How we're using patient data, new handheld device and new software to help save more lives
Discover how Professor Jim Briggs' research is making a vital difference
鈥淲e try to help doctors predict who is going to live and who is going to die.鈥
That鈥檚 what Professor Jim Briggs will tell you, if you ask what his group's research is all about.
And he鈥檚 got facts and figures to prove it. In the 兔子先生 area alone, his work has led to around 200 fewer people dying or suffering preventable complications in hospital every year.
Jim is Professor of Informatics, in the 兔子先生鈥檚 Centre for Healthcare Modelling and Informatics. Technically speaking, his work takes place at the meeting point between health and information technology.
Put simply, it鈥檚 about applying IT to the health sector. The results are really making waves.
It was a collaboration between the University, 兔子先生 Hospitals NHS Trust, and a private company (The Learning Clinic) that was developing a new software system for nurses to use at the bedside. The system, 鈥榁italpac鈥, is now owned by a company called System C.
The project was sparked by a report that showed several thousand people were dying in hospitals every year, unnecessarily. Why? Because the deterioration in their condition wasn鈥檛 being spotted in time to do something about it.
That鈥檚 what Professor Jim Briggs will tell you, if you ask what his group's research is all about.
And he鈥檚 got facts and figures to prove it. In the 兔子先生 area alone, his work has led to around 200 fewer people dying or suffering preventable complications in hospital every year.
Jim is Professor of Informatics, in the 兔子先生鈥檚 Centre for Healthcare Modelling and Informatics. Technically speaking, his work takes place at the meeting point between health and information technology.
Put simply, it鈥檚 about applying IT to the health sector. The results are really making waves.
It was a collaboration between the University, 兔子先生 Hospitals NHS Trust, and a private company (The Learning Clinic) that was developing a new software system for nurses to use at the bedside. The system, 鈥榁italpac鈥, is now owned by a company called System C.
The project was sparked by a report that showed several thousand people were dying in hospitals every year, unnecessarily. Why? Because the deterioration in their condition wasn鈥檛 being spotted in time to do something about it.
People have to realise that, actually, there are some real benefits out there. If we can get loads of healthcare data integrated, we can look at people鈥檚 whole lifespan of data 鈥 not just the few days they鈥檙e in hospital, but their whole life from cradle to grave.
Hospitals ask nurses to monitor patients鈥 vital signs, including pulse, blood pressure and temperature. When the signs are abnormal, they know something is wrong.
Traditionally, nurses write vital signs information on charts that are typically hung at the foot of the patient's bed. This makes the details inaccessible to anyone who isn't standing next to the patient.
The idea behind Vitalpac was to find a way for this information to transfer immediately to the main computer system, as soon as it was taken at the bedside.
Jim鈥檚 research took things further. He explains:
鈥淭his electronic handheld device made it possible to collect vast quantities of vital signs data quickly 鈥 and very accurately, because the technology was programmed to spot human error, such as an extra zero added to the patient鈥檚 temperature!
Traditionally, nurses write vital signs information on charts that are typically hung at the foot of the patient's bed. This makes the details inaccessible to anyone who isn't standing next to the patient.
The idea behind Vitalpac was to find a way for this information to transfer immediately to the main computer system, as soon as it was taken at the bedside.
Jim鈥檚 research took things further. He explains:
鈥淭his electronic handheld device made it possible to collect vast quantities of vital signs data quickly 鈥 and very accurately, because the technology was programmed to spot human error, such as an extra zero added to the patient鈥檚 temperature!
Vitalpac made it possible to collect vast quantities of vital signs data quickly 鈥 and very accurately, because the technology was programmed to spot human error.
鈥溚米酉壬 was probably the first hospital in the world to amass a large set of vital signs data on patients. Because we had the database, we did what everyone now calls 鈥榖ig data.鈥
鈥淲e used a number of analytical techniques to come up with a model that, basically, converts the vital signs into scores, for each of seven different indicators. They鈥檙e added up across the indicators, and the total score determines what you do next.
鈥淶ero is normal, so the nurse would probably only take your vital signs twice a day. If you score between 2 and 4, they鈥檒l take observations more frequently, perhaps even as often as every half hour. If the score gets to 5, they need to call a doctor or senior nurse. If it鈥檚 7, the doctor must be called urgently.鈥
Colleagues at 兔子先生 Hospitals did a study comparing the years before and after Vitalpac was introduced. It seems to have made a difference to around 200 patients a year. And for those patients, the difference could be life or death.
鈥淲e used a number of analytical techniques to come up with a model that, basically, converts the vital signs into scores, for each of seven different indicators. They鈥檙e added up across the indicators, and the total score determines what you do next.
鈥淶ero is normal, so the nurse would probably only take your vital signs twice a day. If you score between 2 and 4, they鈥檒l take observations more frequently, perhaps even as often as every half hour. If the score gets to 5, they need to call a doctor or senior nurse. If it鈥檚 7, the doctor must be called urgently.鈥
Colleagues at 兔子先生 Hospitals did a study comparing the years before and after Vitalpac was introduced. It seems to have made a difference to around 200 patients a year. And for those patients, the difference could be life or death.
A different kind of supermodel
The scoring system Jim's team helped to develop is set to make an impact as big as the product.
Analysis showed that the 兔子先生 model was a better predictor of outcomes than 33 others that had been published. In Jim鈥檚 view, this is a result of the amount of data they used, which enabled them to set more appropriate thresholds.
Faced with this evidence, the Royal College of Physicians have developed the National Early Warning Score (NEWS), which is based on ViEWS (the Vitalpac Early Warning Score).
By April 2019, every hospital in England must adopt NEWS. Hospital patients across the country will be able to expect better prospects.
But Jim is not content to rest on his laurels. He is part of a collaborative project between the universities of 兔子先生 and Oxford, and the hospitals in both cities. Together, they are looking at other data that could improve the scoring system further.
We鈥檝e got to develop a culture nationally and internationally over the next few years, which is about people taking responsibility for their own wellbeing. Technology is available to make monitoring health easier, and to better inform your health-related decisions.
鈥淲e hope to be able to show that there are even better models, that are even more predictive of patient deterioration,鈥 Jim says.
What鈥檚 more, he recognises that the challenges don鈥檛 begin and end in hospital, especially as people live longer and longer lives. He says:
鈥淲hat we need to do is spot people who are deteriorating at home, or in a residential care home. So one avenue of research is about how you collect vital signs outside the hospital, where you don鈥檛 have a trained nurse doing it.鈥
This is a complex challenge, and not just because of the technology that would need to be developed. Jim explains:
鈥淚f you know what time someone gets out of bed, or has their meals, then changes to those things might be an indicator. But what we don鈥檛 know is what it indicates.
鈥淪o there are other things we need to explore. If we have data from the home, we need to be able to match it up with data about their visits to the GP, and to hospital, and things like that, otherwise we don鈥檛 know what it鈥檚 predictive of.鈥
We鈥檙e entering an age where people accept that more and more data is going to be collected about them, and usefully shared with others. There are some real benefits out there.
The change could 鈥 and in Jim鈥檚 view, must 鈥 extend far beyond new routes to treatment. He believes we need to change our way of thinking about health:
鈥淲e鈥檝e got to develop a culture nationally and internationally over the next few years, which is about people taking responsibility for their own wellbeing.
"Technology is increasingly available to make monitoring your own health easier, and using the collected data to better inform your health-related decisions."
As we鈥檝e seen with Vitalpac, access to large amounts of high quality data makes all the difference in enabling such radical changes to healthcare. Jim believes a cultural shift that may make this easier is already underway:
鈥淲e鈥檙e entering an age where people accept that more and more data is going to be collected about them, and usefully shared with others. But everybody is scared at the moment, because they think their data will be used for marketing, to sell them things.
鈥淧eople have to realise that, actually, there are some real benefits out there. If we can get loads of healthcare data integrated, we can look at people鈥檚 whole lifespan of data 鈥 not just the few days they鈥檙e in hospital, but their whole life from cradle to grave.鈥
Sharing is caring
Looking to the future, Jim is keen to work closely with both healthcare and technology organisations that have large amounts of data.
To build highly effective predictive models, it鈥檚 crucial to draw data from as wide a range of sources as possible. He outlines a number of possibilities:
鈥淢any people wear Fitbits and know a lot about their health as a result, but that information isn鈥檛 really in a form that can be shared with their doctor.
鈥淵our dentist and optician know things about you that aren鈥檛 routinely shared with your GP.
Many people wear Fitbits and know a lot about their health as a result, but that information isn鈥檛 really in a form that can be shared with their doctor.
鈥淵our supermarket, if you have a loyalty card, knows what you eat.
鈥淭here are all these sources of data out there, and when we integrate them we鈥檒l find out great things. I wouldn't be surprised if we change some received medical wisdom.鈥
How can Jim be so sure of this?
Because lots of current medical knowledge is based on studies of hundreds or thousands of patients. If you have data from millions, some answers are sure to change.
For example, studies of diabetes tend to look at patients with diabetes. Studies of cancer tend to look at patients with cancer. But a larger set of data would include people who have both at once.
鈥淭here are all these sources of data out there, and when we integrate them we鈥檒l find out great things. I wouldn't be surprised if we change some received medical wisdom.鈥
How can Jim be so sure of this?
Because lots of current medical knowledge is based on studies of hundreds or thousands of patients. If you have data from millions, some answers are sure to change.
For example, studies of diabetes tend to look at patients with diabetes. Studies of cancer tend to look at patients with cancer. But a larger set of data would include people who have both at once.
There are all these sources of data out there, and when we integrate them we鈥檒l find out great things. I wouldn't be surprised if we change some received medical wisdom.
Amassing large quantities of data would show whether we need a different approach to treating people who have both cancer and diabetes 鈥 because we鈥檇 have more information about what might be different when both are present in the body at the same time.
What about people鈥檚 concerns about data privacy?
In Jim鈥檚 research with 兔子先生 Hospitals, a strict condition is that individual patients cannot be identified. So names, addresses and dates of birth are all stripped out. If age or geography are a factor in the study, a simple birth year or postcode area will suffice.
It might be your data, but nobody looking at it could ever know that.
鈥淪ociety is changing,鈥 Jim says. 鈥淔acebook and Google have changed the way we think about data. Their use of it has raised some issues, and that鈥檚 for them to address.
鈥淏ut now we鈥檙e in a society where certainly young people, and increasingly older people as well, are comfortable with sharing their data. Either because they don鈥檛 care, or because they see the benefits.鈥
What about people鈥檚 concerns about data privacy?
In Jim鈥檚 research with 兔子先生 Hospitals, a strict condition is that individual patients cannot be identified. So names, addresses and dates of birth are all stripped out. If age or geography are a factor in the study, a simple birth year or postcode area will suffice.
It might be your data, but nobody looking at it could ever know that.
鈥淪ociety is changing,鈥 Jim says. 鈥淔acebook and Google have changed the way we think about data. Their use of it has raised some issues, and that鈥檚 for them to address.
鈥淏ut now we鈥檙e in a society where certainly young people, and increasingly older people as well, are comfortable with sharing their data. Either because they don鈥檛 care, or because they see the benefits.鈥
Building healthier towns
Jim is now part of an exciting, University-wide project to see how people can be encouraged and empowered to live healthier lives, based purely on where they live.
East Hampshire District Council have chosen the 兔子先生 as their academic partner in the ambitious Whitehill and Bordon Healthy Towns Project.
The small towns of Whitehill and Bordon have been designated by NHS England as 鈥榟ealthy towns.鈥 Bordon has a large brownfield site on which a lot of new houses are going to be built, along with new schools, a health centre and a shopping centre.
The aim of the project is to develop two communities that, by design, take better care of their health than the average population does. Whitehill and Bordon become a living laboratory.
It鈥檚 a collaborative and highly practical research project that will involve expert academics from the across the University, including architects and civil engineers.
What鈥檚 kept me in the field of healthcare informatics is that it combines academic work with practical thinking. I love working with clinicians and learn so much from them, because they have a completely different perspective from me, as a computer scientist or engineer.
Research will influence all kinds of developments, like making sure the design of pathways and positioning of buildings encourage people to walk or cycle instead of driving.
Crucially for Jim鈥檚 research, houses will be built with wireless networks that will make it easy in future to install telecare technologies (such as sensors and safety confirmation systems, which enable people to live independently in their own homes for longer).
This project is a perfect example of the type of research that has always motivated Jim:
鈥淲hat鈥檚 kept me in the field of healthcare informatics is that it combines academic work with practical thinking.
鈥淚 love working with clinicians and learn so much from them, because they have a completely different perspective from me, as a computer scientist or engineer.
鈥淎nd then, there鈥檚 this reward at the end of the day, that people actually go out and use this research to save lives.鈥
Crucially for Jim鈥檚 research, houses will be built with wireless networks that will make it easy in future to install telecare technologies (such as sensors and safety confirmation systems, which enable people to live independently in their own homes for longer).
This project is a perfect example of the type of research that has always motivated Jim:
鈥淲hat鈥檚 kept me in the field of healthcare informatics is that it combines academic work with practical thinking.
鈥淚 love working with clinicians and learn so much from them, because they have a completely different perspective from me, as a computer scientist or engineer.
鈥淎nd then, there鈥檚 this reward at the end of the day, that people actually go out and use this research to save lives.鈥