- This project will commence in September 2017 and run for 15 months.
- Run by Imperial College Healthcare NHS Trust, in partnership with the Centre for Health Policy, and the Big Data and Analytical Unit at Imperial College London.
- Will involve analysing patient feedback given via the Friends and Family Test at the Trust.
- ‘Natural language processing’ will be used to analyse the free text elements.
- Will generate intelligence that will increase understanding of patient experience and enable managers to devise quality improvement interventions.
The NHS Friends and Family Test (FFT) was introduced in 2013 as an anonymous way for patients to provide feedback to service providers and commissioners on the service they have received. The form includes free text, unstructured fields.
At Imperial College Healthcare NHS Trust, there is a huge amount of valuable patient information data collected from the FFT. However, this is currently underutilised due to a lack of human resource and there being no systematic way of extracting useful insights to facilitate quality improvement.
To address this issue and tap into this rich source of information, a team at the Trust is implementing a project aimed at streamlining the analysis of free text patient experience feedback, and using this information to devise effective quality improvement interventions.
‘Natural language processing’ (NLP) is a computer science technique that can transform unstructured text into a structured format that can then be analysed. Using this system, real-time analysis of patient feedback will be available to frontline staff and quality improvement project leads at the Trust.
NLP will mean that visually interesting, easily digestible, interactive reports of patient experience will be produced at ward/service level. The reports will enable a culture of ‘measuring for improvement’ and will provide managers with the vital intelligence they need to devise effective quality improvement interventions.
Free text FFT comments are collected in outpatient, inpatient, maternity and A&E services, and the project team envisages that the intervention will improve care across these four areas.