A reset for the FDP – 3 things the CDAON wants to see next to create short term wins and longer term strategic alignment
Marc Farr
Chair – Chief Data and Analytical Officers Network
Association of Professional Healthcare Analysts
8th April 2025
Ming Tang
Chief Data and Analytics Officer
NHS England
Dear Ming
SUBJECT: A RESET FOR THE FDP – 3 THINGS THE CDAON WANTS TO SEE NEXT TO CREATE SHORT TERM WINS AND LONGER TERM STRATEGIC ALIGNMENT
Thank-you for your reply to our recent letter, which has been received very positively across our network and across the wider NHS, academia and industry. Congratulations too on your new appointment – we wish you the very best of luck.
With recently announced changes to NHS England and Integrated Care Boards, the need for a shared understanding of the role of the FDP in the short-term has become urgent. The coming months are likely to see changes in the structure, relationships and roles of ICBs and NHS providers, and the FDP has the opportunity to be a critical enabler during this key period. There are significant data engineering economies of scale that could be delivered if the FDP laid out a common platform for us all, removing the need for DSCROs and other ancillary services.
Though figures on sign-up to the FDP are promising, our conversations among members suggest that utilisation may be much lower. As things stand, we do not feel that the FDP is in the position to fulfil its potential here and that a reset is necessary.
In this context, we have set out below what we think are the key themes and asks from our network that still need to be addressed. We would welcome a detailed and time-specific response on each of the points raised. The three key themes are:
• Vision and Roadmap
• Partial Adoption
• Support, Training and Evaluation
1. Vision and Roadmap
For such a high-profile programme it is important that there is a clear strategy with timeframes for what the FDP aims to achieve. Currently we have a series of very specific acute provider use cases, ambitions around ICBs using the tool for PHM when the IG issues around GP data seem insurmountable and the existence of a range of mature solutions already in place. Given that the number of ICBs is likely to reduce this means that is likely that an even higher number of ICBs will have access to one of these mature solutions.
a. Enabling functions
For ICBs, if FDP is to work as an enabler over the coming months, then it will require:
i) patient-level commissioning data availability;
ii) functioning ability to link additional datasets (via PET/DSCROs);
iii) simple IG infrastructure for federating across areas;
iv) compatibility with existing analytical tooling (SQL, Power BI, R, Python);
v) provision of data at PCN level and below.
Please provide specific timeframes for how FDP will be able to support each of these.
b. Relieving the reporting burden
There is a huge opportunity to finally ‘relieve the burden’ of regular returns that has been promised for so long. The FDP could become the catalyst for automating Sitrep and similar reporting for us by forcing EPR providers to make data available in the right way. We hope that given the investment in the FDP we can leave behind a system of manually uploading excel templates through a portal that only allows a limited number of users which is the current status quo.
Please provide specific timeframes for how we will be able to utilise FDP to relieve reporting burden on providers.
c. Function
There are a range of functions that lend themselves to a centralised approach. Notwithstanding local nuance, these include as examples demand and capacity modelling, performance and contract management, board reporting, benchmarking, cross border analysis, the servicing of key governance meetings (urgent care boards, improvement boards etc). With strategic commissioning cited as the new function of the ICBs it would be really helpful to see a suite of tools within the FDP being developed to support this. It is our view that these areas are where the roadmap should develop into.
While PHM is clearly important in the left-shift that we need to deliver we are not convinced that the FDP is the vehicle to deliver it, particularly given the sensitivity around the use of GP data. We cannot continue to pretend that PHM is possible without access to GP data in a local linked dataset and in any resultant risk-stratification.
That said, there are exciting opportunities for the FDP to contribute across the ICS landscape helping link data across local authorities. We believe it is currently a missed opportunity not to work more closely with local government and the public health industry.
Please provide a comprehensive detailed roadmap on FDP future functionality to include strategic commissioning as well as the next phase of provider and ICB/ICS use cases.
d. SDE
Is the FDP the pre-cursor to linked data within an SDE? Across the country we need clarity in the FDP/Palantir roadmap and how it relates to the SDE so that we can plan accordingly. We need to understand how programmes such as Opensafely can link to both, so that we can design how we implement different data models for example.
Please supply a clear statement on how the FDP relates to the SDE, covering relevant technical infrastructure implications.
2. Partial Adoption
a. Switchover and Dual Running
It’s really exciting that the government sees the benefit of federating data and has made such a large investment in this area. It’s really important though that no providers or systems are directed to turn off existing systems until comparable utility is available with the FDP. As a network we seek assurance that providers and systems can continue to use existing systems that are proven to deliver patient benefit until this point. Anecdotally we are aware of systems being directed to close down existing systems because the functionality is planned within the FDP roadmap, however we are not convinced that the functionality is imminent and therefore that a risk to patient safety exists – we can cite specific examples. In your response you said that ‘We are exploring the feasibility of a partial adoption model and how local capabilities can be integrated into the FDP platform to ensure a smooth transition.’
Please provide specific timeframes for when details of a partial adoption model will be released. Please also confirm that existing systems will not be directed to end until there is assurance of a smooth [and risk-assessed] transition to a new service.
b. Open-source approach
We welcome your commitment to aligning FDP to the principles of the Goldacre Review. Key to this will be that many providers and systems will want to use FDP as part of a wider data and analytics infrastructure. There is still no write-back function available within the tool.
Please provide details for when and how we will be able to export data and/or link from the FDP into other data warehouses and clinical systems.
c. External Providers
Finally, we need to understand how we can work with external providers (beyond Palantir) within FDP.
Please confirm when commercial terms for industry partners will be available so that we can progress our long-term strategies in this area.
3. Support, Training and Evaluation
a. Support and Training
We welcome your commitment to comprehensive training and support for key skills within FDP. Our experience so far has been that the offer has fallen short of this, consisting of videos and short technical sessions.
The investment thus far has been in a software platform, not in the analysts that will need to learn it, share within it, liaise with clinical and operational staff in how to maximise the benefit from it. We need to see proper consideration to how we build the analytical workforce of the future [as promised in Jim Macky’s recent letter [to invest in analytics] funded from within the FDP programme, this cannot simply be left to the private sector to resolve.
Through APHA and the CDAON we are launching a range of initiatives to deliver analyst training but these are simply not at the scale that they need to be to leverage the opportunity that comes from an expert analyst function.
Considering the significant investment in management consultancy around the programme we would be keen to offer our support in developing the training offer from our professional network; designed by analysts, for analysts. Funded properly, the CDAON is offering here to oversee implementation and utilisation.
Please outline how the training offer will be developed in the coming months. Please set out how the analyst community will be utilised to develop training materials for users and to oversee the implementation of the FDP.
b. Analyst development and structures
Given the importance continually attached to the development of the analyst community and the huge growth in the data science industry we now need to see stronger strategic direction in the professionalisation of the analyst. We need to see a move from this key role from admin/clerical to scientific.
In the ICB operating model we would encourage you to clearly articulate the need to ensure systems are designing integrated digital, data and insight functions that bring together the BI and PHM functions.
Please set out a timetable for moving analysts into the scientific classification.
c. Evaluation
Finally, given the scale of investment in the FDP it’s really important that we see in place a comprehensive set of resources of case studies and detailed evaluations so that we can understand the impact and share it to more and more potential users.
Please share materials setting out the most significant impacts of the FDP to enable us to develop business cases and consider how we organise our reduced resources in an open and transparent way that can be verified (i.e. with open-source methodology).
Please commission or develop a dashboard to show usage by organisation so that we can be signposted to those furthest along and contact for advice on implementation.
Thank-you for your attention in reviewing the letter and we look forward to your reply and to helping to shape the direction of the FDP going forward through analysts being at the heart of its design. We believe the partial adoption model you mentioned represents the best way forward for delivering insight to the front line and for strategic commissioners and the CDAON are willing to work with your team to ensure that we can leverage the national investment that has been made, reduce unwarranted duplication where it exists and ensure we grow the skills and knowledge of professionals appropriately.
Kind Regards, on behalf of the CDAON,
Dr Marc Farr LFEDIP
Chief Data and Analytical Officer
East Kent Hospitals NHSFT and NHS Kent and Medway
National Chair, Chief Data and Analytical Officer’s Network, APHA
https://www.aphanalysts.org/nhs-chief-data-and-analytics-officer-network/
E: marc.farr@nhs.net