Tech suppliers and NHS providers need to work together to improve the sharing of information at discharge. They should begin by embracing the PRSB’s e-Discharge standard, writes Dr Nilesh Bharakhada
As a GP in North-West London, my practice receives significant numbers of transfer of care notifications in the form of discharge summaries and clinic letters from local and regional acute trusts and community care providers.
Every day around three (whole time equivalent) admin staff spend at least half a day processing discharges, working with our pharmacists to reconcile medication, code procedures and diagnoses and forward summaries which require GP actions such as reviews or onward referrals. This labour-intense process is vitally important to ensure patient safety, yet one can’t help feeling frustrated that this information is not more easily ingested by GP systems, in a structured way.
Nationally, 1.5million people use NHS services every day and many hundreds of thousands of those contacts involve discharges.
When and how people are discharged from hospital matters hugely to the individual and can have a profound effect on how clinicians support them. Discharging a person too early without the support they require is unsafe and runs the risk of readmission. Equally, delayed discharges increase the risk of adverse outcomes such as depression and frailty, and can cause capacity issues for new admissions. Even after discharge, people who have experienced poorly coordinated transfer of care can struggle to return to living independently, requiring much higher levels of support than before.
Urgent need for improvement
Given the importance of sharing good quality information at discharge, why are we, as a system, not better at it? The problem is far from new, but there is a renewed sense of urgency to improve transfers of care and care coordination, given the unparalleled strains the health and care systems are under.
Sharing information between services and care settings is integral to good discharge practice, enabling better care continuity and mitigating against unplanned exacerbations or escalations of care.
The PRSB e-Discharge standard, widely endorsed by Royal Colleges, is recognised as representing best practice and defines what information should be shared digitally to enable better transfers and continuity of care between services.
However, the adoption of the standard has been slow, partly because the potential benefits to primary and secondary care are not well understood. There is little awareness of the standard’s impact on managing admissions into secondary care.
Information received in General Practice does not consistently meet GPs’ needs. Discharges are often unstructured, the quality is mixed, and people don’t always receive discharge information that enable them to play a meaningful role in their own care.
So, what can be done? GP systems and workflows need to make it possible to review and update records quickly and efficiently. Suppliers and providers need support to implement SNOMED CT effectively; and PRSB standards and SNOMED CT need to work together.
Better patient outcomes
Finding solutions to common problems would advance transfers of care by making it easier to process important information on medications, allergies, diagnoses, and procedures.
Above all, if we are to win hearts and minds, and change behaviours, we must reframe the importance of good discharges in terms of continuity of care and better patient outcomes.
A national commitment to improving information sharing for transfers of care, ensuring supplier systems are conformant and providing capability to enable better interoperability, would dramatically improve integration of care.
Fundamentally, it’s about working together and recognising that by sharing good quality information at discharge, we can dramatically improve continuity of care, leading to improved patient outcomes. We can all get behind that.
Dr Nilesh Bharakhada is a GP partner and executive clinical director for health and care at the Professional Record Standards Body. He is also a PCN clinical director, chair of quality and clinical governance for Hillingdon Integrated Care Partnership, chair of the RCGP North and West London Faculty and North West London clinical lead for personalised care.
I agree that we need to exchange transfer of care data in a standardised way. Interoperability standards need to be exact, so we require precisely specified FHIR implementation guidelines, including terminology binding wherever needed.
Transfer of care is non-trivial because it handles such a broad subject matter. We need pre-coordinated SNOMED CT codes for everything that might be exchanged. This will require a lot of detailed work, buyt it only has to be done once.
Remember that optionality is the enemy of interoperability.
Tim Benson
Progress has been inversely related to the complexity of the solutions. From my 20+ years of sending discharge summaries electronicially, all a GP wants at discharge is: what was the patient admitted for, what does the GP have to do to continiue the care pathway, what tests/investigations are outstanding, who to contact with queries and what medications was the patient discharged on. The PRSB e-Discharge standard is massively overcomplicated.
Welcome, Charles
Jonathan
Can I play devils advocate?
What is the difference between sharing records while a provider is giving care and at the end of care? With modern technical standards it is possible to share care data with other providers (not just GPs) throughout a patients pathway. Several big EPR providers in the UK already have API’s that allow this (it is known as IHE QEDm, FHIR API and also Care Connect API).
I know from working in several trusts it is good practice to inform other systems and providers if the patient is discharged/admitted. This is just a simple event (known as HL7v2 ADT) which has considerable support in the acute sector and at least one GP provider supports this. Many EPR is other sectors support this (especially if they are US products).
I don’t believe using both of these (complementary standards) deviates from the PRSB dataset. We just have two technical standards working together to meet one record standard. They are reusable to meet other PRSB standards.
However these technical standards aren’t trying to transfer around EHR extracts (which can be quite difficult to keep current and implement), one is supporting sharing of care records (has high concurrency) and the other is supporting clinical workflow.
If a GP supplier wished to import this structured data, they still can (on receipt of the discharge notification).
NHS Digital has delivered a set of technical specifications, known as the Transfer of Care FHIR APIs to allow secondary care providers to send their discharges and outpatient letters as structured messages to the registered GP Practice of the patient. These technical specifications are based on the Information Models defined by the PRSB.
Both TPP and EMIS Health have already enabled a FHIR message receive capability for their entire GP estates in England, and slowly secondary care IT suppliers are adopting these APIs for handover correspondence. If the sender-side just wants or can only do “simple”, then encoded content can be excluded.
The availability of encoded content from source systems, such as EPRs and specialist systems is subject to inadequacies around what information can be gathered, who inputs the information, how it is represented and how it can be output. These problems conspire to make the current flow of FHIR messages minimally structured.
In turn, the current quality of the messages built discourages GP IT suppliers from doing more with consumption of encoded message content. Improvements are only likely to come by recognising that this work needs to happen incrementally, with all secondary care providers initially implementing a minimum viable product and then gradually improving the quality of the message, in a nationally directed manner, so it becomes realistic for GP Foundation IT suppliers to also innovate incrementally.