Kadoorie Centre Critical Care Research Blog: October 2016

Human Factors (HF) is an important part of innovation design. Good use of robust HF principles can make the difference between a new system or device that is welcomed by the healthcare industry, and one which is ignored and unused.

The Oxford University Hospitals NHS Foundation Trust is one of a few but growing number of NHS Trusts in the UK integrating human factors into the hospital work stream to improve patient safety and the working lives of its staff.  A good example of a local project with significant human factors input throughout the development life cycle is SEND, the electronic recording of vital signs system that was designed and developed by the University of Oxford in collaboration with the NHS Trust. You can read more about the SEND project here.

So what does the term human factors actually mean?

Human factors is about designing equipment, work systems and processes, and the environment around the human-being, putting users’ needs and requirements at the centre of the design process. Human factors have been considered in the development of technology in other safety-critical environments such as aviation, the military, railways, and the nuclear industry for some time, and lessons from these areas are now being brought into healthcare. Indeed, those of you reading this who work in healthcare may well have come across the term ‘human factors’ within the hospital context. Crew Resource Management for example, which hails from the aviation industry, has become quite well known within the medical domain.

The Post Intensive Care Risk-adjusted Alerting & Monitoring (PICRAM) project, funded by the Department of Health and Wellcome Trust Health Innovation Challenge Fund, was designed to identify the hospital patients who are most at risk of deterioration after they have been discharged from the intensive care unit (ICU). Just over 400 patients were recruited from the Oxford University Hospitals NHS Foundation Trust and the Royal Berkshire NHS Foundation Trust between January 2013 and December 2014. In addition, electronic records were collated from around 10,000 patients treated at both Trusts between 2008 - 2015. We used the data collected to calculate the ‘normal’ recovery trend for patients who had been admitted to ICU.

The final stage of the PICRAM project has involved designing a working prototype screen interface to display an alerting system to support nurses in the task of following up patients within the hospital after they have been discharged from the ICU. In effect, helping them to identify the patients most at risk of needing further treatment in the period after they have been discharged from the ICU.

To design a usable screen to assist the nurses it was important to understand what the task of following up an ICU patient involves, and what the goals and thought processes of the follow up nursing team were.

The Critical Care Research Group human factors team conducted research which included a series of observations of nurses following up patients who had been discharged from the ICU and admitted to general wards across the John Radcliffe, Churchill and Royal Berkshire hospital sites.  Group interviews were also conducted to understand the nurses ‘experts mental model’.  Hierarchical task analyses were produced from the observational work and a functional abstraction network was created from the group interviews.  Both of these in-depth analyses helped inform what the nurses did, and what their goals were.  Based on these findings a user requirements specification was produced.  This specification was then used throughout the design process to ensure the screen interface designed remained user-centred.

Card sorts (organising pieces of information into categories that make sense to the user) were conducted to understand how the various pieces of information that would be displayed on the screen interface should be organised, so they fitted with the nurse’s usual workflow.  .

This early stage research along with other requirements i.e. the display must include some visualisation of the trend of the patients health formed the basis for the future design process.

The PICRAM team consisting of clinicians, data analysts, software developers and human factors researchers collaborated with a local design company, White October, to produce the screen interface.  A sketching workshop hosted by White October produced a number of initial designs which were reduced to six.  Paper prototype sketches of these six designs were tested with the user group of nurses and the preferred design option taken forward.  A series of tests were carried out and comments on first impressions, what they liked and did not like about the design were fed back to the design team at White October who then made several design changes as the nurses made suggestions.

Following this paper prototype testing phase a software prototype was produced and this will be tested with the nursing team over the next few weeks.  This phase of testing will involve more formalised measures to assess efficiency (time to complete a task), effectiveness (task completion measured against model answers, number of use errors recorded, and user satisfaction), task feedback questionnaires, and the System Usability Scale (SUS) score.

The final version of the system will be presented to representatives from the funding body towards the end of this year.

This work has been featured in the current issue of Ergonomics, the magazine for The Chartered Institute of Ergonomics and Human Factors.