you’ll get an idea of what might happen, and I think some of the early Imperial College models, for example, that’s what will have happened, because if no action is taken, here’s what things might look like. Whereas more useful modelling is to try and say: well, okay, what if we did this type of intervention or that type of intervention, can we do some sort of scenarios or estimates as to what difference that might make?

Counsel Inquiry: Just to go through a sort of very basic modelling approach, could we have, please, on screen INQ00029254. I hope this doesn’t hurt everyone else’s eyes the way it hurts mine, but this is a document produced on 24 March, so the day after lockdown, and it’s a – is it an early attempt to model what is happening and what is likely to happen in the reasonable worst-case scenario in Scotland when assessed in terms of infections, deaths, need for hospital beds and the like?

Dr Audrey MacDougall: Yes. Yes, put simply, yes, and it gives you two scenarios, one with no social interventions, one with – the “do nothing” scenario, if you like – one with social interventions, and the social interventions are listed out here, the kind of things that you might do. Or if not listed here, they will be listed in –

Counsel Inquiry: They’re at line 21, the social interventions applied –

Dr Audrey MacDougall: Oh, that’s right, sorry, apologies, there you are. So yeah, that’s an initial attempt to –

Counsel Inquiry: Just as a sort of worked example, as it were, am I right in reading this that the way this – that no interventions, when we look at reasonable worst-case scenario, the – it’s anticipated that there will be a peak of infections in Scotland per week of just over 1.1 million?

Dr Audrey MacDougall: So you would have hit a peak of 1.1 million, so that was based on a range of assumptions about the percentage of the population that was likely to be infected, how quickly the pandemic would spread and how transmissible the pandemic was, and then there was a hospitalisation data, what percentage of those who got sick would need hospitalisation.

Obviously because this was very early days, we didn’t know – obviously we didn’t know about vaccines, we didn’t know we were going to have vaccines, so this was a kind of very, just native, if we did nothing and, you know, the epidemic just spread –

Counsel Inquiry: And the version within – with those social interventions effectively was based on, was predicated on all of those interventions just being lifted after three months, or whatever the figure was –

Dr Audrey MacDougall: Yeah.

Counsel Inquiry: – and then in that position the modelling was still saying – was saying effectively the same thing is going to happen it’s just going to be –

Dr Audrey MacDougall: Shifted.

Counsel Inquiry: – 15 weeks later, 14 weeks later in the piece?

Dr Audrey MacDougall: That’s correct, because what you would do is dampen down but then you would shift.

Counsel Inquiry: You’d see the ping back or the bounce back that we hear about.

Dr Audrey MacDougall: Yes.

Counsel Inquiry: This document was then, I think, presented effectively to Scottish Government by way of a sort of slide pack, I think by your colleague Mr Halliday.

Dr Audrey MacDougall: Yes.

Counsel Inquiry: If we can just go to that very briefly, INQ000292555. I think this is an attempt to make the information slightly more accessible to those who …

(Pause)

Dr Audrey MacDougall: Yes.

Counsel Inquiry: So we can see at this stage, so very early on, but there have been some updates from a previous – this is I think version 1.8, but it’s still early doors.

Dr Audrey MacDougall: It’s very early days, yes.

Counsel Inquiry: On page 4, we look at the key assumptions that have been applied in that modelling.

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Who decides on how – these assumptions and how they’re applied?

Dr Audrey MacDougall: So the assumptions that we were using at that point were assumptions that would have come through SPI-M or SAGE. We weren’t certainly creating our own assumptions.

Counsel Inquiry: And it may be in those circumstances you can’t assist with this question, but it says there at the bottom:

“Assumes care home residents are not moved.”

Dr Audrey MacDougall: Yes. I think the position, and I’m fully appreciating the issue around care home residents, I think in this situation it was rather the other way round: are care home residents going to be moved into hospital, rather than are people being moved from hospital into care homes.

Counsel Inquiry: And then if we just turn to the next page, page 5, we see the caveats that are being highlighted as not being factored into things. Were these – again, are these from SPI-M-O or SPI-M, or are they …?

Dr Audrey MacDougall: So –

Counsel Inquiry: Are they you and your team saying, “Well, hang on, these are things that we might need to factor in”?

Dr Audrey MacDougall: It was – this would have been a little bit of both, because some of these were things that, you know, wouldn’t have been factored in by any modelling group doing modelling, whereas for example number 10 there would have been quite specific to us. And number 7, we haven’t adjusted for Scottish geography, so they would have been quite specific to us.

Counsel Inquiry: If we move to a slightly separate section of your evidence, please, in terms of the route out, the route map.

Dr Audrey MacDougall: Yes.

Counsel Inquiry: So we know that the Scottish Government published their route map, their framework for decision-making, in April 2020, and is it right that your team were then asked to provide data and evidence in respect of the issues that they were looking – the metrics?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Do we see the first – if we bring up INQ000131026, this further – you look puzzled.

Dr Audrey MacDougall: I tell you why I look puzzled. It is true that we were asked to produce information and, if you like, a measurement framework in and around the four harms, but it’s not this document. There is another document that was published around about the same time that did that.

Counsel Inquiry: Can I – I’ll take you through some of the evidence in this –

Dr Audrey MacDougall: Apologies.

Counsel Inquiry: It deals with metric, I apologise if that’s not the direct document, but I think it deals with metrics that we can look at.

Dr Audrey MacDougall: Certainly.

Counsel Inquiry: So if we look at page 9 of this document, we see – so this is evidence that is being provided in terms of health impacts and –

Dr Audrey MacDougall: That’s correct, yes.

Counsel Inquiry: So spread of the virus, our lockdown rules are working, we get various statistics and then if we look at just at the bottom of that page, if we scroll down, the charts show a range of key measures in the pandemic in Scotland, new cases, hospitalisations, numbers. Are those the sorts of data that you were providing from the MAH, even though it might be in a different document and a different approach?

Dr Audrey MacDougall: So key measures of the pandemic came from different places. I think you’ve already heard from Mr Heald from PHS, who were actually responsible for a lot of the daily data in terms of cases and hospitalisations, ICU and so on. So actual data was being fed through PHS to ourselves.

Counsel Inquiry: And then in an analysis of, in support of or in measuring –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – the way the interventions were working –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – or whether lockdown needed to continue and the like, that was being fed through your team in the MAH with those sorts of information and being able to model through?

Dr Audrey MacDougall: So we could use PHS data to model through what might happen. So particularly case data, for example, was a typical input to a model, and we would use PHS case data for that.

Counsel Inquiry: Again, given this isn’t quite the document, you might not be able to assist, but if we look at page 12, so we see here, this is a document from May 2020, just for clarity, the –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – sort of headline figure, headline title, “Limited headroom to change restrictions”, and it says:

“While precision on the R number is difficult, it’s likely to be 0.7 to 1.”

R number something, of course, that the MAH was particularly –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – keen and interested in in terms of analysing the ebbs and flows of the pandemic, and the –

Dr Audrey MacDougall: That’s correct.

Counsel Inquiry: – applications.

It says, the second half of that:

“This is an average for all of Scotland. The R number for community transmission in Scotland is estimated to be below R number in care homes and hospitals. This is a matter of critical concern.”

Dr Audrey MacDougall: Yes.

Counsel Inquiry: So in other words, general community perhaps closer to 0.7, for example, but because it’s an average –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – knowing that there were higher – there was higher R number –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – in care and hospital settings was dragging the R number up in the average R number up –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – is that right?

Dr Audrey MacDougall: I’d have to think about the actual proportions to say what exact difference it might make. But, I mean, obviously what you’re seeing here is you’ve got an R number but then you’re seeing actual data in terms of what’s happening in hospitals and happening in care homes, that’s not aligning with the community R number, so you can see that the R number may be higher.

Counsel Inquiry: And I think in due course you – again, it might just be easier to bring it up on screen. INQ000249321, page 7. This is a document, sort of sets out how the pandemic is being modelled in Scotland, broadly, and it says there:

“Estimating R in different settings.

“There are at least three different epidemics in Scotland …”

Dr Audrey MacDougall: Yes.

Counsel Inquiry: So effectively quite early on you were –

Dr Audrey MacDougall: We were aware.

Counsel Inquiry: You were aware through different –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Community – am I right, is it community, hospitals and care?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Is that the three?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: The communities that you’re looking – the three metrics.

You’re looking at the whole population model. It notes other types of models are needed to analyse those three segments of society, as it were, and you’re working with academic groups from around the UK to develop modelling for those settings.

Did they ever come about, those models?

Dr Audrey MacDougall: Care home modelling came about, there was care home modelling done through SPI-M, we didn’t do it, it was done through other people, through SPI-M, and –

Counsel Inquiry: Was that SPI-M modelling Scottish specific?

Dr Audrey MacDougall: No, not Scottish specific.

(Pause)

Dr Audrey MacDougall: I think we did – there will have been some analysis done of care homes, because modelling is a very particular tool, when actual data starts to become available one can also be looking at actual data and, you know, getting some better estimation of what’s happening rather than trying to model forward.

Counsel Inquiry: We heard evidence earlier today about the sort of concerted effort to publish lots of data, to be open and –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – produce the data and the modelling.

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Was that something you in the MAH were particularly concerned about?

Dr Audrey MacDougall: Yes. We were very keen – we realised our material wasn’t for everybody, that some people may not wish to engage with us, but we were very keen to try and get material out when we felt appropriate. So from a very early stage, our Modelling the Epidemic report was produced, and we produced that on a weekly basis for most of the pandemic, moving to fortnightly nearer the end. And that was put out there to give a degree of transparency about what we were doing, to recognise the caveats, recognise what it could and couldn’t do, and to invite comment in the sense of we were very happy if people wanted to get in touch with us and make suggestions for improvement or suchlike, so, you know, that was why we put that out there. And indeed I think it served its purpose in that sense. We did have a lot of people get in touch, ask questions, get in touch, offer advice, offer help.

So, yeah, I was – I felt that that achieved what I wanted it to achieve. It wasn’t a document that I thought everybody, you know, in the general public would be reading, but for what I wanted and for the audience, I felt it was – it worked.

Counsel Inquiry: Is there a risk in publishing so much data that it becomes overwhelming and sort of it’s the only show in town, as it were, because it’s the most – it’s the thing that’s shouting the loudest?

Dr Audrey MacDougall: Yeah, it is interesting, that, because, I mean, we did publish a lot of other material. There wasn’t just modelling. We published a State of the Epidemic report, which we didn’t start until slightly later on in the pandemic, but then we did a weekly round-up, if you like, for the public, we had our four harms website, and we published quite a range of research reports.

Now, it’s a really good question whether everybody just really then focused in on this modelling and didn’t perhaps pay due attention to some of the other forms of evidence and analysis that were published and available, and that’s always a risk, but preferable I think to put as much out as possible.

Counsel Inquiry: Because there’s always a risk that something like modelling and its extremely difficult concepts run the risk of looking like it’s a crystal ball, like you’re able to predict –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – the future, and therefore some may say that, “Well, the modelling says this”, and become fixated simply on the modelling rather than taking it in the round with other, less attractive crystal ball approaches?

Dr Audrey MacDougall: And that was part of the reason for the four harms approach, to try and provide a framework within which you could wrap up, if you like, or encapsulate quite a range of evidence and analysis looking at the issue from different angles. So it wasn’t just looking at modelling, harm 1, that’s all. It was trying to take into account a much wider range of approaches, evidence and perspectives.

Counsel Inquiry: And at the other end of the telescope in terms of public interaction, you were carrying out polls and –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – studies in terms of –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – contacting people?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: (a) to – and that feeds into the modelling in itself, doesn’t it?

Dr Audrey MacDougall: So the Scottish Contact Survey certainly fed modelling, one could model based on contact survey. What the Scottish Contact Survey did was survey people to find out how many people they’d been in touch with during the week. So it would be quite literally, “Well, you know, I met one person in a shop, I met one person in …” and by that you build up a picture of the population, of the amount of contact people are having with each other, that has a impact on transmission – the more people you have contact, the more, you know, high risk for transmission – and that can feed into a model.

So that was one sort of survey, if you like, interaction with the public, but we were also doing work such as polling to try and find out a little bit more about the public’s attitudes, public’s level of compliance, public’s sense of wellbeing, a range of different issues.

Counsel Inquiry: Because things like compliance will also feed into whether the NPIs are likely to continue working or –

Dr Audrey MacDougall: Indeed.

Counsel Inquiry: – the effectiveness changing over time. If people are less likely to comply with lockdown rules as time goes by, that feeds into the modelling –

Dr Audrey MacDougall: That’s correct, and that would feed into – that would be one of the considerations within our four harms approach was that consideration of compliance. Actually “compliance” is really not a very nice word, but adherence. But, you know, the general idea would be one wouldn’t put something in place that people would find it impossible to adhere to.

Counsel Inquiry: And presumably some of that polling was particularly useful in the context of harm 3?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: And was that a major source of information and data for harm 3? Because it’s quite a difficult thing otherwise, I would imagine, to –

Dr Audrey MacDougall: Yeah, yeah, it was a major source because it was a weekly source so, you know, it was one thing that we could use to get regular information. Other information that fed harm 3 came to a variety of different research projects or reports, but certainly wouldn’t have been available or updated on anything like a weekly basis.

Counsel Inquiry: In terms of the polling and the studies that were being undertaken, are you aware of to what extent were Scottish specific polls and studies carried out in respect of minority groups, in respect of the particular effect that – for instance, that we’ve heard about this morning – NPIs having on ability of people in the disabled communities to access care or food or …?

Dr Audrey MacDougall: So, I mean, through our own polling we could only split down so far in terms of different groups within the population because the poll simply wasn’t large enough to get every group that one would – underrepresented group that one would like, which was why then in – individual research projects were launched which involved focus groups, interviews, different types of interactions with people from different groupings, and that would have included ethnic minority groups.

We also drew on the work of representative bodies themselves, so where representative bodies themselves had undertaken their own research and, you know, polling or surveying or whatever, we always invited them to, you know, share that with us and that became part of our evidence base.

Counsel Inquiry: Very, very briefly, a whistle stop tour of the four harms strategy –

Dr Audrey MacDougall: Okay.

Counsel Inquiry: – which I’m going to apologise in advance for. INQ000131028.

Dr Audrey MacDougall: Ah, yes.

Counsel Inquiry: This is hopefully more familiar in terms of that’s what you were expecting to be drawn up. This is the 11th – it’s dated 11 December, I think, setting out the detail of the four harms.

I appreciate lots of evidence will be heard about the four harms and the way it was put, how it was dealt with across government, but just very briefly in terms of your role –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – you – is it true that – is it true to say that your – the harm that you probably had in your researcher – wearing your researcher hat, that was harm 3 in particular, as we’ve just said?

Dr Audrey MacDougall: Yes.

Counsel Inquiry: If we just look at page 22 of this document, that sets out sort of broad analysis, broad description of those harms.

Dr Audrey MacDougall: Yes.

Counsel Inquiry: Wearing your respective hats, to what extent were you involved in trying to balance different harms, or were you simply being required to provide evidence in each, under each heading of each harm, as it were?

Dr Audrey MacDougall: So, I’ll try and explain my role here. My division were obviously responsible for a lot of the information that fed harm 1 through our modelling work and through our scientific review work that we undertook, and we provided a lot of input into that, along with our scientific and clinical and medical colleagues. That was harm 1.

Harm 2, the main work was undertaken by my colleagues in health analysis, they undertook that work.

Harm 3 was myself, our Chief Social Policy Adviser, who was ultimately responsible, and our colleagues in areas such as justice, education and so on. Including some of our own work, though, because you’ll notice topics here like loneliness, anxiety, trust in government, social capital. These were all subject to – these were all part of our polling and our own research, so that was a mixture. As you can imagine, because harm 3 was that very broad, so it took a lot of inputs.

Then harm 4, which was the economic analysis, was carried out by my economics colleagues and led by our Chief Economic Adviser.

Counsel Inquiry: In terms of your role in MAH, you weren’t being asked – or were you – to try and undertake that balancing between the various harms as opposed to “Please provide the evidence of harm 3, harm 4”?

Dr Audrey MacDougall: Well, not as an individual but I was a – one of the core attendees at the four harms meeting. So that extent, you know, I would have a perspective and I would give my views. But I was also responsible on a regular basis, if you like, for co-ordinating and bringing together all the material that was produced on all the harms and writing it up in a way that was then presented to ministers.

Counsel Inquiry: So providing advice in respect of that balancing exercise and –

Dr Audrey MacDougall: Yes.

Counsel Inquiry: – where there were more red lines, as we were looking at earlier?

Dr Audrey MacDougall: One of many colleagues who were involved. This was a very collaborative effort.

Ms Arlidge: My Lady, may I just bend down to check?

(Pause)

Ms Arlidge: I have nothing further for you. Was there anything you would like to add, Dr MacDougall?

The Witness: I – two things I just want to say, really.

First of all, I really do want to get across the idea, if we can, that this was an incredibly collaborative effort, and although by the very nature of the Inquiry you’re interviewing lots of other individuals, all our activity was terribly collaborative. So you obviously just get partial views from everybody you speak to.

But the other thing that I would like to do is to pay tribute to the analytical staff within the government who were responsible for the production of some really sophisticated and new and, you know, really high quality analysis and the amount of work and effort and the hours that people put in was phenomenal. So I would really like to pay tribute to my colleagues.

Ms Arlidge: My Lady, do you have any questions?

Lady Hallett: No, I have no questions, thank you very much.

Thank you very much for your help, Dr MacDougall.

The Witness: Thank you.

Lady Hallett: Very helpful.

(The witness withdrew)

Lady Hallett: Right, 10 o’clock?

Ms Arlidge: My Lady, just before you rise –

Lady Hallett: Oh, yes, you wanted to ask about publication.

Ms Arlidge: – I’m going to need to ask about some – make an application for publication in terms of all of the witnesses’ statements that have been heard.

Lady Hallett: Certainly, all the documents that you wish to have published from today shall be published.

Ms Arlidge: I’m very grateful. Thank you very much, my Lady.

Lady Hallett: Thank you. 10 o’clock tomorrow, please.

Ms Arlidge: Thank you.

(4.30 pm)

(The hearing adjourned until 10 am on Thursday, 18 January 2024)