Skip to content

Q logo

Meet the team: OPSSI (Oxford Patient Safety Sepsis Initiative)


  • Jo Murray, Sepsis Patient Safety Manager, Patient Safety Collaborative, Oxford Academic Health Sciences Network (AHSN)
  • Dr Andrew Brent, Clinical Lead, Sepsis Stakeholders Group, Oxford AHSN; Infectious Diseases and Sepsis Clinical Lead, and Consultant in Infectious Diseases & General Medicine, Oxford University Hospitals NHS Foundation Trust (OUHFT); Honorary Senior Clinical Lecturer, University of Oxford
  • Professor Peter McCulloch, Patient Safety Academy lead, Professor of Surgery & Fellow of Trinity College, Oxford.
  • Clare Hird, Sepsis & AKI Specialist Nurse, OUHFT
  • Louise Rawlinson, Matron, Emergency Department (ED), OUHFT
  • Dr Ravi Pattanshetty, ED Consultant and ED Sepsis Lead, OUHFT
  • Geoff O’Donoghue, Sepsis Patient Representative, OUH Sepsis Working Group & Oxford AHSN Sepsis Stakeholder Group
  • Matt Woodward - Human Factors Practitioner, Patient Safety Academy (PSA)
  • Dr Clare Dollery, Deputy Medical Director, OUHFT; Chair, Sepsis Stakeholders Group, Oxford AHSN


Sepsis is a major cause of avoidable death. Early recognition and treatment of sepsis might save up to 10,000 lives in the UK each year, but survival falls by up to 7-8% for every hour delay in treatment (Dellinger et al. Intensive Care Med 2012). There has been a national Commissioning for Quality and Innovation (CQUIN) in place for the last 3 years encouraging acute organisations to ensure that patients who have suspected sepsis receive IV antibiotics within 1 hour.

We developed a standardised approach to sepsis management based on NICE sepsis guidance and successful elements of existing systems (Murray & Brent. Clinical Medicine 2018 in press). This has been successfully implemented in acute hospital Trusts in the region. However the proportion of patients with sepsis who receive antibiotics within 1 hour remains below the 90% target locally, regionally (in the AHSN partner organisations), and nationally.

Delays occur at several parts of the patient pathway (figure 1). Further improvement requires a deeper understanding of the remaining barriers to early identification and treatment of patients presenting to the emergency department (ED) with sepsis. Given the substantial efforts and advances already made, significant improvement will require additional approaches.


To understand better the remaining barriers to more rapid recognition, assessment and treatment we propose to apply a formal Human Factors (HF) approach to analysing and redesigning the ED sepsis patient pathway.

Our detailed prospective data (figure 1) show the segmentation of the problem and immediately point to possible areas for study.  In each segment, attacking the “tail” of outliers is likely to be more effective than attempting to move the entire distribution leftwards, and the segment most in need of improvement is Alert to Prescription.  A HF and ergonomics approach will allow us to understand better the reasons for ongoing treatment delays, and to develop, implement, assess and share effective solutions.

1. Location & stakeholders

The project will be conducted at (OUH). OUH Strengths include:

·         Clinical pathways representative of many acute Trusts, including a rapid assessment pathway;

·         Electronic patient records and vital signs allow reliable process and outcome data collection to correlate with the qualitative work proposed;

·         Unique access to the Patient Safety Academy (PSA), providing a critical mass of professional ergonomic and clinical expertise to ensure delivery. The HF/QI approach described has been successful in recent projects at OUH and elsewhere

·         Broad clinical and patient stakeholder involvement (see team details);

·         project management and network support from Oxford PSC including access to patients and relatives with lived experience of sepsis.

2. Process Mapping and Human Factors analysis.

Patient Safety Academy (PSA) ergonomist(s) experienced in HF healthcare assessment will develop a detailed process map of the sepsis patient pathway in ED using:

·         Ethnographic observation of the current clinical pathway over several days distributed over several weeks

·         Structured interviews with staff, patients and relatives with thematic analysis

·         Development of Hierarchical Task Analyses for pathway segments allowing a Failure Modes & Effects Analysis (FMEA) to identify areas where improvement would yield maximum benefit

3. Intervention

HF specialists will convene multidisciplinary groups including frontline staff and patients to co-design the pathway focusing on improved procedures at identified “hotspots”.  The revised pathway will undergo rapid iterative testing using a PDCA cycle approach.  An implementation strategy will be developed with management to overcome potential barriers to widespread adoption.

4. Shared learning

We will share learning widely with the following organisations, emphasising the need for local testing and modification by staff co-design groups:

·         OUH departments;

·         Oxford AHSN partner organisations via regional sepsis stakeholder group and regional ED collaborative

·         National Deteriorating Patient Workstream via 15 Patient Safety Collaboratives (PSC) & Academic Health Science Networks (AHSN), feeding into NHS England Cross-System Sepsis Programme Board;

·         UK Sepsis Trust

  • Potential academic publication
  • Q community via Q Exchange pages, blogs, news items and relevant SIGs

We will use the findings to develop a sepsis training package for which we are currently exploring separate funding (see below).

5. Timeline

If successful the timing of this bid will allow observation and data collection to be undertaken during the busy winter period. Analysis and process mapping will take place in the spring with interventions piloted in the slightly less busy summer period, prior to full implementation in time for next winter.

Benefits for Q:

Shared learning of project findings (see above).

We are also actively exploring separate funding to co-develop a local and regional (± national) training package with Oxford Simulation, Teaching and Research (OxSTAR).

Successfully implemented, the project will be an exemplar for other projects seeking to apply HF and ergonomics approaches incorporating staff, patients and relatives in a busy ED department setting.

How you can contribute

  • • Has work of this nature been conducted for sepsis elsewhere in the UK?
  • • Are there any specific elements that would be important to include to maximize the transferability/generalizability of any learning to other settings?
  • • What other elements should we consider as part of this project?

Further information

Figure 1 OPSSI (PDF, 203KB)


  1. Hi Jo. Have you seen this project? The final report (at the bottom of the page) details quite a lot of the work, but if you'd like to chat about it, then do let me know and I'll help where I can (I was the designer on the project, working with a big team lead by Dr Chris Subbe from Betsi Cadwaladr University Health Board). Good luck :)

    1. Hi Hawys, I shared this with the team. Feedback has been

      "It's an interesting project, though the sepsis 'icons' have now largely been overtaken by colour coded early warning scores in most settings. The 'wee wheel' may be of interest in relation to AKI so will share with renal colleagues". 

      Thanks gain for sharing Jo

    2. Hi Hawys. Thanks for sharing this and I've forwarded to the team. There will be some challenge with this in the ED setting rather than wards and the sepsis icons are based on SIRS where NEWS plus suspicion of infection is now being seen as more appropriate. Always good to revisit though as may spark some ideas going forward. Many thanks Jo

  2. So...not shortlisted :( but on the reserve list :)

    Please keep your comments coming in case we get a second chance!

    Many thanks Jo

  3. worth remembering the new CQUIN change means that the time to antibiotics begins at the diagnosis of suspected sepsis in a hospital setting. Ambulance trusts will have different metrics to test their screening, speed of transfer.

    The hope is that it will obviate the reflect, give anything with a temperature- broad spectrum antibiotics, embed NEWS2 everywhere and ensure that clinical judgement is prioritised in the assessment process.


  4. Guest

    Very supportive of what sounds like an interesting project. In our practice we have been able to get screening rates to high levels but still struggle with the antibiotics in one hour metric as you do. At times, even in a small town like ours, suspected sepsis can be so frequently called in that there is "pre-alert fatigue" and some analysis of the HFs involved in the process will certainly add to the understanding of what the drivers are in sepsis care.

    1. Thanks Ed and Matt - hope this project can contribute to the national (and international) efforts to improve recognition and management of deteriorating patients and sepsis. Jo

    2. Guest

      worth remembering the new CQUIN change means that the time to antibiotics begins at the diagnosis of suspected sepsis in a hospital setting. Ambulance trusts will have different metrics to test their screening, speed of transfer.

      The hope is that it will obviate the reflect, give anything with a temperature- broad spectrum antibiotics, embed NEWS2 everywhere and ensure that clinical judgement is prioritised in the assessment process.


  5. Dear Jo,

    Great to talk to you on Friday and specifically really good to hear about this project. I have shared this with our Human Factors experts in the HSIB and one is involved in an investigation that is looking at the recognition of a sick adult in ED. Really good project to engage HF expertise.

    1. Thanks Nichola - great to see how this may be able to link with wider learning via HSIB. Jo

  6. In a demanding service the patient flow and human factors aspect is critical to timely intervention.  More work in these areas are crucial to improved patient outcome. Really interesting project.

    1. Thanks Katie for comment and support. Learning should hopefully be transferable to many projects/pathways. Jo

  7. Thanks Ian for your comment and support! Jo

  8. Hi,

    I think this would be a really interesting project as we should be considering HF influences when designing new pathways or investigating problems in pathways.

  9. Hi Ian. Thanks for your comment and suggestion re different models of care. I'll share this with the team to add into the planning.

    Many thanks Jo

  10. Good idea

    In terms of generalisability I would consider the impact of different models of ED care: for instance, if Rapid Assessment Pathways are in operation they should in theory be able to significantly impact on sepsis performance

  11. Guest

    Debbie Brazil 3 years, 4 months ago

    I should add that our work is more ward focused than ED, however some of the system issues we find may affect both settings - ED and wards

  12. Guest

    Debbie Brazil 3 years, 4 months ago

    We are initiating a collaborative within the L&D on identifying and addressing barriers to delivering antibiotics within one hour. We know they don't get prescribed and we can quantify what %, but we still don't know what the root causes are. That's the focus of our attention in the coming months before we even try and consider solutions. Would be very interested to collaborate and share learning both ways.

    1. Hi Debbie

      Thanks for your comment. Very happy to collaborate with you and share our learning both ways. Would love it if you could indicate your support for this project by clicking the "support" button at the top of the page! You can find my contact details on our web page. Thanks Jo

  13. @jomurray- Well done- A much needed analysis. Sepsis care lends itself beautifully to a human factors study, and Oxford is blessed with having some of the finest HF researchers. I am very excited to see where this project goes!

    Presumably SCAS use NEWS- will there be an opportunity to study the impact of this at acutes who use alternative EWS (if they persist with non NEWS)

    @Anne Maloney- completely agree. I hope future measurements nationally will look increasingly at end-to-end, home to hospital metrics. The minutes in hospital pale in comparison to the days at home.

    @stephanieross- LD groups are a critical area of focus, The NHS Med Director for Effectiveness is chairing a working group to get to the heart of this. Jo, please can you forward on the email I sent.


    1. Thanks for your support Matt. We will be sharing via our regional and national PSC work streams so you will be kept updated that way as well as through Q (if we are successful that is!)

      I've sent Stephanie the information about the 2nd July NHSE LD and sepsis meeting.

  14. Dear Colleagues

    I'll be on annual leave next week, so please do keep your comments coming and I will respond on my return.

    Many thanks


  15. Guest

    Stephanie Ross 3 years, 4 months ago

    Hi, Sepsis is a significant health issue for people with learning disabilities, this report gives some idea of issue with 11% of deaths of people with learning disabilities reported related to sepsis. We do have some local data but  I think we would need to do a little more work to establish figures.

    I am aware that it can take longer to get a history or carry out investigations for some people with learning disabilities as it may also be for people with Dementia for example. Has this been considered as an issue contributing to a delay in getting IV antiobiotics started?



    1. Guest

      Stephanie Ross 3 years, 4 months ago

      Thank you for following up and the information on the upcoming workshop. I will aim to ensure we have a local representation at the meeting if possible and can ensure we are addressing locally.

    2. Our clinical lead Dr Matt Inada-Kim let us know that the NHS England are holding a Sepsis and Learning Disabilities workshop on July 2nd where the focus will be on:

      To gain a shared understanding of the prevalence and specific issues people with Learning Disabilities have in relation to sepsis.
      To learn about examples of existing good practice that aims to improve outcomes for people with Learning Disabilities at risk of / who develop sepsis.
      To develop specific actions that can be implemented locally or nationally to improve outcomes and experience for people with Learning Disabilities at risk of / who develop sepsis.

      To make connections with existing programmes of work and ensure alignment in approach or desired outcomes where possible.

    3. I've also just flagged to our national deteriorating patient work stream via the 15 Patient Safety Collaboratives for awareness.

    4. Hi Stephanie

      Thanks for this helpful suggestion and link/report. I've fed back to the team to consider. Jo

  16. While I appreciate it is not the remit of this group to be thinking outside the secondary care boundaries, it is important to remember that sepsis starts in the community, not in the ED. This is where we could reduce risk of deterioration in the first place by early identification and admission before the patient starts to decompensate. 'Presentation to door' would add an interesting new dimension to the 'Time to Treatment' figures.
    Sepsis is a major cause of avoidable death. Early recognition and treatment of sepsis might save up to 10,000 lives in the UK each year, but survival falls by up to 7-8% for every hour delay in treatment.
    What is desperately needed is the development of tools for use by GPs and primary care teams.  Even routine use of a validated NEWS is not routine in GP – this could potentially identify patients who risk deteriorating from infection to sepsis.  Earlier admission would undoubtably reduce morbidity, mortality and costs. All patients seen in the hospital have a GP and we are often the first clinicians to see the patients - we should be part of the sepsis team, too.

    1. Hi Anne

      Thanks for your comments. As you note, this particular project is focusing on the ED pathway. We have a regional stakeholder group that has representatives from across the pathway, including primary care, care homes, ambulance service and other community settings. There are other projects (and alternative funding sources) that cover this part of the pathway. I will continue to keep you updated with these as I'm aware you have work related to deterioration training in primary care.

  17. Jo, Have you had a look at

    Regards Tom

    1. Hi Tom.

      Yes I have thanks and just shared with them the regional tool we have adopted across Wessex and Oxford PSCs. Paediatrics have specific challenges, not least that there is no national approach yet to match NEWS for adults.

      It has prompted me to confirm with our team if we are focusing on adults or including all ED attenders (paeds and maternity) so thanks for that!


Leave a comment

If you have a Q account please log in before posting your comment.

Read our comments policy before posting your comment.

This will not be publicly visible

Please note that you won't be able to edit or delete comments once posted.