Business challenge

To improve the efficiency and quality of nursing care, Liège University Hospital needed to better understand its performance. How could it unlock valuable insight from large and complex Excel files?


Liège University Hospital uses IBM® Watson Analytics™ to analyze and visualize data, making it immediately much more accessible and gaining insight into the most relevant questions to ask.



insight into complex data, without the need to build a model in advance


time and resources in analysis, enabling a single user to perform advanced queries


visibility of key performance metrics throughout the organization

Business challenge story

Bringing data to life

With 895 beds, Liège University Hospital delivers its services across several hospital and polyclinic sites in order to provide patients with specialized care and consultation across all medical disciplines both in Liège and the surrounding areas. Alongside its clinical activity, teaching and research constitute two important elements in the development and propagation of medical innovation. The activity undertaken by employees of Liège University Hospital places the organization in an internationally renowned network of experts and scientists, in addition to contributing to diagnostic and therapeutic advances to the direct benefit of the community.

The level of public funding for each hospital in Belgium is determined using a dynamic approach that takes into account all activities related to patient care. The country’s 120 general hospitals gather a large volume of data concerning patients, conditions, activities and clinical staff. This information is submitted to the Public Health Service via a federal portal, where it is subjected to a variety of tests and validations. Patient diagnoses and treatments are recorded using the 14,000 diagnostic codes and the 4,000 procedural codes in the ICD 9-CM coding system. Nursing activity is codified using a set of 78 activities recorded for each patient hospitalized during the period of data collection. Data on staff is also recorded.


Once both the federal authority and the hospitals have agreed on the final submissions, the data is integrated in the financing mechanisms. It is then used to determine each hospital’s “justified activity” as a function of admissions (by number and type of pathology) and of the national averages for length of admission by pathology group, severity of the patient’s condition and age group.

Based on how its activities compare with the national average, each hospital is assigned a number of points, and the total hospital budget is shared out according to the defined monetary value of each point (currently +/- EUR75,000). Each hospital also receives several complex Excel files annually containing feedback that shows, specifically, the points—and therefore the funding—which have been allocated. The feedback notes the details of every patient’s admission, including the length of admissions as well as the number of points generated by medical activity undertaken during a patient’s stay.

To help improve both nursing resource planning and overall performance, Liège University Hospital wanted to be able to profile its activities based on the information it had submitted.

Olivier Thonon, Senior Head Nurse and research fellow at Liège University Hospital, says: “We knew very well that there was a goldmine of information in these reports that could help us improve our nursing practices and performance; we wanted to be able to understand it more easily.”

The Nursing Department of Liège University Hospital also wanted to integrate the supplementary information in the feedback reports with its management systems, so that it could optimize efficiency and productivity.

Olivier Thonon says: “We had all the information, but we needed to make the federal feedback really speak. The first objective set by the senior nursing management was to increase the internal understanding of the feedback and enhance its practical value, not just for the department but for all other relevant groups. The second objective, once we had a clearer picture of our activities, was to improve the allocation of nursing resources based on objective calculations of requirements.”

We had all the information, but we needed to make the federal feedback really speak.

Olivier Thonon, Senior Head Nurse and research fellow, Liège University Hospital

Transformation story

Visualizing performance

Working with Aexis Group, a long-term analytics partner for the hospital, Liège University Hospital decided to use IBM Watson Analytics to help it unlock the value of data in the feedback reports. Watson Analytics is a smart, cloud-based data discovery service that guides data exploration, automates predictive analytics and enables the effortless creation of dashboards and infographics.

Nancy Laport, Consultant in Hospital Management Analysis at Aexis Group, comments: “We had seen a demonstration of Watson Analytics, and we liked the fact that you can do a lot of analysis without having to build a formal model. In practical terms, this means that you don’t need to know in advance what kind of questions you want to ask, and this was a good fit with the challenges Liège University Hospital was facing. Olivier Thonon had given us a copy of the feedback file, and we fed it straight into Watson Analytics to see what the outcome would be.”

“The solution immediately produced what were clearly relevant questions and topics for investigation,” recalls Olivier Thonon. “Watson Analytics makes the Excel files much easier to interpret, and enables us to proceed with data discovery and the development of analyses. When we saw the possibilities, we realized that the solution could bring us a major return on investment.”

After experimenting with a free trial version of the solution, Olivier Thonon showed the potential value-add to his executive management and gained approval to subscribe to the professional version, which is designed for enterprises and includes a multi-user environment for collaboration and more data connectors.

“At present, I use the solution two or three times a week,” he says. “First, to enhance the readability of the feedback and to put into context elements that could help us improve. I’ve built several scenarios based on the feedback file, which are used to provide simple visual performance reports to the different care units. We’re beginning to use the solution to build profiles of each care unit in terms of workload, and we’re also working to feed Watson Analytics with a whole set of data that is potentially relevant for analysis.”

Watson Analytics makes the Excel files much easier to interpret. When we saw the possibilities, we realized that the solution could bring us a major return on investment.

Olivier Thonon, Senior Head Nurse and research fellow, Liège University Hospital

Results story

Awareness drives optimization

Using Watson Analytics gives Liège University Hospital the ability to understand and query complex data rapidly and with minimal effort.

Nancy Laport comments: “One of the major advantages of Watson Analytics is its ability to go rapidly from complex data to visual reports that can be used immediately at the management level. When you analyze, you want to get results as fast as possible and to build awareness among decision-makers about the problem or the action that needs to happen as a result.”

Olivier Thonon adds: “Our first objective was to be able to understand the feedback reports, and we have clearly achieved that. In fact, we’ve gone further: we knew that the feedback file was rich, but we didn’t realize it was quite this rich. We already understand the key points, and this will enable us to make practical modifications. That’s the second objective, and undoubtedly it will take longer to modify our working practices. The ability to visualize the data is a big help here, making it easier to communicate the results of analysis so that everyone in the organization can understand what needs to be done.”

The ability to see the hospital’s performance in detail and to compare it to national standards will help ensure that Liège University Hospital first maintains and then attempts to improve its standing. As Liège University Hospital ramps up its use of the solution and potentially extends it to other scenarios, Olivier Thonon expects greater awareness of performance to spread through the organization: the first step towards making improvements.

He comments: “Awareness comes from a good understanding of the available data. With the same effort as before, we can achieve far more in our analysis, and this gives us both the desire and the time go further. We can ask ourselves new questions, make new connections, and base our decisions on objective data. Of course, it’s a step-by-step process: we make sure we understand each scenario, and then we look to add in more aspects as suggested by Watson.”

Ultimately, Olivier Thonon envisages automating the profiling of each care unit’s workload and performance based on the feedback data and on data from operational systems. This would help the organization understand its changing workload and how best to allocate available nursing resources.

For the time being, the focus is on mining the feedback for new insight into hospital performance. Olivier Thonon concludes: “Although we receive the official feedback only once a year, we can keep using Watson Analytics as a tool for improvement, putting new data in to analyze how our performance is changing over time. This doesn’t give us the full financial picture, because it doesn’t take into account the performance of the other hospitals, but it’s a very useful interim benchmark. The first step towards improvement is to understand where your challenges lie, and that is certainly something we can now do.”


Liège University Hospital

Liège University Hospital (CHU de Liège), is a pluralist and public teaching hospital, and the only university hospital in the Wallonia region of Belgium. With more than 5,300 staff, it is the largest employer in the Liège area. Professionals from numerous specialties, including 900 physicians across around 50 services, use their skills to ensure high-quality patient care. Every day, the hospital provides consultations to more than 3,000 people and treats the same number as out-patients, as well as handling 150 conventional admissions.

Solution components

  • Cognos Analytics
  • HC: Intelligent Platforms

Take the next step

IBM Analytics offers one of the world's deepest and broadest analytics platform, domain and industry solutions that deliver new value to businesses, governments and individuals. For more information about how IBM Analytics helps to transform industries and professions with data, visit Follow us on Twitter at @IBMAnalytics, on our blog at and join the conversation #IBMAnalytics.

Aexis Group, an IBM Premier Business Partner, is a software and services organization headquartered in Belgium and with additional business units in France, Luxemburg and the Netherlands. Offering solutions across corporate performance management, business intelligence, IT service management and healthcare/hospital systems, the organization has more than 500 clients. To learn more about Aexis, visit

View more client stories or learn more about IBM Analytics