Switzerland

Back to Search

International Partner Search

Innovation & Technology Offer

Predictive software for an early detection of high-risk readmission patients.

Country of Origin: Spain
Reference Number: TOES20161110001
Publication Date: 10 November 2016

Summary

A Spanish medical research institute looks for a partner commercializing hospital management ICT solutions (Information and Communication Technologies). It offers a license agreement to exploit a package of predictive software delivered as a website and aimed at detecting and classifying patients according to the risk of 30-day readmission. Technical cooperation agreement to make easier the marketing of the tool as a turnkey project is offered too.

Description

It is estimated that between 10% and 20% of the patients being hospitalized and subsequently discharged require an urgent readmission within 30 days, depending on the type of hospital and the disease treated, which represents an additional cost for both, the patient being again hospitalized and the hospital itself, which has to reallocate human and material resources to cope with a number of unexpected admissions. 
For instance, assuming a hospital having an average of 120 hospitalization discharges a day, taking a minimum rate of 10% 30-day readmission and a minimum cost of 3.000€ per day for each patient being hospitalized, this hospital would then incur in an extra cost of 1.080.000 € per month originated by 30-day readmissions. In this circumstances, the simple fact of reducing 50% of the total 30-day readmissions would save this hospital 540.000 € per month. It is therefore a major issue in hospital management finding the way of diminishing the number of 30-day readmission as much as possible.
Many hospitals are still today approaching this problem with assessment and follow-up protocols implemented directly by the healthcare professionals, which is having a direct limiting effect on the load of 30-days readmission risk patients that can be properly handled by the hospital.
In this context, a Spanish research medical institute has developed a predictive software aimed at minimizing the risk of 30-day readmission of hospitalized patients, consisting in an early detection system that allows the identification of those patients that statistically have the highest probability of being readmitted within 30 days, allowing thus to incorporate them in a specific transition program planned to monitor the medical factors leading to 30-day readmission.
The technology is based on the practical application of predictive models, which are developed using machine learning algorithms (Naïve Bayes, Random Forests and logistic regression). It consists in a software application developed with general purpose programing languages (Java and/or C++) that classifies patients according to 30-day readmission risk criteria, plus a package with the required associated software needed for a correct integration and deployment of the application into the information system of the hospital. It is ready to be delivered as a web-based application, can be offered as a generic 30-day readmission model, for all hospitalized patients and regardless the type of disease, or customized as 30-day readmission model for three types of specific diseases: a) patients with respiratory problems, b) oncologic patients and c) cardiac patients.
The medical research institute is looking for partners interested in commercializing this technology through a license agreement. The application can be commercially exploited as a web-based central service offered to several hospitals or it can be sold and installed individually in each client hospital. The partner sought shall have the capacity to introduce this application in the market of hospital ITC (Information and Communication Technologies) management tools. The medical research institute is also offering technical cooperation agreement with the aim of assisting the commercializing partner in completing the adaptation and final configuration of the tool, allowing this way the commercializing partner to sell it as a turnkey project.

Advantages and Innovations

The technology allows hospital managers in the early detection and classification of patients in terms of 30-day readmission risks, and serves as a definitive guidance to implement effective transition programs, which results in a +/- 50% reduction of 30-day readmissions and the associated costs.
The application is based on advanced predictive modeling developed through machine learning algorithms (Naïve Bayes, Random Forests and logistic regression).
The software is developed in a general purpose programming language (Java y/o C++).
The technology is ready to be commercialized as a web-based application.
It is an easily adaptable technology that can be configured in different sites and systems due to the use of incremental algorithms.
The technology enhances the efficiency of medical monitoring and makes easier the implementation of better transition programs, and consequently improves the patients’ health.
The commercial exploitation model is not closed, it means the technology can be offered as a central service for many client hospitals or as individualized solution for each client hospital.

Stage Of Development

Field tested/evaluated

Stage Of Development Comment

The Spanish medical research institute has obtained field results from tests conducted with the cohort of patients from a well-known hospital. The sample considered includes patients going through urgent readmission within 30 days after discharge. The collection data spam of time is from 01/01/2010 to 31/12/2013. N=28.065 (sample size), which allows the developers to adjust parameters, to control and to analyze the systems predictions.

Requested partner

The ideal partner would be a provider of managerial solutions, in particularly those specialized in offering hospital management ICT tools. The partner would be in charge of commercializing the application under a license agreement. The technology can be marketed as a central service for many client hospitals or as one single solution to each client hospital interested in this predictive technology. The commercializing partner shall then decide which marketing strategy better suits its hospital portfolio. The medical research institute is offering as well a technical cooperation agreement, intended to provide to the commercializing partner the required support in the adjustment and final configuration of the tool, enabling this way the commercializing partner in marketing this technology as a managerial turnkey project.

Cooperation offer is closed for requests