Embracing digital disruption using intelligent decision support for more responsive care, better patient outcomes and greater efficiency

Flagship Program: Intelligent Decision Support to Improve Value and Efficiency

Project Description

This (approximately) three year program of research includes seven objectives that are important for laying the foundation for implementing intelligent decision supports that promote responsive, safe and efficient patient care. It will include important use cases and produce implementation frameworks supported by empirical data that can be used by others in the future who seek to develop and implement intelligent decision supports in hospital and health services. The project includes key partners in the intelligent decision support field: Metro South Health (Queensland) and Queensland University of Technology. This program of research will demonstrate the benefits of digital hospital system(s) in Metro South Hospital and Health Service initially using ‘offline’ records for new analytics and decision support solutions, but also evaluating the implementation of real time analytics and decision support in the health service. The program of research will also include: identifying highest priority clinical areas for implementation in a large teaching and training hospital (Princess Alexandra) and smaller hospital facilities in Metro South Health; investigate the identification of clinical teams or service areas that are well prepared and willing to engage in the use of intelligent decision support solutions to address one or more high priority problem areas; demonstrate the extent to which machine learning can be used to predict key problem area events and produce actionable intelligence to guide clinical decisions if implemented in real-time; identify clinician preferred approaches for receiving and interacting with intelligent decision supports in real-world contexts; produce economic models to understand the potential cost-effectiveness (including effects on patient outcomes, and relevant health service outcomes) of potential live implementation of intelligent decision supports; and produce a prioritised list of opportunities for intelligent decision supports to improve patient outcomes and improve hospital efficiency to be carried forward for further live testing (in a subsequent project(s) to follow this one).

Project Objectives

Aims:

  1. Understand stakeholder ‘high value problem’ priorities for the use of intelligent decision supports to address common, impactful problems in hospital environments.
  2. Evaluate the implementation of, and staff engagement with, intelligent decisions supports that have been (or will be) implemented within the health service’s digital hospital system, including but not limited to those with potential for integrated application of artificial intelligence for supporting decision making.
  3. Investigate contextual factors (including human resource-related factors) that are likely to contribute to success or failure of the implementation of intelligent decision supports to improve patient outcomes and improve hospital efficiency.
  4. Investigate how machine learning or other appropriate analytics can be best used to predict important events or identify opportunities for clinical care enhancements to help improve important patient and health service outcomes.
  5. Understand key implementation-related barriers and facilitators, as well as how to best operationalize decision support information in a way that will engage clinicians and promote responsive clinical care that leads to better patient outcomes.
  6. Investigate the likely cost-effectiveness of intelligent decision supports for impacting on the ‘high value problem’ area(s) identified in Aim 1 or Aim 2 for the benefit of patients and health services.

Note: there is scope for these aims to be expanded or amended if additional partners join this project after commencement.

Objectives:

  1. Identify the highest priority ‘problem’ areas for the application of new analytics for intelligent decision supports to help clinical teams improve patient and health service outcomes (e.g., proactive and responsive patient-centred care to prevent hospital acquired complications including falls and pressure injuries; improving patients’ journey (and flow / reduce ‘bed block’); prevent health service waste (e.g., non-attendances at scheduled appointments or procedures, wasted bed-days, unwarranted clinical variation/ low benefit care etc.)).
  2. Estimate the effect of implementation and staff engagement with intelligent decisions supports that have been (or will be) implemented within the health services’ digital hospital system.
  3. Identify clinical teams or service areas that are well prepared and willing to engage in the use of intelligent decision support solutions to address one or more high priority problem areas (identified in Objective 1 or 2). Page 3
  4. Demonstrate the potential for machine learning (or other appropriate analytic approaches) for identifying problem areas (identified in Objective 1) and produce actionable intelligence to guide clinical decisions for actions that can be implemented in real-time.
  5. Identify the approaches most preferred by clinicians / health service staff for receiving / viewing / interacting with intelligent decision support information that are likely to overcome identified barriers in order to operationalize actions with potential to improve care for patients.
  6. Produce economic models to understand the potential cost-effectiveness (primarily patient outcomes, but also relevant health service outcomes) of potential live implementation of intelligent decision supports to address aforementioned identified ‘problem’ areas (objective 2) for informing investment (or disinvestment) decisions)
  7. Produce a prioritized list of opportunities for intelligent decision supports to improve patient outcomes and improve hospital efficiency to be carried forward for further live testing (in subsequent project).

Industry Participant

Metro South Hospitals & Health Service (Queensland)
Cameron Ballantine, Chief Information Officer & Stephen Canaris, Director BI & Analytics, Clinical Informatics, Metro South Health


Other Project Participants


Research Participant

Queensland University of Technology (QUT)
Prof Steven McPhail, DHCRC FRED & Academic Director of Australian Centre for Health Services Innovation

Project Value:

$2,653,000