Dear Lewis,

WHAT SHOULD PRIMARY CARE LOOK LIKE FOR THE NEXT GENERATION? OSR INQUIRY RESPONSE

We have today published our response to the Health and Sport Committee’s inquiry looking at the future of primary care in Scotland (see annex).

Statistics published by public sector bodies should be produced in a trustworthy way, be of high quality, and provide value by informing answers to society’s important questions. Our submission outlines our view that primary care statistics in Scotland do not currently deliver public value because many important questions cannot currently be answered.

The Information Services Division of NHS Scotland (ISD) has worked with general practices across Scotland to develop SPIRE, a unique asset for Scotland’s health data landscape. It has overcome significant challenges to deliver a system that has the potential to transform understanding of primary care and support a wide range of information needs in a way that has not been possible before now. However, major information gaps remain that need to be addressed to ensure that decisions about the future of primary care in Scotland are evidenced-based, open to scrutiny, and can answer the questions people have about how primary care contributes to improving health and wellbeing. We have made some recommendations for ISD to help meet these aims.

I look forward to seeing the conclusions of your inquiry. I am copying this letter to Scott Heald, Head of Profession for Statistics, ISD.

Your sincerely
Ed Humpherson
Director General for Regulation

ANNEX
SCOTTISH PARLIAMENT HEALTH AND SPORT COMMITTEE INQUIRY WHAT SHOULD PRIMARY CARE LOOK LIKE FOR THE NEXT GENERATION?SUBMISSION FROM THE OFFICE FOR STATISTICS REGULATION

What we do

1. The Office for Statistics Regulation (OSR) is the independent regulatory arm of the UK Statistics Authority. We provide independent regulation of all official statistics produced in the UK, including those in Devolved Nations and the NHS. Our regulatory work is underpinned by the Statistics and Registration Service Act 2007.

2. We set the standards official statistics must meet through the statutory Code of Practice for Statistics. We ensure that producers of official statistics uphold these standards by conducting assessments against the Code. Those which meet the standards are given National Statistics
status, indicating that they meet the highest standards of trustworthiness, quality and value. We also report publicly on system-wide issues and on the way statistics are being used, celebrating when the standards are upheld and challenging publicly when they are not.

3. We have staff in three locations: Newport, Wales; London; and Edinburgh.

This submission: the public value of primary care statistics in Scotland

4. Statistics published by public sector bodies should be produced in a trustworthy way, be of high quality, and provide value by informing answers to society’s important questions. To answer the question “what should primary care look like for the next generation?” it is essential to understand how primary care is delivered, what outcomes it achieves for people, and to be able to monitor changes in both of these over time.

5. This submission outlines our view that primary care statistics in Scotland do not currently deliver public value.

What do we know about primary care in Scotland?

6. The Scottish Parliament Information Centre briefing – Primary Care in Scotland – provides a helpful overview of information about this topic. The Information Services Division of NHS Scotland (ISD) publishes comprehensive statistics about the primary care workforce (e.g. GPs, dentists, optometrists, pharmacists, allied health professionals). Detailed information is also available about GP practice populations and contracts and out of hours provision. Extensive prescribing data is provided via statistical reports, a dashboard and open data. Information about eye-care in primary care settings is published annually. Patients’ views about primary care provision are
captured by the Scottish Government’s biennial Health and Care Experience Survey of people registered with a GP practice.

What don’t we know about primary care in Scotland?

7. While the size and composition of the primary care workforce is well documented, the main statistics based on administrative data only include headcounts. Information about whole time equivalent workforce numbers, which are more useful to users for workforce planning and understanding capacity in the system, has to be collected via surveys of GP practices, which have been conducted biennially in recent years. The most recent statistics are for 2017 and there is no published information about plans for future data collections.

8. We know more about primary care provision when GP practices are shut than when they are open, thanks to the data collected about out of hours provision which includes numbers and types of contacts, contact outcomes, and contact rates by age and gender.

9. Our analysis, and the knowledge we have accumulated about health and social statistics users’ needs through our regulatory work, has identified a range of questions that illustrates the kinds of issues that, at present, statistics cannot answer. We are grateful to the statistics users who shared their views with us on this topic. We recognise that not all of these questions will have the same level of interest from users
and, as with all areas of statistics production, some prioritisation is necessary. We also appreciate that there are significant challenges involved in collecting data from a federated system with over 900 independent contractors. These include:

• How many GP consultations and other primary care contacts take place per year?
• How does primary care service use vary by socio-demographic group, health status,
protected characteristics and area?
• How many booked appointments are not attended?
• What is the level of unmet demand for primary care services?
• Why do people consult their GPs and other primary care providers?
• What health assessments are carried out in primary care?
• What is the prevalence of individual and multiple health conditions, including frailty, among people registered with GPs – and do the condition codes with primary care correspond with data held in other NHS data systems?
• How many people registered with GPs live in care homes or are house-bound?
• How many people in Scotland have an up to date Anticipatory Care Plan?

10.Beyond these individual questions we also know there is a huge appetite for joined-up statistics that paint a more complete picture of people’s journeys through the different parts of the health and social care system, especially for people with multiple long-term conditions. Scotland is well placed to provide such insights thanks to its longestablished use of an individual patient identifier, the Community Health Index (CHI).  ISD’s recently published statistics on emergency hospital admissions among people living in care homes provide a glimpse of the potential that exists to understand more about the relationship between primary and secondary care via more joined-up statistics.

Why are there primary care data gaps in Scotland?

11.Some of these gaps are partly due to the closure of the Practice Team Information (PTI) programme that had, until 2012/13, provided national-level statistics about GP consultations. PTI used data drawn from a sample of 6% of practices in Scotland which meant it was not a large enough to be used to produce statistics for local areas.

12.A new data infrastructure – the Scottish Primary Care Information Resource (SPIRE) – has been developed that should, in time, enable data to be collected on a universal basis from all general practices in Scotland (currently over 900). Coverage of practices has now reached 95% and complete coverage is expected once IT issues have been resolved in the remaining ones. This data will help to meet a much wider range of users’ needs than the sample from PTI was able to. Initially, SPIRE’s development has focused on supporting the information needs of individual general practices and clusters of practices. ISD informs us that in the year ahead SPIRE will be developed further to meet a wider range of purposes, including support for health and social care partnerships and Scotland-wide data. This would enable the production of populationwide information for research and statistical purposes.

13.The scale of transformation required to build a system capable of delivering data safely from over 900 independent contractors, with multiple types of IT systems in place, should not be underestimated. The progress made by ISD and general practices to date is commendable. It is also helpful that SPIRE’s roll-out will benefit from ongoing investment in a new general practice IT infrastructure. However, progress on delivering data for practices is currently further ahead than plans for using the data to generate population-level statistics about primary care. There is limited information currently available about when new primary care statistics will be delivered and what they might look like. We encourage ISD to address this.

14.It is worth reiterating that some of the questions highlighted above could be difficult to answer using routine systems without placing additional data collection burdens on primary care staff, potentially reducing the time available for direct clinical care. Some might prove to be impossible to answer. Understanding these kinds of limitations is important. Therefore, greater transparency about what statistics SPIRE (and other primary care data systems) can and cannot deliver would be helpful to users.

What is the picture like outside Scotland?

15.All of the UK’s countries lack statistics about what is actually delivered in primary care and about outcomes for patients. For example, much of the Nuffield Trust’s analysis of gaps in English primary care data also applies in Scotland.

16.There are also examples of statistics available in some parts of the UK that Scotland either does not currently have, or does not produce in the same joined-up format:

• In England, new data resources bringing together primary care information are being developed by NHS Digital that will, in time, provide a much fuller and more accurate picture of primary care in England. For example, since December 2018 England is the only part of the UK with statistics about GP appointments (such as volume, type of health professional consulted, elapsed time between booking and appointment) which captures some, but clearly not all, primary care activity. These are currently experimental and still undergoing development.

• In Wales, Welsh Government publishes annual statistics about GP access in a report that brings together information about trends in practice numbers; opening hours; appointment availability; and patient satisfaction with, and experiences of, accessing GPs.
• In Northern Ireland there is an annual compendium report on Family Practitioner Services with information about a range of primary care services: GPs, ophthalmology, dental health and pharmacy. UK-wide comparisons are presented, where available. The statistics are also released quarterly on a provisional basis.

Is comparable primary care information available across the UK?

17.As with other health service and population health statistics, there is no single portal to bring together UK-wide information about primary care. Users therefore need to source information from each country separately, rely on sign-posting between individual publications, and then establish whether the statistics are directly comparable.

18.Workforce data are available in all countries based on headcounts and numbers of GPs per registered patients, but there are issues with comparability which make it hard to answer questions accurately. A UK-wide publication about trends in the general practice workforce was discontinued in 2017. As noted above, whole time equivalent workforce statistics are valued by users, but these are only available biennially in Scotland, are published quarterly in England, and are not available in Wales and Northern Ireland.

19.Scotland, Wales and England all produce statistics about patient satisfaction with GP services. However, as these surveys are not conducted using harmonised questions, levels of patient satisfaction cannot be directly compared between counties.

The Office for Statistics Regulation’s recommendations for Scottish primary care statistics

20.ISD has worked with general practices across Scotland to develop a unique asset for Scotland’s health data landscape. It has overcome significant challenges to deliver a system that has the potential to transform understanding of primary care and support a wide range of information needs in a way that has not been possible before now.

21.However, major information gaps remain that need to be addressed to ensure that decisions about the future of primary care in Scotland are evidenced-based, open to scrutiny, and can answer the questions people have about how primary care contributes to improving health and wellbeing.

22.To achieve these objectives, ISD Scotland needs to:

• publish its plans for official statistics about primary care;
• ensure that primary care statistics are produced in a coherent way that brings together relevant information to tell a clear story;
• work with a wide range of users to ensure that primary care statistics meet their needs.