By Dr Tom Foley, Dr Fergus Fairmichael.
Professor Brendan Delaney, Professor of Primary Care Research Kings College London, Project Coordinator and Scientific Director TRANSFoRm project
Mr Seref Arikan, Software developer at Ocean Informatics
Dr Jonathan Richardson, Chair of Informatics Committee, Royal College Psychiatry
Dr Derek Tracy, Consultant Psychiatrist and Associate Clinical Director Oxleas NHS Foundation Trust and lead on PROM/CROM
Dr Tom Foley
There was consensus among the group, that standards have a significant role to play in ensuring data quality and enabling the learning healthcare system. It was felt that all stakeholders should subscribe to and work within a standard. Discussion focussed around current difficulties working with electronic record systems and disappointing progress in this area over the last 10 years. For example it was highlighted that even within the same trust or hospital, it can be difficult to pull longitudinal data over this time frame because the representation of the data may have changed or the electronic record system has changed. This problem is then worsened when trying to work across different trusts or providers, who may use differing systems. This would ultimately increase the workload and subsequently the time frame when trying to conduct work across these differing systems.
There was discussion around the need to know what data should be recorded that could provide useful insights into care and outcomes. This is often focussed upon by clinicians, with the example of mental health, it is often unclear what data should be recorded and what could be used downstream to assess quality of care and how to improve outcomes. There was an argument that we should gain an understanding of the care pathway and outcomes, before thinking about the format of the data, how it is stored and its potential sources. However, it was ultimately felt that flexibility to alter the types of data collected was more important than knowing what to record, at the beginning of the process. Often there may be a change of mind or it becomes clear that a separate variable may play an important role and this may need to be incorporated into the system. If the system does not have this flexibility and there is a need to incorporate a new element, it may create major problems. The focus of standards should be on how we collect and process the data, not what we collect.
Clinical vs Research Data
Traditionally clinical care and research have had fundamentally different requirements and uses of data, however the concept of the LHS would bring these closer together. At present, in order to obtain data from different EHR systems there is a need to use a data node connector which is individualised to that specific system and works at the level of maturity of that system. This will pull the data out of that system and into a standardised format for the work that is being performed. Data elements, based on an archetype, can be used to serve clinical and research purposes.
There needs to be an understanding of what the data is for and who it is for. For this reason, there is a need to engage clinicians with the learning health system and demonstrate benefit for them. Clinicians are key to the recording of data. Fast access to patient information, as provided by current EHR systems, is useful but this does not influence the clinician to record data more accurately as they often do not see the benefit. Therefore data quality can be poor. Awareness, among clinicians, about the potential secondary uses of data is currently limited.
Typically clinicians want to work solely in a clinical context and informaticians want to work purely in informatics and as such, there is a significant knowledge gap between the two areas which would need to be bridged. There may be a role for mediators in this area. Often clinicians have power in the healthcare system and the role of clinical champions in highlighting the potential benefits of such a system would be invaluable. There also needs to be support and partnership, within the NHS, to allow technical people to understand clinical needs and to allow clinical teams to understand technical needs.
Policy and Government
It was felt that a central body driving standards would be important to progressing the field. Due to the lack of a central body, the health IT domain has been dominated by commercial entities and vendors/businesses aimed at maximising profits rather than necessarily advancing capabilities. There has been some recent important developments in this field with the Health and Social Care Information Centre (HSCIC) beginning to drive the field of standards in the UK. Providing legislation to promote the adoption of standards may prove to be useful and may advance the field more quickly. It was also suggested that progress is achieved faster when there is a contractual framework to adhere to. As part of the ecosystem, central guidance is required to give the right drive.
Trust in the system should be developed incrementally. This can be achieved by demonstrating the use of the system with gradual introduction of standards. These standards require evolution rather than revolution, with incremental standards to increase and improve data quality. It was highlighted that there are often user backlashes and teething problems when there are big transformations in healthcare IT systems which can discourage use and engagement.
Social care systems and data are even less comprehensive than those in healthcare. If they were integrated, we could build predictive models to estimate the future social care needs of patients currently moving through the healthcare systems. This could reduce delayed discharges, improve outcomes and improve patient experience.