In 2020, 37,211 people were diagnosed with lung cancer in England. 68% of this population is at an advanced stage and has limited life expectancy. People with advanced lung cancer have complex care needs and often experience high levels of GP appointments, hospital admissions and extended lengths of stay while awaiting diagnosis and treatment.
Timely genomic testing of the tumour can help identify individualised treatments, which can potentially markedly improve the quality and length of life. Delays in identifying specific gene mutations can result in missed opportunities for patients to receive targeted and more effective treatments, ultimately leading to worse outcomes and higher costs for health systems.
Liquid biopsy, a cutting-edge diagnostic method validated through numerous clinical trials, involves testing blood samples for biomarkers like circulating tumour DNA (ctDNA), among others, to detect cancer-related genetic mutations. This less invasive approach offers several benefits, particularly in vulnerable patients with advanced non-small cell lung cancer.
The NHS in England is working towards being a global leader in adopting liquid biopsy testing into a national health service. Recognising the importance of economic assessment and evaluation of the costs and benefits of broader ctDNA testing, Edge Health was commissioned by NHS England to undertake this work to support an ongoing national pilot involving non-small cell lung cancer testing.
Using health economics to understand benefits and costs
Our initial findings in the early phase of the health economics analysis of ctDNA testing combined academic methods with commercial insight and experienced understanding of how the NHS operates to assess the economic implications. This involved collaboration with clinical experts and synthesis of information from various other sources. As a new technology, our analysis considered various clinical scenarios and sensitivities for critical assumptions.
“Implementing ctDNA testing into the routine diagnostic work up of patients with lung cancer is a huge step forward to improving equity of access to state of the art genomic testing for our patients. This will allow patients to receive the best treatment possible for their condition. The input from Edge Health has been invaluable in mapping out a complex pathway, identifying options for ctDNA implementation and their associated cost benefits”.
Professor Sanjay Popat, Consultant Thoracic Medical Oncologist, Royal Marsden Hospital
Outputs from the initial analysis were extrapolated more generally with national data, which helped identify the potential future costs and benefits.
In the context of stage III and IV lung cancer, from early analysis, the application of ctDNA was found to deliver significant benefits relative to its costs. This finding was primarily driven by ctDNA testing enabling earlier blood testing and potentially avoiding tumour genomic testing, which supported patients to access targeted treatments earlier and more consistently – lowering broader system costs. In the next phase of work, pilot data will be analysed to validate these preliminary findings to support the commissioning of the ctDNA test on the genomic national test directory.
Moreover, ctDNA testing is expected to improve equity in genomic testing access substantially, expanding coverage over a broader spectrum of gene mutations and ensuring the inclusion of patients for whom adequate tissue biopsies might not be viable.
Ultimately, incorporating the latest genomics advances into routine healthcare will help deliver the UK government’s vision in “Genome UK: the future of healthcare”.
“The current work of the ctDNA pilot aligns perfectly with the Genomic Medicine Service goals of delivering equitable genomic testing for cancer patients through accessing cutting edge technology and science. This technology will hopefully, if commissioned onto the national test directory, ensure that clinical services can make better-informed decisions faster, have access to precision treatments which will improve patient outcomes, ultimately leading to more efficient use of NHS resources. The work from Edge Health is vital in helping to demonstrate that this advance in care is also economically viable”.
Paul Ryves, Programme Director, North Thames Genomic Medicine Service Alliance.
Contact us to learn more about our approach and how we can help you.
Developing a System Intelligence Specification in South East London
There are unprecedented challenges facing health, care and communities, and a need to change how we work in response. There is huge potential in the data and capabilities already available within systems, and opportunities to address growing activity, financial and workforce pressures by making best-use of limited resources.
It is critical that systems have a shared vision and understanding for how the effective use of business intelligence and analytics can improve health outcomes
This vision must encompass improving outcomes in population health and healthcare, tackling inequalities in outcomes, experience and access, and enhancing productivity and value-for-money.
The System Intelligence Specification for SEL ICS was co-developed with a wide range of leaders and stakeholders across the system
This included individuals from the SEL Integrated Care Board, Place, local Councils, acute, mental health, primary, community and social care providers, and other clinical networks and partners leading innovation in the system. Edge Health partnered with Public Private Ltd (PPL) to gather experience and insight on current requirements on business intelligence, articulating the future for analytics, and identifying the gaps.
The System Intelligence Specification articulated what South East London will be able to deliver through the better use of information and data, supported by key use case examples
The specification aimed to be ambitious but pragmatic, focussing on key value-add use cases and real-life examples to describe what collaborative working principles, skills and data are required to support enhanced working at all system levels.
This led to short and long-term recommendations to the SEL Board, and plans to deliver on these strategic commitments
The work and engagement provided collective clarity and a focus on opportunities that should be developed as a system. It provided a starting point for workshops that were held to formalise a digital and data enablement group in SEL, bringing together system leaders to turn data into insight into action, and impact individual patient outcomes.
“How many beds?”: Enhancing planning with bed requirement simulations
More than any other, one question we repeatedly get asked is “how many beds?”. Whether planning, improving patient flow, speeding-up discharge, producing business cases or responding to Covid-19, bedded capacity is frequently the biggest factor affecting both efficiency and outcomes. Robustly analysing patient flow and bed utilisation can drive effective decision-making and can facilitate future planning.
Edge Health has been engaged in helping multiple NHS Trusts to forecast their bed requirement and analyse the drivers of bedded pressure to support planning.
Mitigating risks and optimising bed planning
While monitoring bed requirements relative to the Trusts’ capacity is crucial, it is equally important to understand the underlying factors driving the demand for beds in a hospital. Taking this into account, we identify primary drivers of bed demand using detailed patient and ward level data to simulate bed requirements in an NHS Trust. These scenarios have been instrumental in aiding the planning of bed requirements in the Trust as they provide useful insights for effectively managing the Trust’s capacity.
It is very important to recognise the significance of understanding the variation of bed occupancy for the efficient calculation of the monthly bed requirements of a Trust. Hospital bed occupancy varies by minute and hour or each day – particularly for wards with short length of stay and quick turnaround times. Due to this variation, simply calculating the average number of occupied beds at the end of each day can misrepresent average occupancy levels and be unreconcilable with the experience of ward staff.
Variation in hourly bed occupancy over time
Our modelling approach
Detailed patient level data forms the basis of our analysis. To further understand the Trust’s capacity, we’ve also engaged with the staff to dive deeper in the wards structure and capacity, as well as any potential future capacity changes (including opening/closing of wards).
Utilising the available data, the key steps in our methodology approach are:
Calculating total number of occupied beds using hour-by-hour patient level data, accounting for the variation in hourly data.
Using the detailed ward allocations, aggregating the data on a division/ward/specialty level.
Exploring requirements for achieving different levels of target occupancy.
Creating multiple simulation scenarios to explore critical drivers of demand:
Variation and impact of length of stay.
Impact of ‘medically safe for transfer (MSFT)’ patients on capacity.
Effects from changing numbers of elective and non-elective admissions.
Influence of interventions like virtual wards.
Employing a flexible modelling approach enables us to gain valuable insights and understand crucial aspects related to bed planning in a Trust. The main advantages of our approach include:
Capturing the relationship between key variables in the analysis and measuring their influence.
Quantifying the impact of areas of uncertainty in bed planning.
Leveraging the potential outcomes of interventions to develop effective strategies and optimise resource allocation.
This modelling approach serves as a powerful tool in aiding bed planning and driving critical strategic decisions within the Trusts.
See more examples of our work on bed modelling here:
Imaging Productivity: Harnessing RIS data to meet reporting targets
The focus for efficiency gains has traditionally been on theatres, inpatient and outpatient activity and more recently the growing elective backlog. Diagnostics, however, have acquired a new emphasis since COVID. This is because significant backlog in diagnostics is causing delays in finding cancers (you can read more about this on our blog).
As a response, NHSE has included diagnostics activity targets in the NHS Constitution, stating that all tests must be performed within 6 weeks from request. This puts pressure on trusts to understand their testing activity, capacity and bottlenecks in imaging reporting that cause downstream delays in the 18-week referral to treatment (RTT) target.
Trusts have not historically analysed diagnostics data in depth
Gathering these insights requires analysis and data flows which are not yet set-up well across Trusts. Data from radiology information systems (RIS) has not been used as extensively in the past and existed in lower quality than, for example, theatre data. In this context, Trusts now struggle making sense of their imaging data to action NHSE targets.
We were recently asked to support a large specialist NHS Trust in helping with this.
The ask: time-sensitive solutions to guide management
We did this in three stages:
Analyse national data to get a high-level picture.
Engage with team on the ground to understand what actionable insights are needed where and why they are not provided.
Deploying expert clinicians and analysts to provide the insight in a repeatable way.
We outline each of the below in turn and give some detail on the issues under the hood.
First: Public data
First, we explored publicly available data on the Trust’s diagnostics to form the start of meaningful conversation and gain a high-level understanding of some of their challenges. The two charts below showed that our client (Comparator 1 in the chart on the left) had above average waiting times for a scan to be reported, following testing. The time between testing and reporting had also seen a significant increase in 2022/23 compared to 2019/20, particularly for MRI, Nuclear Medicine and Single Photon Emission CT.
Second: Working with team on the ground to understand the problems
We then sat with stakeholders to map the imaging data journey at the Trust and uncovered the key issues within it (outlined below).
Issues we found were the below:
Complicated set-up: Before an end-user could reach any insight, four pieces of software needed to interact: a requesting software like ICE, a RIS software, a PACS software and finally a BI software, QlikView in the client’s case. All insights needed to wait till the analytics team had curated the data.
No quick way around it: Managers trying to extract data from RIS directly were faced with a complex interface that was both hard to work with and at risk of producing unreliable metrics. This step was incredibly time-consuming for managers, adding to undue stress.
Lack of resources: There were no dedicated imaging analysts at the Trust, which meant the imaging team had to compete with other teams to get the insights they needed from the busy BI department.
We supported the trust outlining a variety of solutions, including:
Increasing workforce, such as a specialist PACS or analytics team member.
Changing work practices, such as setting a cap on highly time-consuming MDT requests to focus on internal workload, and upskilling radiographers to report imaging.
Upgrading software to one that included basic analytics, timed with a contract soon to expire.
In order to tie the department over with an urgent need for insights, we also delivered a fast turn-around reporting solution ourselves, embedded in their current BI environment.
Third: Delivering a trusted solution
We were onboarded on the Trust’s system to work within their environment and built a relationship with the BI and analytics team to ensure seamless knowledge transfer and accuracy of outputs.
Engaging with stakeholders also helped build trust in the outputs and ensure that it was truly useful to the team and met expectations.
Early draft of one of our reporting interfaces showing high-detail overview of reporting activity by radiologist and imaging modality, and reporting waiting times at a glance.
By the time the tool was ready to be shared Trust-wide, it had received the seal of approval from the imaging manager, the clinical director, and the BI lead. The final product was fully handed over to the Trust’s BI team to use as a starting point for more analysis and maintain as required in the future.
Specifically, it provided a high-detail overview of reporting activity by radiologist and imaging modality, and reporting waiting times at a glance.
This is enabling the Trust to:
understand pressures on the imaging department
manage workflows more effectively
back department investments and business cases with evidence
improve their performance against NHS diagnostic targets.
Elevating Performance and Driving Clinical Excellence: the Discharge Pathways Model Analytical Tool
As the NHS grapples with ever-growing demand for secondary care from the elective backlog and aging population, establishing an efficient patient flow out of secondary care is fundamental for alleviating system pressures.
Minimising acute capacity required for delayed discharges and ensuring patients have the right level of support available to them upon discharge, at home or in a community bed, will not only reduce costs for the NHS and free up the resources for those who need them most, but also help achieve better outcomes for patients.
The importance of access to accurate and up-to-date data in trying to accomplish this is paramount.
Answer the right questions
Assessing the level of demand, identifying system bottlenecks, and linking local practices with patient outcomes are all crucial undertakings on the path to better managed care.
The important questions like “How many patients are expected to be discharged into the community?”, “What causes the majority of delayed discharges” or “What can I learn from my peers who are performing better?” need to be carefully considered and answered on both system and national levels.
Edge Health has been commissioned by NHSEI to support them in tackling the complexities of the interface between secondary and community care and improving the patient flow.
Working with a multi-stakeholder group, including academics, clinicians and system leads, and in collaboration with the national rehabilitation team, we have developed a Discharge Pathways Model Analytical Tool, hosted on the NHS Foundry Platform, that facilitates access to the most up-to-date data that can be used to address those questions.
Benchmarking interface, part of the Discharge Pathways Model Analytical Tool
A one-stop place to benchmark, learn, and plan
The tool allows systems to benchmark their performance against their peers to identify areas of success and avenues for improvement.
It opens up a conversation around what can the systems learn from each other’s experience and provides crucial insights to the national team, informing a larger clinical effort to develop best-practice for care in the community.
Scenario modelling capabilities for effective bed planning
The modelling element of the tool allows the users to explore different models of care that can be implemented and thus support the planning activities for this winter and beyond.
The tool is now live and accessible to users across all the 42 systems, as well as the regional and national users.
Why We Need to Start Unlocking Change Using Cancer Data Now
We can all agree that cancer does not go on holiday. Yet during the winter months, referrals for suspected skin cancer in England decline steeply.
The National Health Service (NHS) collects extensive data that presents a golden opportunity for improvement and targeted interventions. For instance, by examining the data on skin cancer referrals, we can identify a long-standing pattern that presents a significant opportunity and exemplifies the potential of routinely collected data to enhance outcomes.
In the case of skin cancer, this opportunity translates to 170 livesa year that may have been saved through earlier cancer detection.
Skin cancer referrals: unlocking opportunities through a 10-year trend
The NHS Digital Cancer Waiting Times records provide valuable insights into skin cancer referrals. A simple glance at the time trend reveals a stark cyclical pattern.
The difference between the lowest point in referrals in January and the peak in August is astonishing, with August witnessing a 60% increase compared to January.
If we were to expect referral volumes to be relatively constant across the year, 36,000 patients might have presented sooner than they did in 2022. Given the skin cancer conversion rate of 8%, that’s nearly 3,000 potentially delayed diagnoses.
Moreover, the number of 2WW referrals aligns rather consistently with the number of patients awaiting cancer treatment. Therefore, fewer referrals are a result of fewer cancers presenting and being detected, rather than an artificial summer spike in referrals.
It is not news either – the pattern was present in 2016 and is just as prevalent today as it was then. This is an area that would benefit from targeted intervention.
A Chance to Improve Patient Outcomes
The decline in winter referrals for skin cancer likely occurs because people tend to notice skin changes less often during this season, and others may also point them out less frequently (for instance, as suggested by Walter et al. ). This effect has been observed for several years, not only in the UK but also in countries like Italy and France.
What is notable about the UK, however, is that it lags behind much of Europe for cancer survival rates – this includes skin cancer, though to a lesser extent, given the generally higher survival rates compared to others.
Early detection represents the greatest opportunity for addressing this disparity. Patients who delay their presentation to a GP until the summer may be diagnosed with cancer at a stage later than otherwise.
While skin cancer survival rates are relatively high, late-stage detection significantly reduces chances of survival. For patients diagnosed at stage 1 cancer (55.5% of diagnoses), the 5-year survival rate is 100%. This, however, drops 84% for stage 2 (21% of diagnoses) and 73% for stage 3 (8% of diagnoses). Detecting melanoma at stage 1 versus later stages could save 560 lives for every 10,000 patients. For 2022, that means an extra 170 lives could have been saved out of the 3,000 potentially delayed diagnoses.
Addressing Service Challenges
The surge in summer referrals for skin cancer also strains the capacity of healthcare providers, leading to delays in diagnosis and treatment for all dermatology patients. This issue has become even more significant in the aftermath of the pandemic, as healthcare services struggle to cope with mounting waiting lists.
The following chart highlights this challenge – as the volume of 2WW referrals increases, so do the 62-day target to treatment breaches. Although the time lag between the two is not substantial, it has widened since the pandemic, likely due to the extra burden posed on services by the elective backlog.
Patients referred just before the peak experience the most significant impact, as the capacity for 2WW referrals competes with the capacity for treatment. This indicates a healthcare system under pressure, where there simply isn’t enough capacity to handle a sudden rise in referrals. To address this issue, targeted solutions are needed.
What can we learn from this?
There is a strong case for gaining more insights across various cancer types, examining inequalities, geographies, pathways, and populations to align them with the Core20Plus5 priorities.
This knowledge can help to focus on a broader range of interventions and measure their impact.
For skin cancer, two clear opportunities emerge:
Implementing winter skin check campaigns, akin to successful breast cancer awareness efforts. Targeted outreach could encourage earlier presentation, improving outcomes.
Provider strategies to manage summer surges, such as Teledermatology and cross-site collaborations. New approaches may expand capacity for minor procedures and biopsies, speeding up diagnosis and care.
Data-driven insights can focus and maximize initiatives by revealing where needs and inequities lie. They also offer opportunities to monitor progress and assess effectiveness.
Learning to harness information will be crucial to the future of the NHS. The skin cancer example serves as a testament to how data can uncover life-saving possibilities and point the way toward realising them.
Data Engineering as the Backbone of Modern Healthcare Analytics: Experience from Our Work in the NHS
As a healthcare technology firm, we are known for our expertise in analytical work for the NHS. This includes operational improvement or benchmarking delivered visually via dashboards, and digital tools to improve scheduling, bed planning or empower better decision-making on the ground.
But people are realising that analytics need good engineering
Until recently, we were seldomly asked about how it all works under the hood. However, we are noticing that as trusts and national bodies have developed more and more dashboards, analytical tools and data collection systems, they are noticing errors, outages, delays that are due to back-end processes not being scaled and maintained in line with the increased demands of the analytics teams. In turn, questions that we are asked are pivoting from “can you help us build a dashboard?” to “can you help us make sure our dashboard works and is usable?”.
New tools make it easier than ever to do it well
At the same time, the tools that are available for data engineering, especially in the cloud with Azure, AWS or Gcloud, are making it easier and easier to get it right. And when data engineering is done well, it can ensure that analysts have the right data, refreshed without delay, displaying the right things in the right place.
Data engineering is like plumbing: data flowing to where you need it at the right time.
To deliver high-quality analytics to our clients we have invested heavily in data engineering, both in terms of tools and talent. Our team of data engineers is skilled in building efficient and robust data pipelines using cloud technologies such as Azure, which allows us to process, store, and analyse vast amounts of data in real-time and get our data plumbing right first time.
Example: Work with a National Programme on processing a national data collection
We recently worked with a national programme that collected theatre data from every Trust in the country. The data needed to be organised, flow together into a coherent schema and ultimately be validated, checked and organised into templates and tables. All of this required to be done in a stable, scalable and reproducible fashion. We achieved this using Azure tools, which allowed the data to power high quality analytics and feed into performance improvements for the entire country.
For a detailed case study of this work, click here.
Data engineering is the foundation of analytics
As data and dashboards proliferate, data engineering is becoming increasingly important and more senior leaders are noticing that problems with dashboards may lie under the hood. At Edge, we believe that getting the data right requires even more than engagement on the ground and high quality analysts. It needs the right plumbing to ensure that integrity, scale and reliability of any data products are guaranteed. Talk to us about our data engineers, and rest reassured that any data products we deliver have the highest quality plumbing underneath.
A Proactive Solution to Bed Configurations: Using Simulation to Challenge Narratives
Being able to plan proactively to devise flexible bed configurations is a major challenge for Trusts that strive to improve the services they provide to patients and meet their changing needs.
Edge Health was recently approached by a large specialist Trust that wished to better understand their bed requirements. The Trust wanted to challenge the existing narrative within the organisation on the number of beds required and gain the ability to simulate initiatives such as building new theatres, adjusting case-mix, or achieving national or peer benchmarks for length of stay.
Our expert solution
To meet these needs, Edge followed a tried and tested three-step bed modelling approach, now implemented at several Trusts, which balances efficiencies from using existing frameworks together with bespoke tailoring to ensure a good fit for local circumstances.
Here are the key steps of our approach:
We collected detailed hour-by-hour patient data by ward, team and procedure to gain a deep understanding of the Trust’s current bed usage.
We engaged with individuals in the wards and across the Trust to understand the wards structure, usage, capacity and any future changes, and collect requirements for the bed model.
We built an interactive simulation tool to reflect the Trust’s risk preference and occupancy rates. We built several iterations of the tool to ensure it fully fitted the Trust’s needs.
Screenshots from our bed modelling dashboard
A product built to last
The model is now used within the Trust to scope the building of new theatres, change the booking of patients to achieve more constant bed occupancy, and plan for next year. We received excellent feedback from our client, who praised the interactivity and the built-in ability for lateral thinking.
Finally a bed model that you can play with, use to challenge narrative and look at things from different angles. Not just a report. I really like it
– The trust’s COO
We look forward to continuing to help healthcare organizations make data-driven decisions. If you have any questions or would like to know more about our approach, please don’t hesitate to get in touch.
The NHS’ Balancing Act: Prioritising Urgent Cases While Tackling the Elective Backlog
Over a year ago (February 2022) NHSE set out clear targets for maximum wait times for elective care, as part of the plan for dealing with the pandemic-induced backlog. The aim has been to achieve zero 78-week waits (78WW) by the end of last month, zero 65WW by March 2024, and zero 52WW by March 2025.
Let’s explore this operational challenge
A look at the data will show that around 5% of the RTT waiting list is waiting over 52 weeks, and only 0.8% over 78 weeks.
This seems like a fraction of the total waiting list – why can’t we just treat all of those patients straight away?
The amount of long waits varies across the country
There is significant variation in the waiting lists between systems. In some ICSs, 0 patients are waiting over 78 weeks already. In others, almost 3.5% of patients have been waiting over a year and a half. Crucially, the challenges vary between specialties. An example of this is the ENT operation for septorhinoplasty, or “nose job”, a procedure whose high complexity and relatively low clinical urgency means that patients waiting for this surgery often must wait over 1.5 years.
The question remains – why can’t existing capacity be diverted to those long waiters, to clear the backlog in 1-2 months?
Systems are actively targeting long waits – though there is no “quick fix”
When we compare activity in December with the size of the 78WW cohort across ICSs, it emerges that many systems could clear their 78WW backlog using a fraction of monthly capacity. In fact, on average across all Trusts, 3.3% of one month’s capacity (December 2022) could have been used to clear all 78-week waiters.
However, for other ICSs, around one in every four patients treated would have to be from the long-waiters cohort to reduce 78WW to zero.
Here are two reasons this would not work.
The Patient Tracking List (PTL) for elective surgery is a “live” document
Much like a Google doc, the PTL is constantly changing.
One hospital could have 100 patients waiting over 78 weeks on February 1st. Even if all of those particular patients were treated this month, several more will cross the 78+ week threshold as the month progresses. This means that the activity required to reduce 78WW to zero is actually much higher than just number of 78WW that we have right now. It includes patients who are also about to approach the 78WW mark.
We cannot forget about risk
Typically, Trusts will apportion capacity based on risk – the most clinically urgent cases (emergency admissions, cancer patients) need to be seen fastest and will be bumped to the top of the queue.
A score often used to allocate priority – the “P” priority score, developed by the Royal College of Surgeons -, is a time-based measure of risk that focuses on mortality. Patients within the 78WW cohort were assigned low mortality risk scores on triaging, which has contributed to their growing wait as more clinically urgent patients have been seen first. While there is evidence that waiting for care for prolonged periods can carry substantial morbidity, the issue of mortality risk remains.
Juggling finite resources remains a conundrum
Prioritising 78WW patients may take away capacity from the most urgent patients. On the other hand, not addressing the backlog causes longer and longer waiting lists, and increases risk that some of those patients will need urgent care. Operational and clinical teams handle this balancing act on a daily basis – it is essential that they have access to timely, accurate and insightful data that support their decisions, ensuring that they can safely and effectively managing waiting lists.
In the next blog, we will explore innovative methods that operational teams are using to tackle the elective backlog, and highlight the tremendous work by specialty teams to overcome their unique, specialty-specific challenges