Clinical Informatics needs to be a competency for Intensive care training

Abstract

Clinical informatics is a cornerstone in the delivery of safe and quality critical care in Australia and New Zealand. Recent advances in the field of clinical informatics, including new technologies that digitise healthcare data, improved methods of capturing and storing these data, as well as innovative analytic methods using machine learning and artificial intelligence, present exciting new opportunities to leverage data for improving the delivery of critical care and patient outcomes. However, ICU training in Australian and New Zealand does not adequately address capability gaps in this area, potentially leaving future intensivists without the necessary skills to provide leadership in the application of informatics within ICUs. This highlights the need to examine how competency in clinical informatics can be incorporated into ICU training, potentially through a range of activities such as curriculum redesign, the formal project, and workshops or datathons. Further work to identify relevant informatics competencies and methods to develop and assess these competencies within ICU training is needed.

1. Introduction

It is your first week as a new consultant when the intensive care unit (ICU) director calls you into their office: your ICU has been identified as an outlier in routine benchmarking, falling well outside the funnel plot confidence interval. The ICU director needs someone to look into this and asks if you can help—you take up the challenge, but leave the office wondering—how do you start?

The growth in health technologies that digitise clinical data, the means to store these data, as well as analytical techniques to glean insights from these datasets has led to a growing role for clinical informatics in health care. Clinical informatics refers to the application of information and computer science to improve clinical care,¹ and is an increasingly vital component in the delivery of safe and effective health care in Australia and New Zealand.

Clinical informatics is not new to the ICU but increasingly supports a range of activities, including benchmarking, quality improvement, research, and public health policymaking. Despite its importance, there are significant gaps in the College of Intensive Care Medicine (CICM) of Australian and New Zealand's training program in this area, particularly in developing competencies that enable ICU clinicians to leverage clinical informatics for practice and quality improvement^2,3

We provide a rationale for the inclusion of clinical informatics competencies in the CICM training program and suggest ways forward to achieve this.

2. Informatics is a cornerstone for safety and quality improvement

Informatics is becoming established across industries as a driver for process improvement, including in health care. A range of metrics currently inform the safe and effective delivery of ICU care, and a basic understanding of informatics is needed to fully realise the benefits that data provide.

Within Australian and Aotearoa/New Zealand ICUs, benchmarking measures such as the standardised mortality ratio and central line–associated bloodstream infection rates are routinely reported. The Australian and New Zealand Intensive Care Society (ANZICS) Centre for Outcome and Resource Evaluation (CORE) outlier programme, designed to identify and support ICUs with higher complication and adverse outcome rates, has been demonstrated to be a highly efficient approach and is associated with improved outcomes for patients.⁴

More recently, data routinely collected by ANZICS supported federal responses to the COVID-19 pandemic. Resource data identified limitations, particularly around ventilator capacity, and led to a focused effort towards procurement of additional ventilator equipment.⁵ The Critical Health Resource Information System, developed in partnership with Telstra and Ambulance Retrieval Victoria, also provided real-time visibility of strain and demand, enabling targeted load balancing across facilities.⁶

3. New sources of data are available

Application of new technologies has also facilitated the digitisation of healthcare data. These include technologies that capture clinical information such as electronic medical records as well as technologies that capture data that are previously collected in an analogue format in an electronic format (e.g., radiology images and clinical messaging systems). The digitisation of various types of clinical data, along with the ability to store in them, allows novel insights into a range of daily activities that previously were unavailable.

An understanding of these data sources and the types of data they generate is essential to recognise their potential, and well as limitations, in terms of how they can support clinical quality improvement.⁷ For example, process mining using time-stamped data can be used to measure efficiencies of care and inform quality improvement.^8,9

4. New analytic methods are increasingly common

Over the last decade, there has been a surge in interest around the use of artificial intelligence (AI) and machine learning (ML) for improving processes across a range of industries.¹⁰ Application of AI and ML to the health sector is lagging, but there is an expectation that these technologies will be progressively implemented, with increasing investment in AI/ML research and development.¹¹ This trend has been reflected by an increasing number of AI/ML publications in the critical care literature,¹² including models developed for the Australian and New Zealand context.^13,14 However, the complexity and sophistication that makes AI/ML useful also make these technologies nonintuitive for many clinicians.

Future intensivists, in their role as clinician leaders, will play a pivotal role in deciding how and when these technologies should be implemented and applied in the clinical environment. Clinical appraisal of these complex systems in turn requires basic understanding of the principles of clinical informatics, as well as an appreciation of the unique challenges and issues. Intensivists will be needed to play a key role in mitigating risks associated with these technologies, particularly bias and generalisability of the output of AI/ML models,¹⁵ and implementation issues that may contribute to clinician burnout and workload.

5. Gaps in ICU training for informatics capability development

Building informatics capability across the general healthcare workforce has been recognised as a key priority by the Australian Institute of Digital Health,¹ with corresponding engagement with a range of disciplines including nursing and midwifery¹⁶ and allied health communities.¹⁷ This growing need requires multidisciplinary engagement from all health professional groups, including front-line ICU physicians, to set the direction in this emerging discipline.

Despite the growing role and potential of informatics, there is limited emphasis on the CICM training program.³ Within the context of ICU training, clinical informatics is not addressed specifically, leaving future ICU clinician leaders bereft of the requisite skills to leverage the benefits of informatics. Of note, there have only been two fellowship examination questions related to these topics in the last decade.^18,19

This gap in knowledge between technological capability and clinical practice is a barrier to the safe and effective translation of basic electronic health data into the ICU environment, let alone advances in AI/ML, and may lead to missed opportunities in improving the delivery of care to patients.²⁰

6. Next steps

Considering the competing demands of ICU training, prioritisation of specific informatics competencies is needed. We propose leveraging the Australian Health informatics Competency Framework,¹ aligning it with the competencies required for intensive care medicine as outlined by the CICM.³ In particular, clinical informatics competencies synergise with the Manager (Leadership) and Scholar (Educator) domains of the CICM ICU competencies framework. We propose that an expert trainee can describe the utility and limitations of clinical informatics. Enabling competencies could include

• -Appraising the value of new data sources and emerging technologies and their relevance to critical care
• -Analysing the clinical safety risks associated with health information and technologies in critical care
• -Describing methods of risk mitigation in relation to the analysis and interpretation of data

We further suggest leveraging the existing CICM framework for teaching and learning opportunities, to guide development for relevant competencies. These may include

• 1.Experiential learning with clinical informatics fellowships or rotations, with consideration of this as electives with value towards ICU training
• 2.Courses or learning packages—developing training materials, workshops, and events that provide hands-on opportunities to use data for addressing key clinical safety and quality questions and ensuring that these are supported by both the college and peak bodies (e.g., trainee participation in critical care datathons²¹)
• 3.Formal project—encouraging projects that are multidisciplinary and/or make better use of pre-existing data collections such as registry data held by the ANZICS CORE.

A range of assessment methods will need to be developed, including incorporating informatics-style questions into the examination, such as addressing interpretation of quality improvement–type data, or critically appraising clinical research that applies advanced analytic methods.

Recognising the rapidly developing nature of the field of informatics, we also advocate for incorporating informatics into the specialist continuing professional development program. This would support Australian and New Zealand ICU specialists in developing and maintaining the requisite skills to use clinical informatics to improve the delivery of ICU services.

Summary

Clinical informatics is a rapidly growing field, presenting numerous opportunities for improving safety and quality in critical care. Incorporating a basic informatics competency into the ICU training program equips clinical leaders to realise the potential of informatics in critical care for improved patient outcomes. Further work is needed to identify informatics competencies of priority and develop appropriate training methods and assessment strategies relevant to ICU training.

Credit authorship statement

ST – conceptualisation, writing (original draft), writing (review and editing).

TE – conceptualisation, writing (review and editing).

TH – conceptualisation, writing (review and editing).

MD – conceptualisation, writing (review and editing).

PS – writing (review and editing), Supervision.

DP – writing (review and editing), Supervision.