The power of AI: Birdie and Faculty begin an exploratory research project into polypharmacy
October 29, 2020
At Birdie, our mission is to improve the lives of one million older adults by 2023 - and for our latest project, we’re harnessing the power of artificial intelligence (AI) to begin a research programme that will use insights from our partners to help highlight and address problems that occur as a result of polypharmacy.
First, what is polypharmacy?
Polypharmacy is generally understood to refer to the concurrent use of multiple medication items by one individual.
Polypharmacy is becoming increasingly common. In the past decade, the average number of items prescribed for each person per year in England has increased by 53.8 per cent, up from 11.9 (in 2001) to 18.3 (in 2011). (source)
Problems associated with medication use in general, and polypharmacy in particular, are more prevalent in the aged population because of numerous contributing factors. (source) Research has clearly established a strong relationship between polypharmacy and negative clinical consequences. Adverse reactions to medicines are implicated in up to 6.5 per cent of hospital admissions. Patients admitted with one drug side effect are more than twice as likely to be admitted with another. Patients on multiple medications are more likely to suffer drug side effects. (Pirmohamed et al 2004). (source)
Why do we care?
We know that providing the best care goes beyond our software. That’s why we’re always looking for new ways to innovate in our sector and help to build solutions that aid the entire care pathway.
Our software system has the ability to capture hundreds of thousands of data points a day - and we know we can use this for good. That’s why we’re excited to announce that we are hosting a Faculty fellow, who over the next six weeks will use anonymised data to construct a model that will provide insights into our clients’ medication usage and the effects of polypharmacy.
"The digital social care sector is privileged to have the most extensive view of a person's overall wellbeing on a daily basis. At Birdie, we recognise the potential this kind of data has for allowing analysis from a health perspective. We are thrilled to be part of the Faculty fellowship program, to use our customers' data in an ethical way to improve their care." Kath Middleton, Clinical AI Product Lead, Birdie
How will we do it?
Faculty is a leading applied AI company, with experience working on projects with private sector industry, government and healthcare. As part of its wider offering, the company has a Fellowship Programme that helps the brightest STEM PhDs to make the transition from academia into industry roles by providing the opportunity to work on impactful real-world projects. It’s through this programme that we’re delighted to welcome Oscar Bennett to the Birdie team.
"This is an exciting opportunity to explore the application of data science to an important problem in social care. I hope this collaboration between Faculty and Birdie can be a step towards helping us to find more ways to improve the lives of people with complex medical and social needs."
Oscar Bennett, Data Science Fellow
Oscar studied Medicine at Cambridge University before completing a Masters in Medical Engineering at University College London and worked as a clinician in the NHS before transitioning into digital health and data science. At Birdie, he will begin researching into the issues with polypharmacy and how we can use anonymised data to help us build the bigger picture, tying academic data manipulation methods to real-world impact within health and social care.
Birdie allows detailed daily recording of the medications a care recipient is taking, as well as regular checks on how they are feeling. With Faculty, we’ll be looking at how we can better understand how a care recipient’s medication regime is interacting and flag any unexpected and potentially concerning interactions.
What do we hope to achieve?
The challenge for clinicians is keeping up to date with new drugs as they come on the market and being aware of the interaction between them in patients being treated for a number of medical conditions.
We’re hoping to be able to predict side effects/interaction effects from a particular medication schedule, automatically. That means that you can get an even deeper insight into your client’s health, and instantly be able to understand medication effects and mitigate them where possible.
“We're excited to partner with Birdie on this research project into polypharmacy. We believe there is enormous potential for the application of AI into social care to improve patient outcomes, support carers and ensure timely interventions for changes to care provision. This research is an opportunity to explore the possibilities for using AI to understand the interactions of different medications. For Faculty this is just the start of exploring the potential for accelerating the adoption of AI in social care.” Meghan Leaver, Healthcare & Life Sciences Principal at Faculty
We know that you might have a few questions about our ongoing project. We’ve answered a few below, but if you have a specific question about our research, results, or partnership with Faculty, you can get in touch with us by emailing: email@example.com.
Are you using personal information?
No. All data we use for this project will be anonymised, as part of our agreement with our partners.
Are you giving/selling info to Faculty?
No. Faculty are working only with anonymised data (which means it can’t be linked to a person). They’ll only use this anonymous data to develop a model and algorithm, which will only be used by our Birdie product to identify clients who may be at risk.
How do you know it'll work?
This is just the first step in exploring what insights we can gather from our data. If it’s a success we’ll explore iterating and further co-developing projects with our partners.
This is the start of a research programme to make the best use of data that we can provide as insights to our partners. As part of this, we will also continue to refine our approach and processes, to continue to ensure anonymisation, and compliance with GDPR and other legislation. We'll continue to communicate with our partners about this in line with our value of transparency and integrity.