NHS Safety Thermometer: Evaluating the way data are collected on four common ‘harms’

Timescale: January 2013 - March 2015

Funder: NHS England via Haelo

Key points:

  • The NHS Safety Thermometer is a tool designed to measure improvement in the number of patients affected by four common harms
  • SAPPHIRE was commissioned to evaluate how the data were being collected and used across a range of healthcare organisations
  • The research team will report their findings to the Safety Thermometer team in order to contribute to the tool’s further development

Safety Thermometer cogsBackground

Every year around 41,000 people, or 10% of those admitted to a hospital or care home, suffer avoidable accidents, complications or mistakes. In 2011 the NHS introduced the Safety Thermometer so that organisations delivering healthcare could see where they needed to improve and take action.

The Safety Thermometer is a tool that allows data to be collected on four common harms – pressure ulcers, urinary tract infections in people with catheters, falls and venous thromboembolism, or VTE (a potentially fatal condition where a blood clot that occurs inside a vein can break off and travel to the lungs) – and assembled in a series of 'temperature checks' on a pre-determined date each month

The research study

As part of evaluating and refining the tool, SAPPHIRE researchers were asked to undertake a qualitative study looking at how Safety Thermometer data are collected and used at a range of healthcare organisations. This study, led by Dr Natalie Armstrong, Dr Carolyn Tarrant, and Professor Mary Dixon-Woods, uses a mix of observations in clinical areas and interviews with frontline staff, senior staff within organisations, senior NHS leaders, and experts in the four harms.

Rationale behind the evaluation

Safety Thermometer in useData collection and interpretation has come under increasing scrutiny in recent years: accurate measurement is critical to quality improvement. Without robust and reliable data, improvements may go unnoticed and unrewarded, or worse, poor practice may be reinforced. Dr Armstrong, principal investigator, says: “People don’t always count the same things in the same way. It’s not as straightforward as we expect it to be, as studies such as the Lining Up project have shown.

“We know that there is a lot of variation in the way data are collected and compiled.”


People don't always count the same things in the same way. It's not as straightforward as we expect it to be. There is a lot of variation in the way data are collected and compiled...


The team has completed 126 interviews and looked at data collection and use at 19 different healthcare organisations. The final report was submitted to the Safety Thermometer team at the end of September 2014.  

Research Team

Natalie ArmstrongCarolyn TarrantProfessor Mary Dixon-Woods
Dr Natalie Armstrong Dr Carolyn Tarrant Professor Mary Dixon-Woods

Related links

NHS Safety Thermometer

NHS Institute for Innovation and Improvement, Harm Free Care

Lining up project

Share this page: