Building the future: predictive models.
Our current healthcare system needs a thorough review if it hopes to remain sustainable. We talk about funding as being one of the main obstacles to making public healthcare viable, but this is just one of the aspects to be fixed, albeit a very important one. The trend in demand for health and social services is even more worrying. We need to start discussing what we can include and how we can fund future public service portfolios, and this is something we can only achieve if we work with long-term predictive models.
Other sectors that are extrapolating data to make market predictions are perhaps more unpredictable, but in health, we all know that the population is ageing every day and suffers from more chronic and complex diseases.
Almost 10 years ago in Catalonia, with the competitive dialogue procedure, we were working with big consulting firms specializing in data processing, and in the design, implementation and operation of a management and service model to use the data from the Catalan healthcare sector for the benefit of both the public and private sector (VISC+ project).
For this project, the availability of structured data derived from mass digitalization (in a fully anonymized and secure form, naturally), allied with the analytical potential of the available technology, pointed to considerable added value, including the ability to plan public healthcare policy. The project was never carried out.
Today, artificial intelligence (AI), as in other sectors, is considered to be the most likely area of research to produce the evolution of the system, and it is to be found in all strategic projects linked to the healthcare sector, in support for clinical decisions, personalized-medicine projects, etc.
Yet this is not true at the macro level, where it might provide the healthcare authorities and other agents of the healthcare system with a tool to support decisions on regional planning and positioning healthcare services and priorities going forward. It is also essential for projecting the number of professionals in the different specialties that will be needed and thus guide schools and universities.
This positioning requires the participation of the public and private healthcare sectors as a whole. It is in this workspace between public needs, private enterprise, and innovation that we can find the most opportunities for progress, either globally or in the defined strategic areas.
Before the COVID-19 crisis, analysis of the economic data had already been showing, year after year, a growing gap between healthcare expenditure and the healthcare budget; changing the trend when the difference is this big and some items are growing exponentially is very difficult, if not impossible.
If we were to correct these financial needs from the point of view of the future demand for services, with a long-term view of social and healthcare needs, to 2040 or 2050, using predictive models based on artificial intelligence (AI), we would see alarming increases.
In Finland, and other Scandinavian countries that have worked on this, the increases in demand are above 300%. Of course, this is, like here, largely due to the ageing of the population and to chronic disease.
We cannot be unmoved by this information – we say that we have a good healthcare model, but if it has no future, what is good about it? The situation we have experienced with COVID-19 has stretched the elasticity of the model to the limit, revealing many of its strengths, particularly the strength of its professionals, and many eHealth solutions that were available but were not in the schedule had to be implemented.
Despite resistance to change, the crisis has forced our hand and we have advanced in implementing telemedicine, telecare, and other solutions.
But the digital transformation is not just about implementing technologies that help us to be more efficient. It is far more than that. It is about technologies that are transforming organizations and that can guide our future. This requires using the data we already have to our benefit by turning them into useful and intelligent information.
This new knowledge, based on predictive models and with the help of AI, at the macro level, will allow us to propose new future scenarios, new ecosystems in which, as well as building alliances between the different partners involved in healthcare, we will have to redesign healthcare services based on future demand for services, and on the available resources.
If we can envision these future needs for services, we will have to seek out innovative solutions among the different agents of the healthcare ecosystem to strike major accords on shared risk, new funding mechanisms, and other aspects, accords that allow us to meet these new challenges and minimize their impact.
This will only be possible if we have predictive models that can tell us what kind of increases we are looking at. To build them, as well as the databases on demand for services available among the different partners in the healthcare sector, we will also have to look for correlations with other variables such as lifestyle, genetic makeup, working conditions, and environmental factors, or a combination of all of them.
If we want to maintain the benefits of the current healthcare model to the extent possible, we must anticipate the future; and we have all the skills we need to build it: the best consulting forms, companies specializing in developing the algorithms, processing power to treat the data securely and anonymously, which is of supreme importance, and large series of structured data to be able to create this knowledge.
Anticipating the future using predictive models will allow us to correct current trends in demand and seek out, where necessary, alternative priorities and funding sources, with more preventive and perhaps more efficient medicine.
But this will only be possible if we work together to build the future with predictive models.
Roser Artal, healthcare economist
Article for Diario Médico / 02 April 2022