Trust / The Whittington Hospital, London
Background and initial problem / What are you trying to improve and why?
The Whittington Hospital ED department has 17 junior doctors who previously rotated through a set pattern of shifts and annual leave, each starting on a different week of a fixed 17-week rotation. However, doctors usually worked in the department for 26 weeks, so did not rotate equally through each week of the 17-week pattern. As a result, leave and shifts were unevenly distributed, and every cycle some doctors were not allocated sufficient annual leave, which required additional payments to be made at the end of the post, and was bad for morale. Although doctors were given some choice of which week of the rotation they started on, adjustments for study leave and personal commitments were not accommodated, and required doctors and the rota manager to arrange swaps. Managing study leave amongst these doctors was difficult, and often deemed prohibitively expensive to support whilst on nights, because 2 nights would need to be covered by a locum doctor to support a single day of study leave.
Academics at Cass Business School have dedicated decades of their careers to the Operational Research underpinning complex systems roster design, across many safety-critical industries including rail transport and nursing. Their unique software, Nightglass, for Junior Doctors’ rosters uses mathematically-sound algorithms to prioritise staff ergonomics (adhering to best practice from the literature in terms of fatigue minimisation so improving safety and efficiency at work) and work/life flexibility (e.g. annual and study leave requests, exams, personal commitments and less than full time trainees) within the roster, whilst maintaining safe service cover and meeting the legal requirements of the European Working Time Directive, New Deal and the 2016 Junior Doctors’ Contract. In this way, individualised rosters are produced for each doctor which meet their particular needs, protect patients and doctors through fatigue reduction, ensures fair total hours distribution across the duration of their placement and fits work around life.
The ability of this software to design doctors’ rosters to directly minimise fatigue and protect vital training is therefore unique, and reassures Trusts and doctors that they are providing safe, high quality care 24 hours a day. This case report describes the collaboration between Cass Academics (the Nightglass team) and the Whittington Hospital ED team, who consider this system to represent the Gold Standard in junior doctor rostering.
The relevance of this case study to Junior Doctors’ morale comes from the NHS Constitution’s 3rd Principle: ‘Respect, dignity, compassion and care should be at the core of how patients and staff are treated - not only because that is the right thing to do - but because patient safety, experience and outcomes are all improved when staff are valued, empowered and supported’.
Health Education England’s 2016 listening exercise on improving working conditions for Junior Doctors identified opportunities that fall broadly into three themes:
1. Being supported: for training, service provision and individual career aspirations
2. Feeling valued: by clinical and non-clinical staff
3. Having autonomy: through greater involvement in decisions relating to their working and personal lives.
Health Education England collected this case example and we believe it addresses themes 1, 2 & 3 above.
Specific aims and objectives / What are you going to measure?
Roster quality measures for individual doctors
1. Junior doctor overall satisfaction with roster vs rota and resultant improvement in morale
2. Reported fatigue
3. Satisfaction of doctors’ individual annual and study leave requests
4. Reported reduction in swaps
5. Equitability in the distribution of total and anti-social hours
6. Protection of planned training opportunities and exam preparation
Organisational Measures
7. Legal compliance: EWTR, New Deal & New Junior Doctor Contract
8. Financial savings to the department
9. Administrative time savings to the department
10. Patient safety
Process / What did you do? How did you do it? How were junior doctors involved? How did you measure it?
Initial Discussions
· Develop a mutual understanding of local rota challenges and departmental context, then survey junior doctors to understand their concerns and priorities.
Requirements Gathering
· Capture requirements for shifts needed each day of the week from existing rota, and specific information about different types of shifts
· Capture working practices of individual duties not apparent from the rota, e.g. staff seniority requirements and subspecialist training aspects
· Develop a framework for fairness criteria which are important to the doctors, e.g. fair distribution of antisocial shifts.
· Identify cohort of doctors including individual contractual arrangements, e.g. part-time working
· Collect junior doctors’ requests for annual leave, study leave and days off, graded by level of priority set by the individual, entered through a web interface
· Clinical Lead decides which study leave requests to authorise, considering conflicts, if any, and high priority leave, e.g. weddings
Roster production
· Nightglass team assemble users’ data and quality requirements for the e-Rostering software.
· The e-Rostering software produces three versions of the roster: individual rosters for each doctor, weekly rosters for the rota coordinator, and a master roster to allow an overview in case of future swaps.
Challenges / What challenges did you encounter and how did you overcome these?
E.g. Financial/HR/junior doctors/management/other stakeholders
Technical
· Introducing a new system, establishing working relationships between stakeholders and contacting doctors with enough notice to process leave requests was an initial challenge
· The short timescale provided a challenge to making a case for financial support for project; however, savings far in excess of the cost of the Nightglass service were achieved
· The spread and variation in grades of Doctors results in a complex ED rota, but the software handled this well
· There were no difficulties during the roster period in operating the new roster system
HR
· There was initial reluctance of HR to accept our interpretation of Working Time Regulations where this differed from existing rostering software. This issue was resolved in consultation with an NHS rostering expert.
· Despite carefully prioritised and controlled working parameters, fatigue indices for doctors in the department remained higher than desired, because of the intense nature of emergency medicine. Indices can be reduced by adopting ergonomic practices including reducing shift length, reducing number of consecutive nights, increasing rest periods after nights, as well as employing more doctors.
Outcome, impact and learning / What has the result been – on an individual level for junior doctors and other staff? / on an organisational level?
Overall, the ergonomic rostering system has significantly improved the working experience for the doctors on it. It directly addresses the themes expressed by Junior Doctors in the HEE listening exercise: supporting safer service provision and fair training opportunities, improving autonomy over the working schedule, and valuing career and personal commitments by prospectively incorporating these into their individual rosters. The specific improvements for each measure are further detailed below.
Roster quality improvements for individual doctors
1. Junior doctor overall satisfaction with roster vs rota and resultant improvement in morale
All doctors reported improved morale working on this roster. This was achieved both through the consideration of doctors’ priorities and the other specified measures of roster quality in the design of the roster, which cannot be accounted for by traditional patterns of a fixed rota.
2. Reported fatigue
All junior doctors reported reduced fatigue on the roster compared to on a rota. This has an obvious and direct relationship to patient safety[1],[2].
3. Satisfaction of doctors’ individual annual and study leave requests
Individual doctor’s needs varied greatly, and Nightglass encouraged them to make as many requests as they needed, and to provide as much depth of information about these as possible. Each request for annual leave was assigned a priority by the doctor. Every doctor on the roster was guaranteed their first priority annual and study leave requests. The software was able to handle a variety of specifics including protected weekends either side of requested weeks, and requests to work only during the daytime on a particular date. Doctors reported that this significantly reduced the stress of waiting for leave approval or fixed date allocation, and directly contributed to improved morale.
4. Reported reduction in swaps
By guaranteeing annual and study leave requests, there was a dramatic reduction in the need for swaps within the roster. This was reported as one of the most valuable positive factors in improving doctors’ morale.
5. Equitability in the distribution of total and anti-social hours
The roster was carefully controlled for fairness, ensuring trainees were allocated a very similar number of total and anti-social hours. This removed the unfairness of some doctors working more antisocial shifts than others dependent on indiscriminate allocation of which line of the rota they commenced their placement on.
6. Protection of planned training opportunities and exam preparation
Trainees were guaranteed their in-house training opportunities by them being written into the roster and covered by other doctors who had their time compensated for elsewhere. Trainees taking professional exams were guaranteed 2 no night working for minimum 48 hours beforehand. For Paediatric ED training, the software fairly distributed shifts to ensure doctors gained this valuable educational experience in post. This facility also allowed training opportunities to be monitored, providing accountability to educational supervisors and HEE whose placement fee obliges protected training.
Organisational improvements
7. Legal compliance: EWTR, New Deal & New Junior Doctor Contract
After guaranteeing the ergonomic and quality-related factors described above, the software was able to design rosters which were fully compliant with the legal requirements of the EWTR, the New Deal and the 2016 Junior Doctors’ Contract. This removed the risk of fines and additional banding supplements. Moreover, the New Contract obliges that Doctors should now be paid for the actual hours worked. Equitable distribution of work between doctors is guaranteed by the roster, whereas this is not the case for doctors working on different rows of a traditional rota.
8. Financial savings to the department
Some of the more junior doctors rotate on a 4 month pattern, whilst others rotate on a 6 month pattern. The e-rostering software was able to make the most out of this overlap, ensuring those not rotating were rostered to work during induction, thus improving cover and reducing locum requirements during changeover. Reducing locum requirements during the December 2016 changeover and induction period, when 5 doctors rotated, meant that the department saved the cost of 8 locum shifts, the equivalent of £6000.
Planning shifts around study leave, some of which is part of a mandatory training programme, meant that the department saved the cost of providing locum cover for two night shifts rather than one day shift, saving £900 each time. Currently there are around 30 of these kind of study leave requests in 6 months, and it is reasonable to assume that around 5 of these would have been during nights on a rota, a total saving of £4500.
Further savings were made from not having to pay out doctors for annual leave they had been unable to take at the end of their posts. This is a practice that would be unacceptable with the new contract.
9. Administrative time savings to the department
The senior clinical staff member previously responsible for managing swaps on the rota reported, at a conservative estimate, a saving of 3 full working days’ time per month, which was able to be put to better use.
10. Patient safety
This rostering system can improve patient safety by objectively and subjectively reducing the fatigue of doctors on it1,2, and through improving morale[3] by the mechanisms described above. Furthermore, although registrars were rostered separately, the software was able to ensure that at least one more experienced junior doctor was rostered during the vulnerable nightshift periods. Having more seniority around when there are fewer doctors on the floor improved the safety of the nightshift.
Next-steps & sustainability / What are the next steps (if any)? How measures are in place to ensure that these changes are sustainable?
The new ergonomic rostering system has proved invaluable for satisfying the tighter New Contract conditions, and has been embedded into the Emergency Department’s working practice where all relevant staff are able to operate the software independently. It works smoothly and continues to receive highly positive feedback from senior staff and the Junior Doctors themselves. It is expected to minimise exception reporting, because the roster is less subject to swaps, includes planned study leave and becomes a transparent record of worked hours and training.
The Nightglass team and their clinical collaborators are now in a strong position to further challenge the current NHS culture surrounding shift working, such as length of duties and long stretches of consecutive days and nights.
The next step is to roll out to the rest of the Whittington Hospital, and/or to other Emergency Departments within the NHS. The spin-out company Nightglass Medical Rostering Ltd. is set up for this purpose, with features for multiple grades and wards, and other specialties.
Future initiatives, dependent upon funding, include four parallel overlapping projects:
· Establishing a Gold Standard for ergonomic rostering in the NHS
· Improving the existing user interface for Lead Clinicians, rota Managers and Junior
Doctors at the Whittington Hospital
· Further pilot studies specialising in Anaesthetics
· Extending the range of users to key staff such as the Guardian of Safe Working and
Educational Supervisors
Further information / Who can be contacted for further information?
Professor Celia Glass, Professor of Management Science, Cass Business School.
Dr Duncan Carmichael, Consultant in Emergency Medicine, Whittington Health.
[1] The Mid Staffordshire NHS Foundation Trust Public Inquiry. Report of the Mid Staffordshire NHS Foundation Trust Public Inquiry Volume 3: Present and future Annexes. p1404. February 2013.
[2] Health and Safety Executive. Health and safety in the health and social care sector in Great Britain, 2014/15. http://www.hse.gov.uk/statistics/industry/healthservices/index.htm
[3] West, M; Dawson, J. Employee engagement and NHS performance. The King’s Fund. 2012.