Redefining success: Incorporating long-term survival outcomes into routine benchmarking

1. The importance of long-term survival after critical illness

From the birth of intensive care units (ICUs) during the polio pandemics of the middle of the 20th century, critical care clinicians have focused on the acute derangement of physiology to titrate minute to minute organ support.¹ As a result, there has been an emphasis on short-term mortality as a marker of the quality of clinical care delivered and a standard for benchmarking ICU performance.² Yet there are compelling reasons for ICU clinicians to expand their horizons beyond the walls of the ICU and indeed, beyond the hospital.

With an ageing and increasingly comorbid population, together with increasing ICU survival, it is timely to consider what becomes of patients after ICU and hospital discharge.[3], [4], [5], [6], [7] While ICU and hospital mortality are convenient to measure, they fail to capture the long arc of recovery. There are additional processes above the initial acute insult which plausibly drive increased mortality long after ICU discharge.^2,[8], [9], [10], [11], [12], [13] Indeed, there is a growing body of evidence that surviving a critical illness is only the beginning of a long and often arduous journey.[14], [15], [16], [17] This reflects not only the severity of acute illness but also the compounded effects of factors not routinely measured in hospital mortality risk adjustment scores including frailty, ICU- and health system–acquired complications, cognitive dysfunction, ICU-acquired weakness, social determinants, and persistent inflammation.¹⁸

As such, ICU and hospital mortality offer a limited view.^11,12 Long-term survival—measured in months and years, rather than days and weeks—may offer additional insights into the efficacy and value of intensive care, illuminating patterns that may be invisible in the short term.

The historical emphasis on short-term mortality is being challenged by an initiative from the Australian and New Zealand Intensive Care Society’s Centre for Outcome and Resource Evaluation (ANZICS CORE). Facilitated by a grant from the Intensive Care Foundation of Australia and New Zealand, ANZICS CORE in partnership with the Australian Institute of Health and Welfare and the Aotearoa New Zealand Ministry of Health has now made comparative reports of long-term survival available. The linkage of registry data to a national death index across two countries provides new opportunities for critical care clinicians to evaluate and improve outcomes for ICU patients. This advancement enables the critical care community to redefine excellence in care—by not just who survives to discharge but also who survives in the months and years beyond their critical illness.

2. Comparative reports for all ICUs

The long-term survival reports available through the ANZICS CORE portal (https://coreportal-uat.anzics.org) to all Australian and Aotearoa New Zealand ICUs are presented as Kaplan–Meier plots, displaying the probability of survival up to 3 years after ICU admission. Users can view reports stratified by age or frailty status (Fig. 1) and filtered to diagnostic and other subgroups such as invasive ventilation (Fig. 2). Although an individual ICU’s data can be stratified, filtered, and compared to patients in a variety of hospital and ICU types, the reports are not “risk adjusted” and comparisons should be interpreted accordingly.

Fig. 1.

Screenshot of the ANZICS CORE long-term survival report for “Oldtown Hospital ICU” showing a Kaplan–Meier plot of all patients admitted between January 2020 and December 2023, stratified by frailty status. Time commences at ICU admission. ANZICS CORE: Australian and New Zealand Intensive Care Society’s Centre for Outcome and Resource Evaluation; ICU: intensive care unit; PICU: paediatric intensive care unit.

Fig. 2.

ANZICS CORE long-term survival report for “Oldtown Hospital ICU” showing a Kaplan–Meier plot and survival table estimates for all ventilated patients (excluding cardiac surgery) admitted between January 2020 and December 2023. Time begins at ICU admission. ANZICS CORE: Australian and New Zealand Intensive Care Society’s Centre for Outcome and Resource Evaluation; ICU: intensive care unit.

The information provided is dependent on the integrity and accuracy of the linkage to the death registers which now occurs annually in both countries. Present estimates suggest more than 96% of records are accurately linked, but mismatches and linkage failures can occur. In addition, linkage methods vary slightly between Australia and Aotearoa New Zealand. Delays in registering a death can also vary between jurisdictions, with the most common example being patients for whom the death is undergoing coronial investigation. Foreign nationals, migrants, and admissions where an individual’s identity was uncertain also influence linkage success.¹⁹

This first iteration of the reports provides long-term survival estimates based on approximately one million adult patients who have had over 1.3 million admissions to 181 ICUs in Australia and 20 ICUs in Aotearoa New Zealand since 2018. At the time of writing, data have been linked up to March 2024 for Australian and June 2024 for Aotearoa New Zealand patients. Admissions after these dates are still included but only up to their discharge from hospital. A version for paediatric ICU admissions will also be available in later updates.

3. Implications

The introduction of long-term outcome benchmarking provides a new lens for measuring and improving performance. Hospitals and ICUs will be able to compare their long-term survival rates against their peers. Variations between units can be quantified, reported, and used to inform quality improvement initiatives.

Until recently, the ability to investigate long-term outcomes had been limited to the area of clinical research and had been time- and labour-intensive. ICU care is resource-intensive and often invasive. In an era increasingly defined by value-based care, it is insufficient to focus solely on immediate survival. Long-term survival offers a more holistic and ethically sound quality metric. The ability to benchmark long-term outcomes against ICUs with similar patient demographics and case-mix offers an additional nuance to traditional benchmarking.⁵

The increased knowledge that is now in the hands of bedside clinicians empowers improved communication with patients and their families about the likely long-term outcomes of critical illness and can be used as the basis for ongoing discussion. This is especially so for high-burden interventions, and for patients with a prolonged ICU length of stay, shifting the focus from short-term survival to meaningful and sustained recovery.

Similarly, for clinicians and policy makers, information about long-term survival may inform decisions around interventions in the post–hospital-discharge period, particularly for those in whom long-term outcomes are worse than expected. Interventions may include improved interfaces with primary health care, follow-up clinics, increased access to rehabilitation, or perhaps palliative care integration.

While the ability to benchmark long-term outcomes represents a leap forward, its implementation will require careful interpretation and education. It is essential that critical care clinicians, their ICUs, and policy makers not only apprecaite the data and its limitations, but also translate this understanding into action - balancing performance evaluation with meaningful clinical change.²⁰ There are a number of limitations to the use of survival plots, not the least of which is that the plots are not (currently) risk adjusted, with implications for ICUs with different case-mixes or patient demographics. In addition, the ICU has little control over factors beyond hospital discharge such as access to rehabilitation, socioeconomic status and social determinants of health, and primary care access. It is also likely that attribution of death to the critical illness becomes increasingly tenuous over time.^18,21

Furthermore, it is important to acknowledge that survival is not the only endpoint and that for some patients, it is not even the most important.²² The addition of routinely collected patient-reported outcome measures, supported by surveys that describe the family experience of having a loved one in the ICU, will soon be available to add nuance to the long-term survival data now available.⁵

Evidence suggests that ICU outcomes in Australia and Aotearoa New Zealand are among the best in the world (Fig. 2). The addition of routine reporting of long-term outcomes allows the Australian and New Zealand critical care clinicians to continue to benchmark their outcomes in a transparent manner.²³

4. Conclusion

The availability of long-term survival outcomes provides a comprehensive understanding of patient recovery beyond the short-term metrics of hospital mortality. These data are crucial for identifying factors that influence long-term survival, informing targeted interventions, and improving ICU care. By focusing on long-term outcomes, it is possible to better assess the true efficacy and value of intensive care, ensuring that the care provided not only saves lives in the short term but also contributes to meaningful, sustained recovery. This approach aligns with the ethical obligation to provide care that is life enhancing and reflective of what patients and their families value. The integration of long-term survival outcomes into routine practice represents a significant advancement in critical care, offering new opportunities for quality improvement and patient-centred care.

Author Contribution

Paul Secombe: conceptualisation, Writing – Original Draft Preparation. Ed Litton: Funding acquisition, conceptualisation, Writing – Review and Editing. Shaila Chavan: Funding acquisition, Visualisation, Writing – Review and Editing. Jennifer Hogan: Visualization, Writing – Review and Editing. Sue Huckson: Writing – Review and Editing. Tamishta Hensman: Writing – Review and Editing. Chong Tien Goh: Writing – Review and Editing. Craig Carr: Writing – Review and Editing. Jason McClure: Writing – Review and Editing. David Pilcher: Funding acquisition, conceptualisation, Writing – Review and Editing.

Funding

Funding for the inititial data linkage project was provided through a competitive grant provided by the Intensive Care Foundation.

Conflict of interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Ed Litton and David Pilcher report financial support was provided by Australian and New Zealand Intensive Care Foundation. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.