University of Southampton

Academic Institute

Erik Mayer

Unknown

No

NIBDAPC_2026_0056

Aim Using information about patients already held by hospitals (such as vital signs, care plans and diagnostic information) we aim to estimate the optimum number of nursing staff required on a ward, to provide a better match between patient needs (demand) and number of staff required (supply). Background If there are fewer nurses than needed on hospital wards, more patients suffer complications and more will ultimately die. However, managers do not know when patients will be admitted or discharged, nor how quickly they will recover or deteriorate. So it is hard to plan staff numbers. Most hospitals use the Safer Nursing Care Tool (SNCT) to guide staffing decisions. For this, nurses record the severity of each patient's illness, and how dependent they are on nursing care, sometimes several times per day. Maybe information already recorded about patients could be used instead. What we will do We will develop algorithms (lists of rules for computers to follow) using patient data to predict the number of nursing staff needed on different wards. We will also survey nurses and run workshops to find out how staffing tools are used and what would help. Working with eight diverse NHS general hospitals in England, we will extract SNCT assessments and patient data from IT systems. We will predict the SNCT assessments from information about patients on the ward at the time of the assessment. Our focus is on patients staying in hospital (inpatients rather than outpatients) and nursing staff (registered nurses and nursing support staff). We will also extract staff rosters, patient outcome data (e.g. mortality rates and readmissions) and ask nurses whether there were enough staff on their current shift. We will use this data to test if our algorithms are valid. Our study is retrospective observational (looking backwards at what happened) and longitudinal (tracking patients over time).

Priority-setting and engagement activities preceded and informed project development. As part of our recently concluded study (NIHR128056), we presented to a group of PPIE (patient public involvement and engagement) contributors from the ARC Wessex PPI Academy. The discussion highlighted a strong sense from participants that adequate staffing in the NHS was vital, and a belief from some that nursing assistants might be suitable substitutes for registered nurses. Participants were surprised by previous research results pointing to adverse effects from higher assistant staffing and were surprised at the lack of regulation around staffing numbers. We discussed further engagement opportunities with lay representatives on our advisory group who noted that this PPI group had very particular experience – they were experienced in PPIE, older and affluent. We actively sought out a different group of non-experienced lay people, who were younger and may have fresh viewpoints and experiences of the NHS. Eleven Health and Social Care students (16-19 years old) from a local college and their tutor attended a workshop. They came from a variety of backgrounds and ethnicities, providing a good reflection of the demographics of our city. Participants discussed their concerns about low staffing in the NHS and their expectations that nursing assistants had mandatory training and that there were mandatory minimum staffing levels (both not currently met). We also discussed the issue of using patient and nurse roster data in our projects. There was general support for use of anonymised data for research purposes by university researchers, where there was no profit motive. This encouraged us to pursue the use of routinely-collected data for this study. In a questionnaire funded by RDS South Central Public Involvement Fund we asked 5 people who had recently stayed in hospitals for their perspectives on nurse staffing, e.g. "What does “enough” nursing staff look like?" A commonly raised issue was the importance of staff being visible and available: "Being able to speak to a nurse if you have concerns", "Nurses and HCAs visible, call bells being answered" and "Are they providing interactive patient care or are they filling in charts, sitting at the computer or talking amongst themselves". The conclusion of this was that patients believe that nurses being available to care for patients should be prioritised over documentation. This helped shape the current research project, which aims to remove one of nurses’ documentation tasks, so making them more available to patients. We also asked what patients would not want overlooked regarding safe staffing, and the issue of having enough staff at night was raised. We will develop these questions further with our public co-Investigator as the basis of PPI activities in this project.

Public Health Research

04/02/2026

Not applicable

Not applicable

One-off

TRE