How digital twins of patient journeys can help improve access to innovative treatments

Digital twins can help healthcare providers test service changes, plan capacity and improve access to innovative treatments.

In brief

• Digital twins create virtual models of patient pathways that allow life sciences companies and healthcare organisations to test changes before implementing them.
• Healthcare providers can use digital twins to identify bottlenecks, assess capacity requirements and support treatment planning.
• Case studies involving oHCM and CAR-T therapies demonstrate how digital twins can help improve patient access and resource allocation.

Digital twins are often discussed as a future-facing technology. Their more immediate value may be in helping healthcare leaders test practical decisions about access, capacity and care delivery before those decisions impact patients.

Healthcare organisations today strive to deliver better clinical outcomes while managing rising demand and workforce constraints. At the same time, new and innovative therapies are also creating opportunities to improve outcomes for patients. Traditional reporting in healthcare settings can explain what has already happened, but it is less useful when organisations need to proactively understand and predict where issues might appear next, or run scenarios to understand how changes will affect the system and ultimately the care delivered.

By combining real-world data and leading-edge, potentially AI-driven, analytics with clinical expertise, life sciences companies and healthcare organisations can develop digital twins that create virtual representations of real-world patient pathways and journeys or services which can be used to test different scenarios before changes are made in practice. They give healthcare teams a controlled way to safely explore the consequences of decisions in complex systems, where demand, capacity, protocols and patient flow interact over time.

The term Digital Twin is often associated with personalised medicine and virtual replicas of individual patients. These are important examples, but some of the most valuable uses of digital twins today are more operational and experiential and address targeted patient groups and treatments. This is particularly relevant as healthcare services in tandem with life sciences companies prepare for the introduction of new medicines and innovations.  Recent work undertaken by EY Ireland's Life Sciences AI & Data team show how digital twins can be used to model patient pathways, identify constraints, bottlenecks and test interventions in specialist services where capacity may be limited, there are gaps in the patient journey or demand is rising.

Improving access to innovative cardiac therapies

One application focused on obstructive hypertrophic cardiomyopathy, or oHCM, where the introduction of a novel therapy led to increased demand in a specialty Inherited Cardiac Clinic. Waiting lists and times were growing, and there were concerns about clinic capacity, timely access, continuity of care and patient outcomes. Despite the existence of substantial data in respect of the service, there was no clear or integrated view of what the constraints were in practice and how different interventions would ultimately affect patient access to the new and highly beneficial treatment.

Using data science tools, EY Ireland's Life Sciences AI & Data team developed a digital twin of the oHCM patient pathway. The model combined outpatient waiting-list data, treatment protocols, patient demographics, resource constraints and clinic capacity assumptions. It simulated the journey from initial consultation through acceptance for the new therapy to actual treatment and maintenance, creating a site-specific view of demand and capacity which could be updated as real-world experience evolved.

The model allowed the clinic to test choices which would otherwise have been made without an evidence-based view of future impact. Changes to clinic frequency, staffing models, capacity assumptions and pathway design could be examined virtually before being implemented. This helped identify the key drivers of limited new patient initiation and gave decision-makers a clearer basis for forward planning, resource allocation and improving access to therapy.

Testing new models of CAR-T delivery

A second application shows how the same approach can support decisions about care delivery. CAR-T therapies in the context of cancer care, are resource-intensive and are commonly delivered in inpatient settings, creating pressure on hospital capacity as demand grows. For providers considering greater use of ambulatory care, the planning questions are difficult: what happens to bed utilisation, monitoring requirements and staffing if appropriately selected patients are treated differently, and how might those changes affect patient experience and service resilience?

EY Ireland's Health AI & Data team developed a digital twin model to support CAR-T service planning across treatment centres. The model categorised patients into defined care cohorts, simulated the treatment pathway from lymphodepletion through post-infusion monitoring, and established a baseline understanding using current inpatient delivery. It incorporated clinical assumptions, resource requirements and published evidence validated by healthcare professionals, enabling inpatient and ambulatory models to be compared consistently.

By testing assumptions such as ambulatory adoption rates, monitoring duration and the number of treatment sites, the model gave providers a clearer view of throughput, bed use and resource requirements under different scenarios. Clinical judgement was still essential, but it gave management a more disciplined way to understand trade-offs before committing scarce resources or changing how care is delivered.

A more disciplined way to plan

Across both examples, the central lesson is that digital twins of the patient journey are most valuable when they are anchored in addressing a specific issue, challenge or decision. Their usefulness comes from bringing data, clinical and product knowledge and operational assumptions into a model that can be tested, challenged and refined before action is taken.

The ability to do this is becoming increasingly important, supporting healthcare providers and life science companies to better plan for new treatments while maintaining access, capacity and quality of care. In specialist services, where small changes in capacity or pathway design can have significant consequences for patient access, the ability to test those choices in advance may become one of the most practical uses of digital health.

Such digital twins of the patient journey are an evolution beyond static models into dynamic, intelligent systems that continuously learn and adapt using real-time data from multiple, relevant sources. Through ecosystem partnerships between life sciences companies and healthcare organisations, and with cross functional collaboration, these digital twins can deliver win-wins by reflecting real-world system conditions and supporting optimal patient access to appropriate treatment. Supply and resource capacity and variability planning can help unblock bottlenecks in real time. Patient flow, demand and referral dynamic modelling and scenario planning enable smarter evidence-based and data-driven decisions. They also enhance healthcare organisations’ operational planning including strategic scenario-based decision-making.