Data saves lives
The ability to exchange information in an efficient manner suddenly became very important due to the corona crisis. Kurt Nys ICT Manager FPS Public Health
Self-service analytics with AI and Machine Learning capabilities
Data saves lives for FPS Public Health in Belgium
Of all sectors healthcare has been tested the most by the corona crisis. The FPS Public Health, the Belgian government body that watches over the general health of citizens, has been a SAS partner for decades and has already made substantial investments in data. To cope with COVID-19, however, they had to step up their game. For example, to be able to share data with partners and to predict infection rates or anticipate the occupancy of hospital beds.
The Federal Public Service (FPS) Health, Food Chain Safety and Environment has been using data for a long time – in the context of hospital financing, reporting, and to adjust hospital supplies. However, their system is becoming outdated and the landscape needs to be modernized. Kurt Nys, ICT Manager at the FPS Public Health, compares it to an old Mustang that is still powerful and beautiful, but consumes too much. Moreover, COVID-19 has created a number of unseen challenges.
“It took us by surprise and we were suddenly confronted with new requirements. All governments worked hard to collect the right data in short notice and to improve the quality of that informselviation”, says Kurt Nys. To combat the virus, hospitals and policy makers needed access to predictive models. The government department’s old platform was also not prepared to share figures and insights with the outside world. Because of the urgency of the crisis, SAS decided – as part of a special program for customers affected by COVID-19 – to provide the SAS® Visual Analytics and SAS® Viya® licenses in loan and to support the FPS Public Health with consultancy.
"Predictive models enabled us to better align the hospitals’ emergency plans with the evolution of the figures." Kurt Nys ICT Manager FPS Public Health
Predicting occupancy of hospital beds
The FPS Public Health closely collaborates with several other organizations, such as Sciensano, the National Institute for Health and Disability Insurance (NIHDI), and the federal agency for medicines and health products. The ability to exchange information in an efficient manner suddenly became very important due to the corona crisis. As a result, the government service urgently needed a cloud solution for all its data. Because of the European privacy directives, it was not possible to store the data on one of the big American cloud providers, such as Amazon or Microsoft Azure. SAS, together with its partners and the customer, therefore ensured that everything could run on a private cloud within a time span of just one week.
Thanks to this rapid implementation, the FPS Public Health and other agencies already in the first wave of the pandemic had access to an efficient dashboard to monitor hospital capacity. Predictive models can look up to fourteen days ahead and deliver reliable results to predict when additional beds will be needed. The absence rate of the hospital staff was also closely monitored to ensure that all patients could be nursed.
“These models enabled us to better align the hospitals’ emergency plans with the evolution of the figures. If we had not collected this data during the first wave, it would have really weakened us in the second wave”, explains Kurt Nys. “We were now able to switch to a higher crisis level more rapidly, which meant that bed occupancy in the intensive care units never exceeded.”
The platform also guarantees the confidentiality of the data and even adjusts the information to the user. For example, the FPS Public Health does not need to know the identity of the patients who are occupying the hospital beds, but those profiles are important for Sciensano in order to perform risk management and provide optimal advice for policy makers. So, where one party cannot identify the person behind the data, the other does have access to more individualized data. All information is anonymized and fully compliant with the GDPR legislation.
Company – Facts & Figures (H2)
secondary metric or company fact
(Avenir Light R & XS)
secondary metric or company fact
secondary metric or company fact
Returning from a red zone
The Passenger Locator Form is another application that the FPS Public Health had to launch in the run-up to the summer. Near the end of the first wave, the borders reopened and the government needed the ability to track all traffic of people going in and out of the country. To collect the right data, they created a form that travelers must fill out 48 hours before arriving in Belgium. That way, people could quickly receive an invitation for a test if they had visited a red zone. Thanks to the shared dashboard, the data could again be easily exchanged between the different partners.
“Towards the end of the summer, a lot of people without symptoms were tested preventively. As a result, the positivity ratio dropped and there were suddenly not enough reagents to perform the tests. In response, a self-assessment tool was added to the form in order to gather additional data and to effectively determine, based on that information, whether someone needed to be tested”, says Kurt Nys. As soon as the test capacity was restored, the measure could be reversed to ensure that every traveler returning from a red zone once again received a request for a corona test.
Data is a very important weapon to combat COVID-19, but even beyond this crisis the public health sector has much to gain from efficient use of information and insights. Just before the outbreak of the virus, the FPS Public Health participated in a hackathon for SAS partners and customers. “Our case was about modeling the impact on ambulance response times”, says Kurt Nys. “For example, data allows us to determine which hospital an ambulance should drive to based on traffic information. That choice can save a lot of time when the Antwerp ring road is jammed.”
“In addition, ambulances sometimes transport patients with very complex pathologies. In certain cases it may therefore be appropriate to ignore the nearest hospital and go straight to a more specialized alternative. That way patients immediately receive the best care and you can avoid a secondary transport”, says Kurt Nys. Such a system also offers great potential for the future, for example by working with hospital networks. “Today almost every service is present in most hospitals, but do they all need an oncology department? Perhaps we can improve health services by grouping experts as much as possible?”
During the crisis year 2020, the FPS Public Health experienced the power of data more than ever and the fast implementation was an impressive achievement. “The various taskforces have congratulated us, so we are very satisfied with the collaboration with SAS. Moreover, our application was nominated for two eGov Awards and we won the main prize for best collaboration”, says Kurt Nys.
COVID-19 has accelerated the transition within the FPS Public Health. Previously, the solution served mainly to capture and structure the data. Today this has been extended to a platform that allows employees to perform analyses themselves and on which AI and Machine Learning applications can be deployed. In addition, data can now be exchanged between different partners in a secure and effective manner.
In the near future, SAS will help the government department to switch from the previous system to a modern SAS Viya environment. This will give the FPS Public Health even more possibilities to perform future-oriented analyses for a broad public. And thus optimize health care in Belgium. Or how the old Mustang is still looking as magnificent as ever, but now gets a sustainable and future-proof engine under the hood.