How to build digital twins within healthcare

In May, the NHS announced that waiting lists for routine treatment in hospitals had risen for the first time in seven months, despite being out of winter pressure season. The waiting list sits at 7.42mn appointments at the end of March, a steep increase of 18,751 from February.

While it is wholly acknowledged that waiting lists tend to fluctuate with natural spikes in the system, this data comes at a time in which the Labour government finds itself under the spotlight to reduce waiting lists and improve overall NHS performance.

Wes Streeting, whose mission for this parliamentary term is to revamp healthcare efficiency, has already made some radical changes with a view to delivering on this promise. The new 10-Year Health Plan laid out the importance of making the NHS the most “digitally accessible” system in the world. Moves have been made in this direction by announcing promising upgrades to the NHS App that enable its use as a digital front door to the entire NHS.

The department has also teased its ambition to develop a ‘Single Patient Record’ for each NHS patient, using their appointment history and medical records across care specialties. The ambition is that each patient can then seamlessly access care providers across the NHS, without having to repeat their story.

I’d argue for taking this one step further and creating a 360-degree identity of each citizen within their context. Connecting their data across government departments, educational institutions and other public services will allow healthcare providers a full understanding of each patient case.

What are digital twins?

It’s not enough to address individual problem areas such as growing concerns about mental health or challenges with the demand and strain on the system. Instead, there is increasing pressure on government to tackle these systemic issues at their core – using patient data to create a 360-degree identity of each citizen within their context and improve overall population health.

One of the ways to meet these challenges effectively is by embracing digital twins. Digital twins are gaining significant traction as more industries recognise their value in identifying how we tackle complex problems. Put simply, a digital twin works by replicating a physical asset in a digital world, created using smart sensors and data from the original source to model and predict how something will or could behave.

System digital twins are a virtual representation of a real-world system. The twin is made first as a digital model of the real system, using simulations, 3D models or system diagrams. System data can then be integrated into the model continuously or at regular intervals, allowing it to update in real-time. People, places and concepts can be fed into the twin to mirror the current state of the physical system. That means if something in the real system changes, the twin can reflect this immediately.

Simulating society through data integration

Digital twins are currently most often used across supply chain industries to improve real-time insights around customer journeys, in-transit inventory and staffing, and modelling against supply chain variability (McKinsey). The technology is gradually becoming more commonplace – often used in the manufacturing industry to produce models – and Gartner projects that by 2034, global revenue for simulation digital twins will reach $379 billion.

The healthcare industry is starting to adopt digital twins, with pharmaceutical companies embracing them for shipment operations and forecasting in clinical trial labs. But their potential expands beyond clinical trials, and they can be implemented to forecast trends and make predictive analysis about population health. If NHS patient data is fed into the digital twin, data engineers can then model scenarios on the patient population, to gather patient-level insights in the case of unexpected scenarios. To unpack how we can use this innovation to improve health outcomes, it first needs to be understood.

For example, Britain’s life expectancy crisis is developing, with one community reporting male life expectancy on par with war-torn Syria. Digital twins would allow for analysis of why life expectancy is worsening, using a trial-and-error approach to solutions without the need to run studies on the public. Alternatively, if the UK sees a sudden rapid rise of immigration, engineers can simulate the effect on population health with a rise in the spread of foreign diseases and an increase in pressure on the NHS.

Digital twins can be used to develop clinical trials that will get quick and accurate results. Now that the government has announced it will be automatically inviting patients to join clinical trials based on their health data and additional insights, digital twins can map clinical trial outcomes to optimise design for ideal outcomes. This also allows researchers to scale drug discovery, as they can model patient data to get faster and more personalised research.

A third opportunity for digital twins to contribute to healthcare and patient wellbeing is through overall hospital management design and care coordination. By taking a more predictive lens on patient outcomes, hospital staff can staff their treatment centres accordingly and apply the right level of resource.

There is great opportunity for predictive analytics that traditional models of aggregated intelligence didn’t have. As one of the government’s key pillars is “prevention over treatment”, taking advantage of this technology will give the UK’s health service a leg up in the future of its population health.

Rethinking local healthcare with digital twins

The NHS’s regional split must also be used to the system’s advantage when dealing with population health. Integrated care boards (ICBs) and NHS trusts have access to a whole network of health data that applies to their local region, and regional health datasets can be instrumental to the national health ecosystem.

To paint the picture of why regional health data is important, regions with ageing populations may see patterns in diseases that affect an older age group and therefore may be disproportionately reliant on community care. Urban populations may see more illnesses spike from air and water quality issues than a rural population. While all patient data feeds into a wider system, it’s important for ICBs to understand their own geographical challenges.

Regional and local data is incredibly valuable, but currently underused. By beginning to implement digital twin technology at a regional level, ICBs and trusts can start analysing their regional health trends to understand resourcing through a predictive lens. This means that under increasing waiting list pressure or staff shortages, organisations within the NHS can see where they’re lacking, and apply more resources to specific care specialties that will be most valuable for their patients.

Digital twins can also be used to identify regional cohorts of patients that would benefit from prevention initiatives – for example the child flu vaccine, or a diabetes prevention programme. Through a digital twin simulation, local organisations can identify preventative programmes that will be beneficial for their population. And conversely, understanding the potential outcomes of these patients through a digital twin can help determine the cohort.

If able to tackle public health through greater predictivity and prevention, digital twins quickly unlock benefits and rapid innovation to begin improving outcomes at scale. The potential benefits for digital twins go beyond hospital management and improving the efficiency of patient care – there are additional benefits like drug discovery, R&D and disease prevention that have the potential to improve overall population health.

The National Digital Twin Programme has been developed to ensure that the digital twin market is built on secure, trustworthy and ethical standards, while maintaining interoperability and adaptability. There is still a lot of debate around the ethics of using NHS patient data. The way I see it, is that using patient data is not only acceptable – but necessary – when the outcome goes back into patient health. It’s not about arbitrarily collecting patient data, but the safe and secure management of patient data for good.

It is in all our interests to prevent the impact of the next global pandemic, or even just ensure that our neighbours are all receiving a high standard of care. The NHS is one of the largest banks of healthcare data in the world, and we need to use it to improve care for generations to come.

---

Ram Rajaraman is Healthcare and Life Sciences Industry Lead at Quantexa.