Sue Beecroft, Joint MD, OPEN VIE, takes 15 minutes to answer some key questions on health informatics in pharma.
1. How would you describe “big data”?
Big data is a bit of a misnomer and often used to cover a wide variety of data sets. We prefer to use the term “Population Based Health Data” as this serves to define the vast quantity of data sets available across England, UK, EU and Global.
Health data can then be split into routinely collected data e.g. data from patient records that is used for clinical or financial management, such as Hospital Episode Statistics (HES) or bespoke collections of data for research from a wide variety of sources.
These data sets can range from entire population coverage - HES covers of 60 million patients in England - to small specialised data sets such as registries and audits.
2. Are pharma and device companies maximising all the data available to them?
The simple answer is no. Some companies have embraced the data available and utilise it across a wide variety of departments. For example, they may use it for HTA submissions, research, value and reimbursement, NHS engagement, and publications. But many companies do not realise what is available to them and rely on old and often inaccurate data.
3. Why do pharma and device companies need to access the different data-sets?
There are multiple reasons why companies need and use the insights and analysis the data provide. This is often dependent on the lifecycle of the drug or the device.
They may wish to understand the current treatment pathways, the health care resource use and mortality. Using HES we can determine the pathways, analyse the variance in care, calculate the costs and understand the mortality and cause. This information is vital for our clients in understanding how their drug or device fits into the pathway and what the burden is in terms of HCRU and reimbursement to the NHS.
4. How does OPEN VIE help its clients from a data standpoint?
We are uniquely placed in OPEN VIE from a data standpoint. Our data division Harvey Walsh is licenced by NHSD and holds the entire HES data set which covers 10 years of data across England for A&E, outpatients, inpatients and critical care. This amounts to over 1 billion records.
These data can be used for:
- Uptake and implementation of innovation and national policy guidelines
- Commissioning support
- Patient pathway analysis
- Disease burden analytics
- Business cases
- Epidemiology research
- Health economic research
- Quality and Outcomes analysis
In addition, we also have access to numerous other data sets across the UK and EU including CPRD, CPRD linked to HES, SACT. As well as local research data sets such as DISCOVER, registries and local NHS data sets.
5. Tell us about your two most exciting projects at present?
We are working on a number of really interesting projects but a couple which stand out involve predictive analytics and artificial intelligence. We are using existing data sets to determine algorithms which can potentially flag a patient at risk of disease before symptoms are present. This has huge implications on health and well-being, and it is great being able to work with our clients to deliver programmes such as this.
We use technology wrapped around the AI to create tools that our clients provide to the NHS which allow the AI to be run securely on NHS systems and then have the software required to manage the patient contact and review. This provides a real, tangible benefit for both the NHS and patients and exciting joint working for our clients.