Understanding the impact of a non-endoscopic diagnostic test pathway upon cancer pathways using System Dynamics Modelling

System dynamics modelling (SDM) provides a holistic approach to consider the effect of an intervention, incorporating the relationships and processes within a system. This approach enables the observation of expected and unintended consequences that may occur following implementation, providing a framework that can test a pathway based on varying demands and resource levels, ensuring deployment can be responsibly managed. As part of a wider mixed-methods evaluation, SDM can provide an enhanced view of the wide-ranging effects of an intervention, providing a robust framework to support economic and quantitative evaluation.

Introduction

Barrett’s oesophagus occurs when the inner lining at the lower end of the oesophagus is damaged due to acid reflux and can develop into oesophageal cancer. Unity Insights were commissioned to evaluate the impact of a minimally invasive, non-endoscopic diagnostic test for the detection of Barrett’s oesophagus and early oesophageal cancer for individuals living with chronic reflux.

The two main patient cohorts examined were:

• Proactive case finding: Patients living with chronic reflux who were at risk of oesophageal cancer were invited to take a non-endoscopic diagnostic test in Primary Care
• Barrett’s surveillance: Patients already diagnosed with Barrett’s oesophagus and requiring surveillance who were at risk of oesophageal cancer were offered the test in replace of an endoscopy.

A mixed-methods approach was used to observe the impact of the non-endoscopic diagnostic test, incorporating quantitative, qualitative and health economic analyses to provide a holistic assessment of the intervention and its impact. A key element of the quantitative analysis was the use of systems thinking, and the development of a system dynamics model, observing the impact of introducing an alternative diagnostic pathway upon the wider healthcare system, specifically endoscopic services.

System dynamics is an analytical approach used to understand and simulate complex systems over time. It focuses on the relationships between different components of a system, such as the people, processes, and resources, and how these interactions react to change. The method uses feedback loops, time delays, and a ‘stock and flow’ structure to explore how changes in one part of the system can have wider implications across the pathway.

System Dynamics Methodology

In order to simulate and forecast the long-term impact of implementing the non-endoscopic diagnostic pathway. The following steps were followed:

• Pathway Mapping: Modelling patient flow through the pathway, as well as the traditional endoscopic pathway, including identification, eligibility, uptake, adequacy, and diagnosis.
• Data Integration: Using real-world pilot data from three Cancer Alliances to inform model parameters reflecting case identification, diagnostic and treatment outcomes.
• Scenario Analysis: Three scenarios were modelled:
  • 10-year projection based on pilot activity.
  • Scaling to full Cancer Alliance populations.
  • Incorporating the use of the test for the triage of reflux patients to further reduce endoscopy demand.
• Assumptions: Included population growth, prevalence of Barrett’s oesophagus, and uptake rates. Some data were extrapolated due to limitations in real-world availability.

The system dynamics model was developed in Stella Architect, providing a framework to simulate activity across each cancer alliance over the course of a year.

Key findings

Initial Surge in Demand:

The introduction of Proactive case finding, to screen individuals at-risk of developing oesophageal cancer was identified to create a short-term spike in endoscopy demand, based on the system dynamics model, increasing demand upon secondary care services by 5-12% in the first year. This is primarily due to identifying more patients with Barrett’s Oesophagus requiring confirmatory endoscopies following the initial non-endoscopic screening test.

Long-Term Sustainability:

Over time, demand for endoscopies was projected to taper off and settle at a more manageable level. The number of endoscopies saved through non-endoscopic surveillance of Barrett’s Oesophagus patients over time was projected to outweigh those generated by proactive case finding. These results meant that a phased rollout of proactive case finding was recommended to avoid overwhelming endoscopy services in future, wider deployments.

Using non-endoscopic diagnostic approaches to also triage reflux patients on endoscopy waiting lists was projected to offset up to 95% of the additional endoscopy demand by year 10, highlighting that the full utilisation of the pathway could release pressure from secondary services in the long-term.

Operational Considerations:

The model clarified the importance of balancing proactive case finding with available endoscopy capacity and emphasised the need for efficient triage processes and data-driven planning to manage patient flow.

The potential challenges of proactively diagnosing patients with long-term risk indicators of developing Barrett’s oesophagus or oesophageal cancer, were modelled to observe the potential outcomes of operating the service model at scale. While the service helped identify patients at an earlier stage, allowing for early intervention, and provided reassurance to many more, the model allowed the evaluation team to observe the operational impacts, flagging potential risks to endoscopic services that could be mitigated as the service is deployed.

Overall, the use of a systems thinking approach allowed the project to provide analyses focusing on the operational consequences of the intervention, supporting planning and an understanding of the full potential benefits of future implementations.