Submission - Response to COVID-19 - Chief Statistician (Director of Statistics and Analytics) - 23 F

19-21 Broad Street | St Helier Jersey | JE2 3RR

Constable Karen Shenton-Stone Vice-Chair, Public Accounts Committee BY EMAIL

22 February 2022

Dear Constable Shenton-Stone ,

Public Accounts Committee (PAC) - COVID-19 Response Review Introduction

You wrote to me on 16 February on behalf of the Public Accounts Committee regarding its ongoing review of the Government of Jersey's response to COVID-19. You requested a response by 22 February. I have provided below responses as best I can in the time available.

As I believe that you are aware, I became the Interim Chief Statistician and Director of Statistics and Analytics on 4 January 2021 and secured the post on a permanent basis at the end of 2021. I therefore have no personal experience of the Jersey Covid response during 2020 but have spoken to those involved to gain their insights and have included this information in my responses below.

Most of your questions relate to Statistics Jersey, but I have expanded my response to cover the work of the HCS Informatics and Public Health Intelligence teams who also had a crucial role to play in the production and publication of Covid statistics. I would want to give these teams credit for their hard work on behalf of Islanders, and to also provide you will a fuller picture.

My role as Chief Statistician and Director of Statistics and Analytics includes leadership of Statistics Jersey but also a professional responsibility for statisticians and analysts across government. This includes career and professional development, but also providing advice and support to the Heads of Practice for statistics in departments on professional statistical matters, in line with the Code of Practice for Statistics.

And finally, I should mention that I have sat on STAC as a member since joining Government of Jersey in January 2021 – with a particular role in relation to the statistics and data supporting STAC and some considerations about publication of statistics.

I have repeated your questions below and provided the replies that I am able to in the time available.

Response to COVID-19 Questions and Responses

1. Please could you outline the roles and responsibilities undertaken by Statistics Jersey during and specific to the COVID-19 Pandemic? a. How were decisions made within Statistics Jersey in relation to prioritising work which arose as a result of the pandemic?

Roles and responsibilities of Statistics Jersey during the pandemic:

• At the start of the pandemic Statistics Jersey, as with other stats offices around the world, prioritised work on statistics and modelling of Covid (case numbers etc)

• A member of Statistics Jersey with Public Health experience carried out modelling of possible case numbers in Jersey, based on models from the UK. This included briefing STAC and briefing Ministers

• Statistics Jersey had to stop some of its existing survey activity because of the pandemic

o The Living Costs and Household Income Survey (LCHIS) had started in summer 2019 and was due to run for a year. This involved an hour-long interview in respondent's house. This survey was stopped only partially completed on the introduction of Covid restrictions in March 2020

o In January 2020 Statistics Jersey started to carry out the Exit Survey (of visitors spend and experience) for Visit Jersey. Due to Covid restrictions, and with the agreement of Visit Jersey, this survey was halted in March 2020

o The new Annual Business Survey was stopped. This was going to be the first compulsory business survey which would have included an expanded set of questions specifically to try and start collecting Trade Data. It was judged inappropriate to start this survey at a time of great stress to businesses, and when the results were unlikely to be representative.

• In early-2020 Statistics Jersey carried out an antibody survey which had three rounds to measure antibody rates in the community. Results were published in May, June, and July 2020. Statistics Jersey were able to deploy field-staff who were no longer able to work on LCHIS or the Exit Survey.

• At the start of the pandemic Statistics Jersey produced a weekly 'Where are we on the curve' publication, which was subsequently handed over to Public Health Intelligence in July 2020

• Statistics Jersey also rapidly started an economic indicator report from 28 April 2020, focussing primarily on administrative data including from CLS, Treasury, the Royal Court and (later on) Transport. This report was published weekly initially, then became monthly and Statistics Jersey continue to publish the information quarterly as open data.

• The Public Health Intelligence team was built up rapidly in response to the pandemic, and Statistics Jersey became less involved in reporting of Covid health data. The Public Health Intelligence team took the lead on reporting of Covid statistics to STAC.

• The HCS Informatics team lead on collation of data on, for instance, testing, case numbers and hospitalisations and produced the daily statistics that have been made public to politicians, the media and via Open Data. This responsibility moved to Public Health Intelligence in January 2022.

Decisions on prioritising the work of Statistics Jersey were made in the normal way by the then Chief Statistician, in consultation with Medical, Health and Public Health colleagues. Activities that had to stop because of the public health restrictions (such as the LCHIS, Exit Survey, and Annual Business Survey) freed up resource that could be deployed on new priorities.

2. Please could you outline which organisations and departments you received information from for the production of statistics in relation to the pandemic?

Data was obtained from:

• CLS, T&E, The Royal Court and transport on the economic impacts of the pandemic

• HCS on Covid Cases, hospitalisations, vaccinations, etc

• JHA Test and Trace for testing at the border (until moving in to SPPP in late 2021)

• IHE Contact tracing (which moved to JHA and then to SPPP in late 2021)

• CYPES for information on number of cases amongst students and teachers

• Superintendent Registrar for mortality data.

It is worth noting that the Covid Informatics team also had an operational role – for instance they produced swabbing projections based on intended policy changes to help with estimates of costs. They also used data from Ports of Jersey in relation to predicted arrivals to help with planning of Covid travel testing.

3. What agreements did you put in place to ensure that data was provided to Statistics Jersey within a sufficient timeframe?

Statistics Jersey already had Data Sharing Agreements in place for health data for instance. Where the Health Informatics or Public Health Intelligence teams needed new Data Sharing Agreements (e.g. for new sources of data such as vaccination data) these were rapidly agreed. Agreements were put in place with test providers such as Opencell and Micropathology for operational purposes, with a secondary use for production of statistics.

4. What were your lines of accountability for Government work during this period and did they change from usual practice due to the pandemic?

• As in other jurisdictions, there was lots of joint working to provide the information needed to support public health decisions.

• Stats Jersey is independent; it's lines of accountability didn't change.

• Public Health Intelligence didn't exist at the start of the pandemic; it has been rapidly built, in response to the pandemic.

• HCS Informatics moved back to HCS as part of the Analytics Transformation Programme rather than because of the pandemic.

• Public Health advise Ministers after seeking advice from STAC.

• At the start of the pandemic Statistics Jersey had a senior statistician attending STAC on an advisory basis. From the start of 2021 the Chief Statistician has been a member of STAC.

5. How were you assigned work by Ministers and other Officers and, as above, was this a change from usual practice?

• In line with normal practice, Statistics Jersey were not assigned work by Ministers or other officers. As in other jurisdictions, the Chief Statistician took decisions on the relative priority of requests for help, based on the joint effort to protect lives and livelihoods.

• Public Health Intelligence advised Public Health colleagues; STAC and Ministers (e.g. with data briefings), and Ministers would sometimes request analysis to support decision making in line with usual practice.

6. Please describe how you supported the work of the Scientific and Technical Advisory Cell (STAC), Ministers, and other key stakeholders in providing advice and making decisions to respond to the COVID-19?

• Public Health intelligence provide a data brief at the start of each STAC meeting – with data such as cases, hospitalisations, vaccinations etc.

• Statistics Jersey continue to publish briefing on the economic impacts of the pandemic.

• The Chief Statistician has supported Heads of Practice in Public Health and HCS in responding to requests for briefing on issues such as the vaccine status of Covid cases, and the vaccine status of hospitalisations – to ensure both that the statistics are reliable and that they comply with professional standards on issues such as disclosure.

7. During the course of the pandemic which States of Jersey departments (both executive and non-executive) were you required to provide advice to?

• Statistics Jersey were not required to provide advice but, in line with normal practice, provided professional support to other parts of government when asked and after taking account of priorities. This is described in response to earlier answers.

• The Chief Statistician responded to queries on Covid data received from the Statistics Users Group – including working with colleagues in government to develop the existing reports (e.g. on vaccination statistics) in response to such feedback.

8. Did the presentation of this advice during a pandemic present any new challenges to the objectivity of Statistics Jersey?

• No.

9. How did you work to arrange and facilitate briefings with Ministers, States Members and stakeholders when requested?

• Briefings were provided to Ministers and the media on the R Number modelling undertaken by Statistics Jersey early in the pandemic

• Regular data briefings were provided to Ministers by the Public Health Intelligence team

• HCS Informatics and Public Health Intelligence ensured that a wide range of Covid statistics were published on the Government of Jersey website.

2. How did you determine what statistics to collect and publish in relation to COVID-19? a. What consultation did you seek from Ministers, the Scientific and Technical Advisory Cell, the Competent Authorities, Health and Community Services and other key stakeholders regarding which statistics to publish and disseminate?

• Health Informatics and Public Health Intelligence used professional judgement to decide on which statistics to collect and publish – based on their own expertise and international advice and experience.

• Part of the judgement on publication related to the availability and quality of data. Systems to collect appropriate data had to be developed, working in collaboration with policy colleagues, M&D Technical colleagues, and operational colleagues in Test & Trace and HCS.

• Public Health Intelligence responded to requests from STAC and from other bodies such as the Statistics Users Group.

2. How was a publication schedule for relevant statistics agreed with other relevant parties, including the Government of Jersey's Communications department?

• Publication of Statistic Jersey official statistics continued as normal in line with the Statistics Jersey release schedule, apart from those outputs which had to be stopped because of Covid public health restrictions. As mentioned above, Statistics Jersey introduced a new output on the economic impact of the pandemic.

• HCS Informatics and Public Health Intelligence developed a range of statistical outputs (e.g. on cases, hospitalisations, deaths, etc) in response to the pandemic. The range of statistics evolved as the pandemic developed (e.g. statistics on vaccinations once the vaccination programme started).

• The frequency of reporting depended on the quality of the data. Some statistics were released daily while others were weekly, depending on the robustness of data and the amount of quality assurance work that needed to be carried out.

• Some usual public health analyses (e.g. smoking and drinking) had to be delayed whilst staff resources were prioritised to pandemic related analyses.

3. How did you seek and obtain feedback from key stakeholders (including GPs and hospital staff) and the general public to understand how and whether the quality and reporting of statistics could be improved?

The HCS Informatics team worked very closely with hospital staff and GPs throughout the pandemic to ensure that operational requirements were met. In wave 1 they introduced 7-days per week reporting to meet these requirements and they also introduced automated reporting.

The HCS Informatics team spent a lot of time with, for instance, GPs to ensure that their reporting systems met a range of user needs. Both HCS Informatics, and Public Health Intelligence heads also had very regular meetings with the Deputy Medical Officer of Health.

• Some of the evolution of published statistics naturally followed the progression of the pandemic (e.g. introduction of vaccination statistics).

• Discussion in STAC identified areas of interest.

• Some statistics were developed over time as the collection and quality assurance process bedded down and they became more routine.

• The Statistics User Group provided feedback on Covid related statistics, which were actioned where it was possible to do so.

4. Did your approach to publication change over this period and, if yes, what drove that change?

• The standard approach to publication was to put information on the Government of Jersey website, with key messages also communicated via social media.

• Presentation of the website data was updated periodically as additional information became available, and in the light of feedback.

3. How did you ensure that the statistics produced were accurate, easy-to-understand, and accessible during the pandemic? a. Please can you outline any ways in which your work with the Government of Jersey departments and agencies changed to adapt to the pressures caused to improve the quality and rate of production of statistics at that time?

• Presentation of statistics was in line with professional standards on reporting of statistics and was informed by the professional judgement of the statisticians and analysts publishing the data.

• Covid Informatics were a core part of the operational Covid team – represented on the (twice) daily meetings, and the Test & Trace Board. This ensured the team were kept abreast of policy changes, were fully engaged in technical system changes to ensure data capture and availability, working very closely with M&D colleagues. The team were also able to respond immediately to concerns and questions raised by operational colleagues. If there were issues with systems, the Informatics team's knowledge meant they were often best placed to help understand which part of the process had issues – e.g. if a person did not receive their test result.

• Additional resources were provided to HCS Informatics and Public Health Intelligence teams to enable them to respond to the increased demand for Covid statistics.

• As is normal professional statistical practice, over time it was possible to automate the production of Covid statistics to make their production more routine.

• It was also possible to review and update the presentation of statistics once they were established.

2. Did you seek to gain an understanding of whether the statistics produced and published were appropriate and useful for both decision-making and for informing the public about aspects the COVID-19 pandemic?

• There were regular meetings with the Deputy Medical Officer of Health, especially at the start of the pandemic, on the statistics required.

• As explained above, discussions in STAC informed the statistics produced.

• Media commentary on topics of interest to the public (e.g. the vaccine status of Covid cases) also informed decisions on analyses undertaken.

• The Statistics Users Group provided feedback on Covid statistics.

• The approaches to publication of Covid statistics being taken by other jurisdictions (including Public Health England) were also taken into account.

3. How did you seek to understand the role and use of statistics in other jurisdictions and areas of similar size and standing to Jersey, including Guernsey?

• Public Health England (PHE) had a special advisor for Islands, who provided advice.

• Public Health Intelligence leads in the Isle of Man, Guernsey and Jersey hold a monthly catch up and talk frequently about COVID statistics. This has been helpful in understanding and sharing challenges, and learning lessons from each other, although the different Islands' pandemics have followed different timescales.

• Public Health experts in Jersey reviewed the presentation of data in other jurisdictions, including by PHE.

• There is no single comparator for Jersey; for some issues (e.g. Covid cases) comparison to the UK was most relevant, as that was where most cases were seeded. But for other issues (e.g. travel restrictions) comparisons to Guernsey were relevant.

• The weekly epidemiological report tracks testing rate and test positivity against jurisdictions for which Jersey has significant travel links – such as Ireland, France etc.

4. Could you please provide an overview of the support that Statistics Jersey provided to the Scientific and Technical Advisory Cell, particularly regarding the provision of statistics and other quantitative social and operational research, as outlined in the Cell's Terms of Reference and Code of Practice? a. What was your role as a member of the Cell?

• At the start of the pandemic Statistics Jersey provided briefing on the R number and modelling of Covid cases etc. As explained above, this role was subsequently taken on by the Public Health Intelligence team who provided a data brief at the start of each STAC meeting.

• Statistics Jersey ran three rounds of an antibody survey in early 2020, which provided a measure of prevalence in the community at that point in time.

• The Jersey Opinions and Lifestyle Survey 2020 (run by Statistics Jersey) was reduced in scope/length (reflecting a need to not overburden respondents at a time of significant personal stress) and was given a Covid-specific focus. This survey contained a range of questions related to the impact of the pandemic and provided some useful data which was presented by Public Health Intelligence to STAC.

• The Chief Statistician was a member of STAC from January 2021.

5. How did you ensure that Statistics Jersey was adequately resourced during the pandemic? a. Were additional resources or staff moved to Statistics Jersey? If so, how was this identified and how were they requested?

• No additional resources were provided to Statistics Jersey. As explained above, some Statistics Jersey work had to stop because of Covid public health restrictions. These resources were reprioritised to other work. During 2021 Statistics Jersey's priority was the regular economic statistics outputs, the 2021 Census, restarting the Living Costs and Household Income Survey, and running the bi-annual Children and Young Persons Survey.

• The overall Test and Trace Business Case funded by Covid reserves included additional resources for HCS Informatics and Public Health Intelligence teams to support production of Covid statistics and analyses.

2. How did you track the movement of resources, including funding and officers?

• The Covid reserves deployed through the Test and Trace Business Case were monitored within the overall Test and Trace Programme.

• The real challenge was not funding, but being able to recruit and deploy experienced statisticians and analysts. As in other jurisdictions, the pandemic increased the demand for statistics and analyses on a range of health, economic, and social factors.

3. How did you monitor budgets and total spend during COVID-19? Was it tracked separately to the rest of the spending within Statistics Jersey?

• Statistics Jersey spend was monitored in the normal way.

• The Covid reserves deployed through the Test and Trace Business Case were monitored within the overall Test and Trace Programme.

4. What impact did the pandemic workload have on the ability of staff to undertake other ongoing projects?

• Mainstream Health and Public Health Informatics outputs (such as analyses of smoking and drinking) had to be deprioritised.

• As explained above, a number of Statistics Jersey surveys had to be stopped because of Covid public health restrictions.

6. How did you monitor, track and record the performance of Statistics Jersey and the quality of its output at this time?

• Statistics Jersey met all of publication release dates, carrying out the same quality assurance checks as usual. As explained above, Covid public health restrictions meant that some Statistics Jersey work had to be stopped.

7. Have you conducted an internal review of Statistics Jersey's work and performance during the pandemic and drawn any lessons from that review? If yes, please outline how any resulting changes will be introduced?

• No – but the Chief Statistician has reviewed the workload of Statistics Jersey in the context of having to restart the Covid-delayed Living Costs and Household Income Survey in autumn 2021; alongside the analysis and processing work on the 2021 Census; the autumn 2021 Children and Young Persons Survey; and the broader Statistics Jersey workload. The pandemic has meant that Statistics Jersey has not been able to progress all of these projects with the pace that it would normally aspire to, including not being able to produce 2021 Census results as quickly as ten years ago.

• The Public Health function has been strengthened in response to the pandemic, including strengthening the Public Health Intelligence team.

I hope that this information is of assistance to the Public Accounts Committee, and I am content for my response to be published.

Yours sincerely,

Ian Cope

Chief Statistician and Director of Statistics and Analytics