Rinaldo’s role in the medical emergency team and rapid response systems

Abstract

In the 1990s, there was emerging evidence that patients admitted to hospitals frequently suffered in-hospital cardiac arrest, unplanned admission to the intensive care unit (ICU), and potentially preventable in-hospital death. These events were often preceded by objective signs of instability and suboptimal recognition and response by hospital ward staff. Rinaldo Bellomo collaborated with key Australian and international leaders to develop a novel and paradigm-shifting model of care referred to as the medical emergency team (MET). This team is comprised of senior staff members who are experts in the assessment and management of acutely deteriorating patients.

In Australia and New Zealand, staff members from the ICU are frequently the team leaders for the MET. The team is called when a patient develops objective signs of clinical deterioration, prior to the onset of cardiac arrest. Rinaldo led, mentored, and supervised a systematic and structured research program that evaluated the nature and effectiveness of the MET at Austin Health and throughout Australia. This commenced with single-centre before-and-after studies and progressed to the first Australian ICU cluster-randomised controlled trial. His unique skillset was pivotal in the emergence and promulgation of this model of care worldwide resulting in countless lives saved from preventable morbidity and mortality.

1. The background and leadup to the medical emergency team model of care

The conception and development of the medical emergency team (MET) model of care was the culmination of accumulating evidence for adverse events in hospitalised patients and the meeting of key individuals who were working in this area in different parts of the world (Table 1).

Table 1.

Major milestones in the history of medical emergency teams and the rapid response system.

Date	Milestone
1984	Ken Hillman first discussed the concept of the “MET system” at Charing Cross Hospital in London
1989	Ken Hillman commenced the MET in Liverpool Hospital.
Jul 1992 to Jun 1993	Rinaldo worked with Michael DeVita at University of Pittsburgh hospital as an ICU fellow, including participating in “Condition C”—A response to expedite ICU admission of unwell patients outside the ICU.
Aug 1994	Rinaldo commenced employment at Austin Health.
1995	Ken published the concept of the MET Wilson study on quality and safety in health care published
1997	Mike DeVita modified “Condition C″ response at Pittsburgh to send ICU staff to ward to commence treatment. Michael Buist introduced MET into Dandenong Hospital.
Nov 1998	Donna Goldsmith commenced employment as the research coordinator at Austin Health.
Dec 1998	Commenced recruitment for study of postoperative adverse events.
May 1999	Commenced recruitment for study of IHCAs. 4 bed post-operative HDU opened at Austin Health
Sep 1999	Approval for commencing the MET. Commenced a 1-year period for education and preparation for introduction of the service.
Nov 2000	MET implemented at Austin Health using objective criteria for escalation.
2002	Before-and-after study of Austin MET on postoperative complications published in Medical Journal of Australia.
2003	Before-and-after study of Austin MET effect on IHCAs published in Medical Journal of Australia. Commenced planning for MERIT study.
2004	Discussions and planning for first textbook and conference related to RRT.
June 2005	First international MET Conference, Pittsburgh, USA. First textbook on METs published.
2006	First international consensus statement on RRSs published
2006	Discussions with Joint Commission, USA
2010	2nd consensus conference on RRS afferent limb.
2011–2015	Supervised PhD on RRS uptake, resource implications and patient characteristics.
2014–2018	Member of advisory committee for international society for RRSs.

HDU = high-dependency unit; ICU = intensive care unit; IHCA = in-hospital cardiac arrest; MET = medical emergency team; MERIT = Medical Emergency Response Intervention Trial; RRS = rapid response system; RRT = rapid response team; USA = the United States of America.

Multiple observational studies were published in the 1990s that focused on the frequency, nature, and antecedents to serious adverse events (SAEs) for hospitalised patients. These studies were conducted in multiple countries and reported that SAEs occurred in up to 16.6% of hospitalised patients and were associated with death or permanent disability in approximately a quarter of cases.1, [2], [3], [4], [5]

Other investigators studied the period leading up to in-hospital cardiac arrest (IHCA), unplanned intensive care unit (ICU) admission, and in-hospital death and found that up to 80% had objective signs of clinical deterioration that were present for many hours prior to the event.[6], [7], [8], [9], [10] Such studies also showed that the escalation and response to this deterioration was variable and often was not commensurate to the needs of the patient.^6,9

As early as 1984, Ken Hillman had proposed the concept of the MET at Charing Cross Hospital in London. However, he was not able to implement the model until 1989 when he returned to Liverpool Hospital in Sydney. Ken et al. published the original description of the MET in 1995,¹¹ the same year that Wilson et al. published a study reporting that among more than 14,000 hospital admissions in 28 hospitals, SAEs occurred in 16.6% of instances.⁵

In 1992, Rinaldo worked at the University of Pittsburgh hospital as an ICU fellow with staff specialist Michael DeVita, where he was involved in the “Condition C″ response to expedite ICU admission of unwell patients outside the ICU. This model of care was subsequently refined in 1997 to expand the role of ICU staff to be involved in the assessment and management of acutely unwell patients outside of the ICU and not simply their retrieval to the ICU.¹² In 1997, Michael Buist introduced an MET into Dandenong Hospital in Victoria, using the activation criteria published from Liverpool Hospital.¹³

2. The MET at Austin Health

Preparation and planning for the MET at Austin Health commenced in Nov 1998 when Donna Goldsmith commenced employment as a research nurse. At her interview, Rinaldo asked her “Are you ready to do what it takes to change the lives of people who you will never meet?” Data on postoperative SAEs were collected between Dec 1998 and March 1999. Following 1125 major surgical procedures, 16.9% developed SAEs and 80 patients died (7.1%). The most common of these predefined SAEs were emergency admission to the ICU, respiratory failure, and readmission to the ICU.¹⁴

In early 1999, a four-bed high-dependency unit was opened and colocated within the ICU to increase surgical throughput and early postoperative care. However, this process was associated with neither a reduced risk of in-hospital death nor a mean duration of hospital stay.¹⁵

In September 1999, the hospital executive approval was provided to begin planning for the introduction of the MET. The preparation process for this novel model of care took 12 months, multiple meetings to address concerns of key stakeholders, education of all hospital staff about the new system, preparation for the hospital switchboard to take the calls, and development of the criteria for activation. Importantly, the MET calling criteria contained a “staff-worried” criterion to empower bedside clinicians to call for assistance to any patient they were concerned about, irrespectively of the presence of deranged vital signs. Prospective data on IHCAs were also collected during this period.

The ICU-led MET service commenced operation in November 2000. In a study of more than 2600 major surgical cases, implementation of the MET was associated with a 57.8% relative risk reduction in postoperative adverse outcomes including respiratory failure, stroke, severe sepsis, and acute renal failure requiring renal replacement therapy. There was also a 44.4% relative risk reduction of unplanned ICU admissions and a 36.6% relative risk reduction in the risk of death.¹⁶

In a separate before-and-after study, the introduction of the MET was also associated with a 65% relative risk reduction in the risk of IHCAs, along with improved in-hospital survival and reduced ICU and hospital bed days among IHCA survivors.¹⁵

Rinaldo went on to supervise several studies of the medium-term effects of the MET at Austin Health. These included a before-and-after study revealing that the introduction of the MET was associated with improved long-term survival of patients undergoing major surgery compared to the control high-dependency unit period.¹⁷ In addition, the benefits of the MET in reducing IHCAs were not only sustained but also further decreased out to 4 years.¹⁸ Additional studies included analysis of circadian variation of MET calls and cardiac arrest detection, a pilot study of the role of the MET in end-of-life care,¹⁹ and descriptions of “MET syndromes”²⁰ which would later become the basis of an MET training programme.

3. The concept of rapid response systems and their promulgation worldwide

Despite the positive findings from Dandenong and Austin hospitals, there were concerns about the effectiveness and implications of the MET in Australia and the USA.²¹ Major concerns included the risk of bias with the before-and-after study design²² and the risks of deskilling ward staff.²³

Accordingly, Rinaldo, Ken Hillman, and others commenced planning on the Medical Emergency Response Intervention Trial (MERIT) study which was published in The Lancet in June 2005. This was a cluster-randomised control trial that randomised 23 Australian hospitals to continue functioning as usual (n = 11) or have the implementation of an MET (n = 12). Introduction of the MET was associated with increased incidents for emergency calls but was not associated with a reduction in the primary or secondary outcomes which included IHCAs, unplanned ICU admissions, and unexpected in-hospital death.²⁴ Although this was not a positive trial, it was groundbreaking in terms of the conduct of the first cluster-randomised control trial conducted by ICU researchers within Australia. A post hoc analysis published later showed a linear relationship between the frequency of MET calls and the decrease in IHCA.²⁵

In 2005, Mike DeVita, Ken Hillman, and Rinaldo edited the first book on rapid response systems (RRSs) and with Michael DeVita, held the first international conference on METs in June 2005 (Fig. 1). Michael DeVita also led the first international conference on METs which was published in 2006.²⁶ This consensus conference resulted in the development of the concept of the RRS which describes the four arms of a hospital-wide system to improve the recognition, response, and overall care for deteriorating hospitalised patients: (i) an afferent limb to identify the crisis and to escalate care; (ii) an efferent limb—the MET or rapid response team; (iii) a governance limb to provide administrative oversight; and (iv) a quality improvement limb. In 2006, they also began discussions with the USA-based Joint Commission on the topic. This ultimately led to the Institute for Health Improvement including rapid response teams as an essential component for their 100K lives campaign.

Fig. 1.

Rinaldo presenting at the first conference on medical emergency teams in June 2005. Credit: Edward Rieker.

In 2010, Rinaldo participated in a second consensus conference focusing on the afferent limb of the RRS.²⁷ This meeting saw the introduction of many of the future world leaders in the RRS research community.

Between 2011 and 2015, Rinaldo supervised Daryl Jones in a PhD that evaluated the update resource implications and patient characteristics of the RRS in Australia and New Zealand. Rinaldo served on the international society for the RRS as an advisory committee member between 2014 and 2018 and spoke at several conferences for this society and for the Australian and New Zealand Intensive Care Society on the topic of deteriorating patients and the MET.

4. Rinaldo’s attributes and his enduring legacy to patient safety

While Rinaldo did not conceive the MET concept, he embarked on evaluating the nature and effectiveness of this paradigm-shifting model of care in a systematic and rigorous manner. In an 8-year period, he went from collecting baseline data needed to implement the MET into Austin Health to helping lead the first cluster-randomised trial in Australian ICUs, participating in the first international RRS conference, and advising the Joint Commission on this model of care. His leadership and vision resulted in getting the key leaders working together to develop and evolve the MET into a hospital-wide approach of RRSs worldwide.

His unique skill set allowed him to overcome cultural and logistic barriers and supervise and mentor novice researchers who have gone on to conduct independent research in this space. His gentle and persuasive approach permitted the spread of this novel and disruptive concept.

In Australia and New Zealand, the RRS is a ubiquitous model for recognising and responding to clinical deterioration in hospitalised patients. It has enhanced the profile of the specialty of ICU locally and resulted in the development of consensus statements and conferences in this space. There are now four systematic reviews demonstrating that the introduction of an RRS is associated with reductions in the risk of IHCA and all-cause hospital mortality.[28], [29], [30], [31] At the Austin Health, the risk of IHCA is now 90% lower than it was before the implementation of the MET.³²

It is unlikely that this model of care would have evolved so rapidly or been so successful without Rinaldo’s involvement. As was the case for many other areas of his research, he did “what it takes to changes the lives of people who he would never meet”.

With the evolving profile of patients who are admitted to hospital and the emergence of big data and enhanced analytic techniques, there are opportunities to improve the understanding of and care for patients who are at a risk of clinical deterioration (Table 2). There is also a need to improve training and education, for not only the MET responders but also ward-based staff for the management of clinical deterioration in the period before patients need an MET call.

Table 2.

Future research and quality improvement opportunities in the area of the medical emergency tams (METs).

1. Can big data be used to a. identify the optimal MET calling criteria? b. predict which patients are at risk of clinical deterioration and needing MET review? c. evaluate which patients are at risk of adverse outcomes following MET review?

2. There is a need for an MET registry to permit a. benchmarking of MET patient outcomes between peer hospitals b. identification of areas for future research and quality improvement initiatives

3. Can the care of MET calls be optimised via a. introduction of guidelines for management of sepsis and other common conditions seen in MET call? b. establishing criteria for escalation to more senior staff?

4. Improved training and education a. Enhancing the ability for ward staff to manage early clinical deterioration and conduct goals of care discussion to prevent MET calls b. Developing and promulgating programs for interprofessional training of MET members

5. Evaluating the optimal use of technology including the a. role of continuous monitoring outside of critical care areas b. use of clinical alerts as decision support tools c. use of automated notification of staff regarding clinical deterioration and MET calls

MET = medical emergency team.

Author Contribution

Conceptualization: Daryl Jones; Methodology: All authors; Software: N/A; Validation: All authors; Formal analysis: N/A; Investigation: All authors; Resources: N/A; Data Curation: N/A; Writing - Original Draft: All authors; Writing - Review & Editing: All authors; Visualization: All authors; Supervision: N/A; Project administration: N/A; Funding acquisition: Unfunded study.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflict of interest

Not of the authors have finanical or personal relationships that could be considered a conflict of interest.