Problem
In Europe alone, 2.6 million patients develop a healthcare-associated infection which is the most common adverse event. Apart from the harm this brings to the patients, it also leads to 37 thousand deaths and an additional cost of 2000 EUR per patient to the healthcare institute. P3S provides automated outputs on adverse events to healthcare professionals and healthcare providers. This is done by complex algorithms that scan routinely collected electronic health record data. The algorithms are scientifically validated and hence have high credibility. Currently, P3S focuses on diagnostic algorithms of adverse events, but it will also incorporate predictive algorithms that can detect patients at risk of future adverse events. P3S reduces the costs of surveillance of adverse events, reduces and improves the management of adverse events, and makes it easy and fast to evaluate the effect of changed work processes in healthcare. The system can be tailored to the needs of each healthcare provider.
Solution
Healthcare institutes can significantly improve patient safety and reduce their costs through continuous monitoring of adverse events through algorithms and re-use of electronic healthcare data using our P3S ecosystem. This system costs only a fraction of the costs associated with adverse events and can easily be scaled up from one bed to entire wards and eventually the whole institute depending on the wishes and needs of the institute.
P3S offers an ecosystem consisting of five software modules and consultative assistance around the automated surveillance and prediction of adverse events. Our software as well as the algorithms are developed and maintained by the company itself. This service is offered to healthcare providers who want to benefit from data-driven care, and researchers who want to bring algorithms to clinical practice. All clinically production-set algorithms will be CE-marked if they constitute a decision support for healthcare.
The heart of the ecosystem is the Algorithm Cloud Service where the algorithms are created and maintained based on research. The Data Adapter translates the data of the healthcare provider into the format necessary for the system to be able to run the algorithms. The Engine runs the algorithms from the Algorithm Cloud Service on the translated data obtained from the Data Adapter to generate the results. These results can be trained and validated so that their functionality meets desired accuracy requirements in the Validation Tool. Finally, the approved and validated results can be shown to the healthcare provider via the User Interface. If the care provider already has an analysis tool, it is possible to use this instead, or both.
The first algorithms for adverse events we chose to develop and implement deal with healthcare-associated infections. The algorithms can easily detect which patients developed an infection in the healthcare institute which in real time can be shown to healthcare providers. Just by monitoring infection status in real time, infections can be reduced by 20% (Gastmeier P et al. J Hosp Infect, 2006 Sep;64(1):16-22). Additionally, the system can be used to measure the effect of changed working methods and interventions in the institute on the presence of adverse events. It can also be used to validate the effectiveness of interventions and new products.