The opportunities and challenges of General Practice Data for Planning and Research (GPDPR)

There has been much media scrutiny recently about the movement to standardise access to Primary Care data as part of GPDPR (General Practice Data for Planning and Research). My personal opinion is that this has been grossly misrepresented and symptomatic of the click bait culture that social media has fuelled. That is not to say that the announcement of GPDPR could not have been handled better, but expressing the real benefits that this will bring have been suppressed by scaremongering and falsities regarding who may be able to see your data and what they may use it for.

Quite frankly the only way that we will be able to ensure equality of patient care as we recover from the pandemic will be through being able to understand health and care needs across all sectors, and the steps that GPDPR put in place allow all commissioning organisations to understand specific needs from a Primary Care perspective.

With my little rant over, the thought of other commissioning organisations being able to access Primary Care data led me to consider what opportunities and challenges this may bring. This may sound like an odd statement to make, but the focus shifts from being able to access the data to understanding the data and representing an accurate picture of Primary Care.

This will require a significant mind set change, both for GP Practices (the creators of the data) and consumers of the data, whereby using the data highlights the advantages and disadvantages of Primary Care data and what opportunities there are within the health and care system to resolve these.

I’m deliberately choosing not to focus on the patient opt-out aspect of GPDPR, as this feeds into a much wider discussion about patient confidence in NHS data usage!

Primary Care – the data created is being more widely used

The opening up of Primary Care data represents a significant change in how the data that practices create on a daily basis will be used. With the exception of payments for national schemes such as the Quality Outcomes Framework, Primary Care data has been (for the most part) solely used by practices as a record of clinical care specific to their GP practice and how they operate.

This, by it’s very nature, may lead to variations in how practices record “non-standardised” clinical areas, for example the coding of rarer conditions or more specific patient needs. Practices may each utilise different codes for the same clinical area or may choose to note this information in free text which cannot be extracted and quantified within GPDPR.

Historically some practices have been known to use uncommon codes to meet completely different things, such as bitten by crocodile actually meaning seen in specialist clinic. Whilst a silly example, consumers of GPDPR may be in for a shock if they were to find a high prevalence of crocodile bites in one practice!

Practices will therefore need to, in future, understand how their data is being used in a wider context and, if possible, move to a greater standardisation of information. I must stress that this should not be at the cost of good patient care or turning clinicians into full time clinical coders, but as access to the data opens up it highlights the need for the standardisation of coding.

Consumers of GPDPR – what can and can’t be learned from the data

GPDPR also requires a shift for consumers of the data, accepting that all clinical practice is not reportable in a standardised way and that there may be variations in data quality between practices. Other commissioning datasets, such as Secondary Care data through SUS, are much clearer in terms of understanding the accuracy of all coded information.

It is therefore the responsibility of the consumers of the data to understand what is and isn’t coded in a standardised way, and use the data in a fair and responsible way that does not misrepresent Primary Care.

A tangible example – Long COVID

The challenge described above, both for GP Practices, and consumers of data, can be exemplified through understanding prevalence of Long COVID. Only in the last few months Primary Care IT systems have been updated with the ability to code Long COVID in SNOMED. As a result there are some historical patients whose Long COVID status is only captured in comments, whereas others will utilise the new codes available. Practices may also not be aware that codes are now available and may continue to record the information as comments, leading to disparity in Long COVID prevalence across geographies.

Solving this challenge through Primary Care Data Quality functions

As users of GPDPR we therefore need to take a greater responsibility for the data and work with GP Practices to understand variation, and improve data capture. This can be summarised as

1. Identifying a potential data gap
2. Working with clinicians to understand how the clinical area:
   – is currently coded
   – should be coded
3. Create and share resources to assist practices in coding in a standardised and accurate way
   – searches to identify patients coded in the standardised way
   – data entry templates to aide in new patients being recorded
   – protocols (pop-ups) to alert for patients who have been recorded, or who may be eligible but not recorded
4. Use the coded information to see if the data quality has been improved
5. Utilise the data to fully analyse the clinical area

Working in collaboration with Greater Manchester Shared Services (GMSS) Data Quality Team in Manchester we have successfully utilised this process to increase the quality of data across a number of areas, including:

• Register of carers
• Diabetes 8 Care Processes
• Serious Mental Illness Physical Health Checks
• Protected Characteristics

This has a range of benefits, as accurate recording of this data allows both GP Practices and consumers of data to fully understand the patient need, improve the quality of care delivered and, for locally commissioned services that include additional clinical areas, ensure that they are paid for the work undertaken.

The 5 steps outlined above have recently been undertaken for Long COVID, leading to an increase in the recorded prevalence. This has helped the health and care system in Manchester to better understand the number of patients with Long COVID, what other health conditions they have and how as a system we can support them and, if required, commission new services to meet their specific health needs.

Summary

The opening up of Primary Care data through GPDPR represents a shift in the access of data available to the health and care system, but the utilisation of this data must be undertaken accepting both the opportunities and challenges that it utilises. Through GP practices working in collaboration with consumers of the data, and Primary Care Data Quality organisations, all parities can be assured that the data is being used to fairly represent the delivery of Primary Care and improve outcomes for patients.