Using data to make informed healthcare planning decisions

A team of researchers from Durham University funded by Connected Health Cities has created an app which allows GP practices in County Durham to plan future staffing levels using predictive analytics.

The app means that GP practices can begin to analyse patterns in their data in way they haven’t been able to do before. The app allows managers and clinical teams to see projections of GP practice activity with different population growth scenarios.

The team, led by Professor Graham Towl includes Dr Rachel Oughton, Dr Camila Caiado, Dr Ian Briggs and Clare Collyer.

Healthy New Towns

The software has been developed in partnership with local GPs and practice managers in Bicester, Darlington and Bishop Auckland as part of the NHS England Healthy New Towns  programme. Practices in these towns have been provided with a working app to explore how it can be used. The team who have developed the app are looking to share the work they have done with other health and local authority partners.  The new insights from this app could help address problems, requirements and opportunities that could lead to improvements in the provision of front line care services.

The app allows users to better understand current levels of activity and to investigate how different scenarios are likely to affect their practice. For example, one practice manager involved in the programme noticed a significant number of GP and nurse appointments allocated to women aged in their thirties – more so than any other age group.  This insight enabled her to think about how the practice could plan their clinics to meet this demand in a more efficient way.

This work is linked to Healthy New Town’s where large housing developments are being built in ten UK cities (Bicester and Darlington are both designated Healthy New Town’s). The app user has the opportunity and facility to create scenarios relating to new housing developments, with user friendly drop-down menus and sliders.

The app is freely available to NHS service providers.

The research team are keen that NHS service providers are not charged for the use of the app, but that it is a free good as a direct result of publically funded research. They fully anticipate a growth in interest in the work and are keen to hear from others who are interested in health service improvements too. There is potential for some significant financial savings for service providers alongside improvements to the quality and timeliness of services provided too.

How it works

The app uses existing datasets, such as Office for National Statistics (ONS) population projection data for the area and individual practice appointment data. There are several tabs to view different aspects of the projections:

  • Snapshot – population and appointment projections for a specific year
  • Timeline – a general overview of the next 24 years
  • History – displays past data to reveal trends in appointment rates
  • Long term conditions – projected population and resulting appointments for people with a given condition (or combination of conditions)
  • Did Not Attend (DNA) – The rates of non-attendance at appointments.

Detailed projections can be shown for a specified year. The projected population shows the anticipated practice list size, with demographic breakdowns, for the chosen year, with the additional plots showing predicted numbers of appointments by month, or staff type.

By providing information about staffing patterns and GP locum financial costs, the app will also display the change in demand in terms of full time equivalent GPs and nurses, and the potential locum costs and overall impact of the use of locums upon the financial cost base of services. Practice managers and clinical teams may also be interested in any related clinical or patient service costs in terms of a potentially diminished continuity of patient care.

To find out more about the app, contact clare.collyer@durham.ac.uk

Leave a comment

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>