Radiotherapy in Scotland
Pathways and Processes Group
Jan – Jul 2010
Report on capacity and utilisation of current stock of linear accelerators across Scotland with recommendations for addressing future demand
Final Report
26th August 2010
Contents
1.Background
2.Methods
Pathways and Processes Group Workplan
Workshop
Centre visits
Good practice from other UK centres
3.Findings
Linear Accelerator Utilisation
Number of fractions delivered per hour
Number of hours per year during which a linac is operational
a.Hours operational per day
b.Public holidays
c.Servicing
d.Commissioning and replacing linacs
e.Weekend working
Minimising the number of unused slots
a.Optimal patient selection for radiotherapy treatment
b.Patient transport
c.Machine breakdown
d.Process Mapping
e.Patient catchment populations
Additional linear accelerators
Location for additional linear accelerators
Shaping demand to ensure that the right patients are receiving the right treatment at the right time
Earlier stage treatment
Clinical variability
Patient selection
Service improvement
Demand and capacity monitoring
Process mapping
‘3SI’ Principles
Patient information
Patient involvement
Skill mix and new ways of working
Networking and leadership
4.Conclusion and Recommendations
Recommendations
Appendix 1.Summary of SRAG Workshop February 2010
Appendix 2.Radiotherapy Centre Visits
Appendix 3.Visit to Western Park Hospital, Sheffield
Appendix 4.Map of drive times to radiotherapy centres
Appendix 5.Satellite Centre Questionnaire
Appendix 6.Background reading articles and references
- Background
In 2006, the Radiotherapy Activity Planning for Scotland 2011-2015 report indicated that due to rising levels of cancer incidence there may be a significant increase in the capacity requirements for radiotherapy in Scotland over the next ten to fifteen years. This rise in demand will result from:-
- Increased incidence of cancer
- Potential increase in indications for radiotherapy in clinical practice
- Rapid treatment developments and increased complexity
In 2009, the Scottish Radiotherapy Advisory Group (SRAG) commissioned a report to test whether the predictions made in the 2006 report still held true and new estimates were calculated in the Report on re-modelling of radiotherapy demand and capacity data, May 2010(R. Chalmers, S. Erridge and C. Featherstone). The re-modelling work was based on Information Services Division (ISD) statistics updated in 2010, which projects cancer incidence in Scotland to 2020.
The updated report estimates that at least 237,200 fractions should be delivered across Scotland (or at least 263,600 if an additional 10% is factored in to prevent waiting times arising from fluctuations in demand, quality assurance and introduction of new techniques).
There have been other reports which have attempted to predict demand for radiotherapy including a Royal College of Radiologists report in 2002 which recommended 4 linacs per million population rising by 5% per year to take account of increasing incidence and complexity to 5.5 per million population by 2010.
In 2007 the National Radiotherapy Advisory Group (NRAG) in England made a recommendation that treatment capacity should deliver 40,000 fractions per million population by 2010/11 rising to 54,000 fractions per million population by 2016 (equivalent to 275,000 fractions in Scotland). Figure 1 overleaf shows that the NRAG prediction is only slightly above the minimum of the 2006 and 2010 predictions in the original and updated reports.
Current linear accelerator stock in Scotland is 4.8 per million population and in 2009 201,752 fractions were delivered. This level of treatment provision is considerably less than the minimum estimate quoted in the 2006 reportRadiotherapy Activity Planning for Scotland 2011-2015 (at least 242,400 fractions in 2015) and the estimate in May 2010 updated report referred to above. There is no clear explanation for this difference in the predicted demand and actual fractions given. It is noteworthy that future demand modelling in the Netherlands tends to be based more upon current demand estimates. It would be useful to explore the reasons why there is a difference between the fractions currently being delivered and the predicted demand. This work is likely to support the demand methodology and would also help to inform the work that may be required to ensure that the expected patient base is being captured, and to understand whether there are reversible factors that are likely to impact upon potential future demand. It is worth noting that there has been a year on year increase in systemic chemotherapy delivery recently.
Figure 1 – Chart showing radiotherapy future demand predictions against actual data reported from centres
Linear accelerators are expensive capital items and several years of planning are required before machines can be commissioned. Given the current economic challenges, it is imperative that new machines are purchased only once existing machines are demonstrated to be achieving maximum efficiency, taking into account the factors that influence useful machine life expectancy.
The Pathways and Processes Group was asked to explore possibilities for achieving maximum output from current linear accelerator stock to inform any future proposal for additional linear accelerators. This report identifies possibilities for providing additional capacity that could be applied pending a decision on additional linear accelerator purchasing and commissioning.
Thank you to all who contributed to the content and compilation of the report.
- Methods
In the autumn of 2009 the Scottish Radiotherapy Advisory Group (SRAG) was preparing to update theRadiotherapy Activity Planning for Scotland 2011-2015 report from 2006and around the same time the Cancer Performance Support Team (CPST) identified risks to delivery of the 31-day cancer access target where first treatment involved radiotherapy. A Pathways & Processes Group was commissioned by SRAG and tasked to explore quality improvement in radiotherapy provision in Scotland including productivity and efficiency and linear accelerator utilisation.
Pathways and Processes Group Workplan
With the work carried out on behalf of SRAG by Ruth Chalmers, Sara Erridge and Carrie Featherstone to predict future demand demonstrating the need for additional radiotherapy resource, it was recognised that meeting this need should take account of the capacity released through maximising efficiency and productivity. This was even more important given the economic pressures facing the NHS. Any business cases for additional liner accelerators would need to demonstrate that this had been addressed.
The group’s workplan included the following:
- Benchmark current capacity and efficiency by designing and distributing a data gathering template for completion by radiotherapy centres, arranging a programme of radiotherapy centre site visits to assess comparability of data and obtaining a progress report and comparable data from National Cancer Services Analysis Team (NATCANSAT) in England
- Benchmark practice in radiotherapy provision in other parts of UK. A group visit to Sheffield Cancer Centre was arranged
- Use site visits to benchmark current status of improvement methodology work, assess outputs from this work and identify areas for further work
- Investigate patient transport issues and impact on radiotherapy capacity, meeting with Scottish Ambulance to identify ways in which the impact of transport issues on scheduling could be addressed
- Provide opportunities for networking for radiotherapy staff
Workshop
SRAG agreed at its November 2009 meeting to offer staff working in radiotherapy centres across Scotland the opportunity to meet and to consider how the delivery of radiotherapy services could be shaped to meet demand. More than 50 centre staff attended a workshop in February 2010. A summary of this is attached as Appendix 1.
Presentations were made by a variety of speakers including from the Netherlands, NACANSAT and NRAG. Sarah Erridge and Carrie Featherstone also summarised their work on demand modelling and delegates were then split into small groups and rotated round stations to discuss key questions on organisation and delivery of radiotherapy services.
The outputs from the workshop questions informed the workplan of the Pathways and Processes Group.
Centre visits
In order to ensure as far as possible that the data on capacity and utilisation of linear accelerators was comparable across centres, it was suggested that members of the Pathways and Processes Group should visit each of the centres to confirm the data that had been submitted and to discuss with the clinical lead, management lead and others referral and treatment protocols, demand analysis, how factors affecting capacity were managed, current waiting times, patient involvement and team ethos.
The completed templates are included in Appendix 2.
Good practice from other UK centres
The Pathways and Processes group also wanted to compare practice in Scotland against what was happening at other UK centres. Tim Cooper, Associate Director,National Cancer Access Team in England, recommended a visit to WestonParkHospital in Sheffield and several group members attended a visit in June 2010.
Information on demographics, staffing levels and equipment was obtained. The management structure and systems were compared and innovative practice in oncological emergency handling and ‘3SI’ (Service Improvement, Staff Involvement and Sustained Investment) implementation was explored.
The group members were able to make comparisons with the processes being used and challenges faced by their own centres.
A more detailed report is attached in Appendix 3.
- Findings
There are a number ofkey considerations thatwill influence the ability to meet future needs and provide timely radiotherapy treatments:
Linear accelerator utilisation
Additional linear accelerators
Shaping demand to ensure that the right patients are receiving the right treatment at the right time
Other service improvement measures
Linear Accelerator Utilisation
Linear accelerator (linac) productivity could be maximised by more uniform use throughout the working day. The previous Radiotherapy Activity Planning for Scotland 2011-2015report recommended 90% utilisation of funded clinical hours. The remaining 10% (equivalent to one half day per week per machine) allows for unscheduled downtime,fluctuations in demand, quality assurance, clinical trials, service development and introduction of new techniques.
The following table summarises the current level of linac utilisation across the five centres in Scotland, based on data supplied by centres in a 2009 census.
Figure2–Table showing the 2009 level of linear accelerator utilisation per radiotherapy centre
Aberdeen / Dundee / Edinburgh / Glasgow / Inverness / Scotland2009 Census
Number of linacs / 3 / 3 / 6 / 11 / 2 / 25
Funded clinical linac hours / 18 / 22 / 48 / 85 / 10.5 / 183
Total Fractions delivered / 20,684 / 16,705 / 51,928 / 102,336 / 10,099 / 201,752
Total Fields delivered / 57,899 / 46,174 / 131,313 / 305,944 / 28,908 / 570,238
Total courses / 1,759 / 1,353 / 3,548 / 7,045 / 685 / 14,390
Hours worked per day / 7 / 7 / 8 / 9 / 7 / 8
Fractions booked per hour / 6 / 5 / 5.5 / 5 / 5 / 5
Days worked per year / 259 / 253 / 255 / 259 / 259 / 257
Population (million) / 0.58 / 0.52 / 1.39 / 2.34 / 0.34 / 5.17
Information derived from census data
Fractions delivered per linac / 6,895 / 5,568 / 8,655 / 9,303 / 5,050 / 8,070
Average fields per fraction / 2.80 / 2.76 / 2.53 / 2.99 / 2.86 / 2.83
Avg fractions per patient / 11.8 / 12.3 / 14.6 / 14.5 / 14.7 / 14.0
Avg fields per patient / 32.9 / 34.1 / 37.0 / 43.4 / 42.2 / 39.6
Linacs per million population / 5.2 / 5.8 / 4.3 / 4.7 / 6.0 / 4.8
Funded clinical linac hours per million population / 31.0 / 42.5 / 34.5 / 36.3 / 31.2 / 35.3
Fractions per million population* / 35,570 / 32,277 / 37,340 / 43,683 / 30,048 / 39,035
Utilisation based on funded clinical linac hours
Funded fractions** / 27,972 / 27,830 / 67,320 / 110,075 / 13,598 / 248,583
Utilisation / 74% / 60% / 77% / 93% / 74% / 81%
* NRAG estimated that by 2016 the NHS will need to deliver around 54000 fractions per million population
** delivering stated fractions booked per hour and stated working days per year.
Funded fractions = Funded clinical linac hours x Fractions booked per hour x Days worked per year
In order to deliver radiotherapy in a timely manner (which complies with the Royal College of Radiologists good practice guidelines and with the new national cancer access targets for first treatment) calculation of the efficiency of patient throughput is important to estimate current radiotherapy provision and potential capacity. However additional contributory factors include age of linacs and the availability and working patterns of staff.
Linac utilisation is dependent upon:-
- number of fractions delivered per hour
- number of hours per year that a linac is clinically operational
- number of unused treatment appointments
In the following pages suggestions for increasing utilisation of linac capacity are demonstrated. However, it must be acknowledged that patient safety and the need to maintain the highest quality standards of all aspects of care are paramount. Any measures to improve productivity of the current stock of linacs must be taken only following robust risk assessments.
Number of fractions delivered per hour
There is variability across the five Scottish centres in the number of fractions delivered per machine per hour. This variability reflects adequacy of staffing levels, complexity of radiotherapy set up, infirmity of patients, clinical interruptions and machine breakdowns. Machine breakdowns tend to increase with the age of the machine. Across Scotland the current mean number of fractions scheduled per hour is 5.
The number of fractions per hour depends on case mix, the complexity of treatment plans, the requirement for verification of the plan and the extent to which auto-sequencing has been implemented. These are rate limiting factors which make it unlikely that the number of fractions carried out per hour can be increased or even maintained at the current average level in Scotland, although there was no clear reason to explain why throughput per hour varied across the five centres.
There is variation in the mean number of fractions planned per hour in other parts of the UK and Europe such that no single standard has been set.
Efficient processes will maximise the number of fractions deliverable per hour. Process mapping has already been carried out in most centres but with changes in techniques and the introduction of additional pathway steps, it is recommended that this should be revisited to realise benefits.
Volumetric Modulated Arc Therapy (VMAT)is being piloted in one centre and it is hoped that its use will reduce the time taken to deliver radiotherapy to a wider number of fields per fraction. This technique is being used in some parts of Europe with some success and although the linac machines currently being procured will have VMAT capability,the cost and efficiency benefits should be explored before a commitment to purchase more VMATunits is supported.
The effect on total capacity across Scotland of changes in fractions delivered per linac is shown below for 25 machines working 8 hours per day, 259 days per year at 90% utilisation:-
Figure3–Table showing the effect of hours, linacs and fractions on all Scotland level capacity
Fractions planned per hour / Fractions deliverable4.5 / 209,790
5 / 233,100
5.5 / 256,410
Key Points
Regular updates of process maps are a useful way of ensuring that maximum throughput of patient per hour is being achieved.
An average of 5 fractions per hour is currently being achieved across Scotland but factors such as treatment complexity and the introduction of new equipment is likely to change this. A lower throughput figure may need to be used for future estimates of linac productivity.
Number of hours per year during which a linac is operational
a.Hours operational per day
Different centres currently operate different start and finish times.
It is acknowledged that adequate staffing levels would be required to make any extended working hours sustainable.
Longer working days may result in loss of flexibility when breakdowns occur. Also, when linac use is extended, there is a risk of shortening the lifetime of the machine.
The potential savings have to be balanced against the costs of staff overtime and there has not been any recent detailed work to estimate costs or compare benefits against these additional costs.
The introduction of Agendafor Change has contributed significantly to the difficulties in introducing extended working hours, although this has been partly taken forward in at least one centre through a local agreement. There is evidence of willingness of staff to be flexible but these instances have tended to be in emergency circumstances or short term situations. Although there is probably scope for extending working hours, albeit gradually or using a phased approach, a detailed feasibility study and cost benefit analysis would be useful.
The issues of training and attrition rates and the difficulties of staff recruitment and retention have affected all centres to varying degrees. There is a particular shortage of physicists and radiographers.
Just as unfilled staff vacancies affect the ability to extend the working hours of the current stock of linacs, so the same constraints will have an impact on the need to add to the existing complement of machines.
The staff complement and shift patterns at each centre and across Scotland need to be reviewed regularly to ensure that linac efficiency is optimised in the light of new technologies, new equipment and clinical protocols.
The effect on total capacity across Scotland of increasing the number of hours per day during which a linac is operational is shown below. The calculations are based on 25 machines working 259 days per year at a rate of 5 fractions per hour and at 90% utilisation.
Figure 4 – Table showing the effect of hours, linacs and fractions on all Scotland level capacity
Working hours per day / Fractions deliverable per hour7 / 203,963
8 / 233,100
9 / 262,238
10 / 291,375
Key Points
There are a number of complex issues involved introducing extended working days and a detailed feasibility study may be the best way of assessing all the benefits and risks. The alternative would be to introduce, incentivise and evaluate voluntary pilot schemes.
It is widely accepted that unfilled vacancies and skill mix shortages are a limiting factor in the expansion of radiotherapy services. Regular monitoring of staffing complement for all professional groups across Scotland will quantify the issues and trends.
A phased introduction of extended hours would provide interim additional capacity to allow time for a business case to be developed and approved and subsequent procurement and commissioning of any additional stock of machines that are required to meet the future demand for radiotherapy in Scotland.
b.Public holidays
Many centres have moved to working on public holidays but there is still some scope for small increases in the number of linac hours available across Scotland if all centres were to adopt this arrangement with machines closed only on two public holidays per year.