Browse Prior Art Database

OptCare - A system for for personalizing clinical care pathways

IP.com Disclosure Number: IPCOM000244979D
Publication Date: 2016-Feb-04
Document File: 12 page(s) / 1M

Publishing Venue

The IP.com Prior Art Database

Abstract

A random forest survival methodology is used to predict the probability that a person will experience an adverse event (such as having to visit the emergency department) at some point in the future, based on historical medical and related data and as a continuous function of time. The historical data is typically from the previous 1 to 3 years and users of the system are typically interested in predictions 3, 6, and 12 months into the future.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 9% of the total text.

Page 01 of 12

OptCare - A system for for personalizing clinical care pathways Sandya Mannarswamy & Shourya Roy, XRCI

Sandya.mannarswamy@xerox.com, Shourya.roy@xerox.com

Context & Motivation

Healthcare industry is undergoing considerable changes driven both by policy administrators and from the grounds up. These include a shift towards value based healthcare from volume based health care where healthcare delivery models are changing from fee for services delivered to payment for Accountable Care [20] (in terms of patient reported outcomes) and wellness. Given that 20% of population consumes 80% of the healthcare budgets during acute illness phases, there is considerable business interest for optimizing the in-hospital health care delivery process. There has been considerable focus on improving healthcare delivery within hospitals by means of analytics and automation solutions. Aiding human judgements in critical care settings by means of automatic solutions is expected to improve patient safety considerably.

Instead of healthcare service providers being paid for the individual services and diagnostic procedures, episode specific bundle payment model has been adopted in order to improve clinical outcome and reduce costs in value based care delivery. One of the major requirements for success in episode based bundle payment models is to identify opportunities for care delivery redesign wherever possible, for improving outcome and reducing cost. There exists considerable variation in outcomes and cost across hospitals in US [22], thereby indicating that there is scope for improving outcome & reducing cost by optimizing the clinical care delivery process. Clinical Care Pathways (CCP) which are defined to execute the steps of care delivery for specific episode bundles, are a potential source for improving clinical outcome and reduce costs. CCPs can be analysed to identify care redesign opportunities.

Problem Description

In lay terms, Care Pathways can be defined as a collection of systematic set of concrete elements associated with the care management protocol for a specific surgical procedure outcome/disease treatment outcome. While standardized generic care pathways are intended to reduce variation in treatment across population, there exists a need for providing individualized targeted personalized treatment based on patient specific factors. Putting the onus on the physician's judgement to adapt the care pathway for each patient based on the patient specific factors puts enormous mental burden on the physician and due to time critical nature of healthcare services, physicians find it difficult if not impossible to adapt the clinical care pathway manually for each patient. Instead they end up following the generic care pathway which (while reducing any treatment variation) fails to tailor the pathway for the specific patient, thereby leading to poor clinical outcomes and increased costs. This brings forth a need to provide automation tools which...