Better use of data for medication safety in hospitals

Summary

NHS hospital staff spend countless hours capturing data in electronic prescribing and medicines administration systems. Yet that data remains difficult to access and use to support patient care. This is a tremendous opportunity to improve patient safety, drive efficiencies and save time for frontline staff.

In this blog, Kenny Fraser, CEO of Triscribe, explains why we need to deliver quick, low-cost improvement using modern, open source software tools and techniques. We don’t need schemes and standards or metrics and quality control. The most important thing is to build software for the needs and priorities of frontline pharmacists, doctors and nurses.

Content

A study in 2018 found that 237 million medication errors occur each year in the NHS in England alone. Three quarters of these cause harm and there are 1,700–22,303 deaths from avoidable adverse drug reactions.

Two things immediately strike me about these numbers:

1. Medication safety is a huge issue.
2. The breadth of the estimate suggests that data about the scale and impact of medication safety errors are incomplete and unreliable.

I have not seen a similar published study since. My experience working with NHS hospitals since 2018 suggests slow progress. There are four reasons:

1. The spaghetti medusa of data. Millions of staff hours spent to capture and store medicines data in a variety of legacy siloed hospital systems such as EPMA, pharmacy stock and EPR.
2. The NHS employs 1.5 million people and at least the same number again work in social care. Yet there are almost no tools specifically built for either NHS or social care workers.
3. Slow progress of clumsy digital initiatives that focus on the wrong thing, made worse by the fear of digital monsters.
4. Lack of change and innovation. Lots of noise around schemes and gateways rather than actual solutions for real people.

Layer the pandemic impact over these underlying issues and the position seems hopeless. It's not. 

“Data isn’t oil, it’s sand.”

The tech industry has invested trillions of dollars and the time of millions of the world’s smartest experts. Much of this goes into solutions that capture and use epic quantities of data. Over the past 15 years, multiple standard, open source software tools and techniques have emerged that tackle exactly this kind of problem.

Behind all the hype, hysteria and scaremongering, the current AI boom is just a manifestation of all this money and intellectual capital. It is outrageous that this is not used for the benefit of hard pressed frontline hospital staff.

So what does this mean in practice? How can tools, like Triscribe, actually improve medication safety?

Those 237 million errors include a lot of different things. Adverse drug reactions are just a small portion and the severe reactions are pretty rare.

Using the existing data collected from a multiplicity of systems, we believe that more meaningful analysis is possible by:

• Reporting of adverse drug risks updated at least daily. Note: using a little AI, we can predict the risk of adverse drug reactions and give clinicians the information needed to stop at least some from happening. Much better than just reporting the incidents.
• Monitoring adherence key safety policies and guidelines. For example, VTE prophylaxis, allergy reviews and oxygen prescribing.
• Tracking and reporting late and omitted doses every day across all systems, including ward comparisons to identify learning and share better ways of working.
• Safe use indicators for specific medications; for example, early/ late administration of Parkinson’s medicines and opioid deprescribing.
• Reporting key compliance measures, including IV to oral switching for antibiotics, high dose prescribing of opioids and usage of methotrexate

The possibilities are limitless. There is no shortage of data in the NHS. However, the ability to share that data between systems and organisations is something the health and care sector still lacks.

It’s a solvable problem. Deliver quick, low-cost improvement using modern, open source software tools and techniques. We don’t need schemes and standards or metrics and quality control.

The most important thing is to build software for the needs and priorities of frontline pharmacists, doctors and nurses. Keep learning and keep improving every day. 

About the Author

Kenny Fraser is the CEO of Triscribe. Prior to joining Triscribe, he was a senior consulting partner with PwC.

Triscribe is an early-stage digital health company, a spinout from the University of Edinburgh. It builds analytics and AI-based on data about hospital medication usage. Triscribe are experts in combining data from all EPMA and pharmacy stock systems in hospitals to provide a complete picture of medicines’ usage to improve pharmacy ward rounds and performance across all aspects of medicines stewardship and medication safety.