Better patient discharge decisions with machine learning

About ePsychiatry Unit at Västra Götalandsregionen

The ePsychiatry Unit (ePsykiatrienheten) is a joint resource for the entire adult psychiatry service at Gothenburg’s Sahlgrenska University Hospital. Its mission is to develop digital treatment support for use in the unit’s everyday operations.

Psychiatric care in Sweden has limited resources. Nursing wards often operate at full capacity and one patient may have to be discharged to allow another to be admitted. Doctors therefore have to prioritise which of their admitted patients they believe are ready for discharge. If the wrong patient is discharged, there is every risk that he or she will soon return to the nursing ward. This leads to considerable suffering for the individual and significant costs for the health-care system.

Capgemini Insights & Data carried out an Analytics Jumpstart with the aim of predicting early readmissions to addiction care wards in Region Västra Götaland (VGR).

The current manual process was expanded with machine learning

Today doctors manually prioritise which patients to discharge. They have to take into account a wide variety of data about each and every patient – both in the form of medical journals and as data about patient demographics, care history, diagnoses and use of medication. The ePsychiatry Unit wanted to find out if advanced analysis using machine learning methods could utilise this data to predict which patients were likely to be readmitted after discharge. If such a process worked, it would be possible to introduce a system to help doctors make calculated patient discharge decisions.

Pilot project with knowledge transfer as the milestone

Machine learning projects are very special since the prerequisites for being able to succeed are hidden in the data itself. It is not possible to know in advance if there are any clear patterns that differentiate groups of patients, the system simply has to be tested. The Capgemini Insights & Data Analytics Jumpstart Method gave the ePsychiatry Unit the opportunity to undertake a pilot project to examine – quickly and without major initial investment – whether machine learning was a usable tool to assist doctors in the patient discharge process.

Another important parameter for the ePsychiatry Unit was the scope for knowledge transfer during the course of the project. The staff wanted to learn how projects in the field of Artificial Intelligence can be implemented, and Capgemini Insights & Data offered close cooperation featuring continuous updates and discussion on the project’s progress. These discussions led not only to knowledge transfer, but also to the use of subject-specific insights to develop better machine learning models.

“For us this has been an immensely exciting and educational journey. We had the opportunity to test whether all the data we save can actually be used for this kind of analysis. What’s more, we have learned a lot about how to undertake machine learning projects,” says Mikael Mide, licensed psychologist and project manager at the ePsychiatry Unit.

Successful pilot delivery and secured continuation

During the course of the pilot project, clearly discernible patterns were found in the data that could be used to predict which patients would soon be readmitted if they were discharged. The machine learning models that were created thus delivered reliable results, not just random success. In addition, a lot of insight was gained into which factors impacted a patient’s risk of readmission. These insights can be applied directly to everyday operations and can help doctors take better manual discharge decisions.

One common problem with the implementation of pilot projects is that it can sometimes be difficult to get beyond the pilot phase and actually benefit from the achieved results. For this reason, in addition to the machine learning models and project report, the project also delivered a proposal for how the results could be used in everyday operations. As a recommended subsequent step, the project suggested an investigation into the prerequisites for implementation into operational production. The ePsychiatry Unit approved this step and has now authorised financing to continue with phase 2 of the project in 2020.

We are very keen to launch the next phase of this project. There have been some pilot projects but few have actually implemented this type of tool in clinical operations. That’s why it feels as though we now have the opportunity to break new ground and use the very latest technologies to offer our patients the best possible care,”

– Mikael Mide, licensed psychologist and project manager at the ePsychiatry Unit.

Better patient discharge decisions with machine learning


Insights and Data

In a world where change happens in a split second, organizations of today and tomorrow must...

Contact us

Thank you !

We are sorry, the form submission failed. Please try again.