Ed – it seems like everything is about Covid-19 at the moment and that’s understandable, but we should not forget other aspects of our work. Whilst patient flow may seem less important as we deal with Covid-19 I would argue very differently. It’s absolutely essential that we have good systems at a time of increased EM resource requirements. Although this blog and paper were written before Covid-19, there are important lessons to be learned here that are very relevant to the coming weeks.
Co-authored with Caroline Boulind.
Hospital flow has never been more relevant than now. With the Corona virus potentially depleting space, staff and resources in a hospital near you, it may be worth considering the design science described in this paper as a solution to improving flow.
I’ll be honest with you, this is not a paper you can read in one sitting. It is published in the Journal of Operations Management1, which clearly does not have the 3000 word cut off we are used to for clinical papers. Then again, this is not a clinical paper, but an operational management paper. It describes how design science was used to turn around something most currently consider impossible – hospital flow – within a relatively short time frame (months). Caroline and I thought that was worth a few sittings to deliver this JC.
I have a conflict of interest, in that I am currently part of the team that performed this original work. I was however, only vaguely aware of the existence of Yeovil District Hospital when this work was done. You can read the abstract of the paper 1 below, but as we always say PLEASE read the full paper yourself and make up your own mind before changing practice.
What kind of paper is this?
This is a design science research paper. It is a bit of a break from the type of research we are used to reading. Design science research is a type of qualitative research. It aims to solve real, complex field problems (such as access block in one hospital) through real-time development of interventions, that can then be used in similar and related contexts (such as access block at other hospitals). Design science research bridges the gap between theory and practice by recognising the relationships between context, intervention, mechanisms and outcomes.
This particular design science research required interventions and mechanisms of implementation, that acknowledged the social and technical complexities inherent in a hospital and the interrelated healthcare professions that inhabit a hospital.
Tell me about the sample.
There was no specific patient cohort in this study. Rather, the research aimed to develop interventions that would improve the experience of all patients on their journey through a district general hospital, by targeting improvements in patient flow.
The case hospital was Yeovil District Hospital, a small district general hospital with approximately 2,000 staff, 350 beds and an annual budget of £120 million. At the start of the project in March 2016, performance of the hospital and Emergency Department (ED) against national targets put it in the lower quartile; only 88% ED patients were seen and discharged within the 4 hour target, and cancellation of elective surgery was high, with an annualised loss of income of approximately £2.1 million.
What did they do?
Although it is easy to get distracted by the eventual interventions devised, it is worth noting the process by which these came to be. The theory behind this is described in great detail in the paper. It is also notable that the lead researcher was the director of operations at the case hospital. But before anyone starts getting excited about bias, bear in mind that qualitative research thrives on bias. It makes sense that a senior-driven intervention would be more successful when real, complex field problems (such as access block) need solving.
This design science research was not the first attempt to address poor flow at the hospital. A prior project using management ideas such as Lean or Theory of Constraints failed; largely as these attempts were met with scepticism and resistance from clinical stakeholders. The authors linked the failed project to a disconnect between clinical and non-clinical stakeholders – specifically where attempts were made to find common ground between the perceived goals for patient care and how to achieve it.
The design science research ran from March 2016 to September 2017. The first intervention of the research was to connect and engage healthcare professionals in a shared goal of improving patient experience. A core team was assembled from senior managers, nurses, physicians and allied professionals; identified for their ability to influence and promote change, rather than for their position of authority. This approach established a dynamic team who could identify goals and interventions based on the shared social goal of improving patient experience and influencing change.
The second intervention was for the team to create a bundle of inter-related, outcome-specific routines that promoted swift and even patient flow through the hospital – from ED (or elective admission) to discharge. This was done through an iterative process over several months. This allowed the goals and processes that underpinned the routines to become clear and unambiguous; maximising consistency in performance of the steps involved in each routine. The project became locally known as the patient flow project.
Identification and implementation of the routines, as well as maintenance of open communication and successful interdisciplinary relationships were led by the team and facilitated through formal weekly meetings and informal daily huddles. This workflow allowed adjustments and improvements to routines to be calibrated in a rapid fashion.
The output of the project was six outcome-oriented routines that were shared across the whole organisation. Each routine was associated with a day of the week to allow staff to become familiar with the routines (although the routines in fact continued every day). Staff members of all seniority and professional backgrounds were encouraged to take ownership of the relevant interventions aligned with routines in their areas, rather than a focus on a top down management-led approach.
The six outcome-oriented routines were as follow:
- Morning Monday – Monday mornings were identified as a substantial bottleneck for patient flow: ED crowding and hospital bed occupancy was high after the weekend, and multiple elective patients started arriving despite low discharges over the weekend. The focus of Morning Monday was for each ward to discharge at least one patient prior to midday to facilitate flow from ED and minimise elective surgery cancellations.
- Next Step Tuesday – focus on those patients who were ready for, but hadn’t yet received, the next step in their care. This might be further imaging, speciality review or discharge to another care facility. Next Step Tuesday facilitated active identification of solutions to delays, rather than waiting around for the next thing to happen.
- 14-day Wednesday – focus on patients who had been in hospital for 14 days or more (often elderly patients). Some of these patients will be waiting for discharge (to a social care setting), for a package of care or sometimes next steps in treatment or investigation. Later-on 14-day Wednesday expanded to the ambulatory emergency care (AEC) service, allowing some patients to have a facilitated discharge with some investigations and treatment continued from home. Of note is that 14-day Wednesday acts as a bridge between Next Step Tuesday and the next routine, DTOC Thursday (delayed transfer of care).
- DTOC Thursday – traditionally DTOC patients (those awaiting transfer of care to social care settings or packages of care at home) were reviewed on a Thursday. Previously this involved a meeting between the patient flow manager and social care manager. This was replaced with a meeting between the director of operations, director of social care, social care manager, discharge team manager and community services lead. This allowed rapid internal escalation of DTOC patients to the director of operations and director of social care.
- 30 bed Friday – Recognition of the challenges with facilitating discharge over the weekend, and a shortfall of 24 discharges relative to admissions, led to a goal to have 30 empty beds by the end of the day on Friday. This allowed the shortfall in discharges over the weekend to be absorbed by a reduced bed occupancy on Friday evening, thus reducing ED crowding.
- Weekend Flow – A redesign of staff schedules allowed a new discharge hit squad to be developed. This consisted of one senior and one junior doctor who joined the weekend huddle with the site manager to discuss a plan for the day. They targeted patients who had been identified by weekday teams as potentially suitable for weekend discharge pending review. The hit squad completed necessary prescriptions and discharge paperwork to remove this load from the medical team receiving the acute take.
What were the main results?
Improvements were rapidly reflected in a number of areas following introduction of the routines. Improvements included reduced elective cancellations, length of hospital stay, delayed transfer of care, staff vacancy rate, and increased ED 4 hour performance and admission avoidance through AEC.
Cancelation of elective surgical procedures (due to lack of bed availability) was very significantly reduced: from nearly 80 per month in early 2017 to fewer than 10 per month by the end of 2017. This resulted in savings of approximately £1.75 million for the year.
Length of stay was reduced over the same period from 5-6 days to approximately 4 days. Over the life of the project there was a sustained reduction in non-elective length of stay of 14%, making beds available for elective admissions and facilitating flow of patients from ED into the hospital.
Changes in the routines for patients experiencing DTOC allowed Yeovil District Hospital to achieve amongst the lowest levels of social care related delay in the region. All of these changes across the whole hospital system had an impact on performance in the emergency department, freeing up beds in the hospital and facilitating patient flow.
The result was that Yeovil was one of only a small number of hospitals across the UK to achieve the 4 hour target in 2017. At the start of the project, only 88% of patients were seen and treated/ discharged within the 4 hour target. At the end of 2017 the target was achieved for 96.9% of patients, and this improvement has been sustained in the period since the end of the project.
At the same time, there was a significant expansion of AEC services. This resulted in more than 350 patients seen per month by December 2017, as opposed to around 60 at the beginning of the project. Avoided admissions increased from around 50 per month at the start of the project to more than 150 by the end of 2017.
These changes, and the new routines giving ownership of the changes to members of staff working on the various clinical shop floors, translated to an increase in morale and a better working environment. As a result, staff satisfaction increased (reflected in staff satisfaction surveys), and a reduction in staff vacancies for both physicians (from 57% to 0% vacancies) and nursing staff (23% to 4%).
What does this mean for our patients in Virchester?
This project demonstrates that improvements are possible – even in the current challenging environment with increased demand on all services, especially the ED. There is a lot of pressure on hospitals to make savings, whilst seeing more patients as well as maintaining standards. Problems related to access block have been perceived as insurmountable. Previous work in this area has often focused on single departments in hospitals or individual problems.
This research considered a whole system approach using design science research. The routines that were developed in Yeovil may be generalisable to other hospital environments with some changes relevant to individual contexts. More importantly, every hospital can create their own routines through design science research principles. This provides hope that the situation in Virchester and other hospitals might also be improved.
The bottom line
Organisation-wide approaches to patient flow bottlenecks are possible even in constrained environments, and can result in rapid, significant improvements in patient flow as well as financial savings.
- 1.Johnson M, Burgess N, Sethi S. Temporal pacing of outcomes for improving patient flow: Design science research in a National Health Service hospital. Jrnl of Ops Management. December 2019:35-53. doi:10.1002/joom.1077