ICU performance level before pandemic impacts COVID-19 mortality

Key takeaways:

• The performance level of an ICU prior to COVID-19 impacted mortality in 2020 and 2021.
• Fewer patients died in more efficient vs. less efficient hospitals.

Patients with COVID-19 admitted to an ICU classified as less efficient before the pandemic faced a higher likelihood for mortality than those in more efficient ICUs, according to results published in CHEST.

“Results suggest that focusing on improving ICU performance (by enhancing organizational aspects) is associated with better outcomes and should be a part of the quality improvement process and preparedness for future pandemics,” Leonardo S.L. Bastos, PhD, assistant professor in the department of industrial engineering at Pontifical Catholic University of Rio de Janeiro, and colleagues wrote.

Mortality outcomes

In a multicenter cohort study of 35 hospitals in Brazil, Bastos and colleagues evaluated 35,619 adults (median age, 52; 40% women) in the ICU with COVID-19 between February 2020 and December 2021 to determine if high-performing ICUs sustained this level of performance during the COVID-19 pandemic in terms of in-hospital mortality.

Of the total cohort, 7% received invasive mechanical ventilation when admitted, and death was more prevalent among these patients vs. those who did not receive mechanical ventilation (54% vs. 13%).

Mortality across the studied hospitals ranged from 3.6% to 63.2% (median, 11%) following adjustment for severity and case-volume, according to researchers.

Using the standardized mortality ratio (SMR) and standardized resource use (SRU) to figure out performance levels of each ICU prior to COVID-19, researchers identified 12 hospitals as the most efficient (SMR and SRU < medians), 13 as the least efficient (SMR and SRU medians) and five hospitals each as either overachieving (SMR < median, SRU median) or underachieving (SMR median, SRU < median).

A little less than half (40%) of patients from the total cohort received care at the most-efficient ICUs, whereas nearly a quarter (24.5%) received care at one of the least-efficient ICUs. The number of patients admitted to an overachieving (17.8%) or underachieving (17.4%) ICU was comparable.

Notably, researchers found the highest adjusted in-hospital mortality rates in underachieving (14.8%) and the least-efficient (12.4%) ICUs, with lower rates in overachieving (7%) and the most-efficient (6%) centers.

Researchers further evaluated hospital performance based on variable life-adjusted display (VLAD), a measure of the accumulated sum of adjusted risk, whereby increasing values indicate risk-adjusted improvement in a system’s mortality risk.

The most-efficient ICUs had VLAD values that exceeded the median 18% of all weeks. In contrast, underachieving centers had values below the median for 66% of all weeks, as did the least efficient for 84% of weeks.

When evaluating this measure against the progression of the pandemic, researchers found only least-efficient or underachieving centers had a reduced VLAD in 2020 when cases started to rise. However, ICUs of all performance levels demonstrated reduced VLAD at the start of 2021, representing the second surge of COVID-19.

Recovery of VLAD occurred quicker among the most-efficient and overachieving ICUs vs. the least-efficient and underachieving ICUs, according to researchers.

Associations with mortality

After accounting for patient characteristics in a mixed-effect logistic regression model, researchers observed elevated odds for mortality among those treated in centers labelled as the least efficient before the pandemic (OR = 2.3; 95% CI, 1.45-3.62).

Heightened mortality odds among patients treated in these centers were also found when assessing only those who initially received noninvasive respiratory support (OR = 2.16; 95% CI, 1.22-3.59) and those who received invasive mechanical ventilation (OR = 2.12; 95% CI, 1.22-3.68).

Using the most-efficient ICUs as a reference, researchers did not find a difference in the odds for mortality among those who went to an overachieving (OR = 1.66; 95% CI, 0.92-2.98) or underachieving (OR = 1.73; 95% CI, 0.97-3.1) center.

However, patients who initially received noninvasive respiratory support had an increased likelihood for mortality if they were admitted to an overachieving vs. most-efficient center (OR = 1.73; 95% CI, 0.97-3.1).

Researchers also looked at the odds for mortality by year and found a drop in the likelihood for this outcome in the least-efficient vs. most-efficient centers from 2020 (OR = 2.47; 95% CI, 1.54-3.96) to 2021 (OR = 2.16; 95% CI, 1.26-3.7).

One limitation of this study is that the findings may not speak for ICUs in other parts of the world because it was conducted in a single private hospital network in Brazil.

“However, if such variability in outcomes can be observed in a single health system, then it is plausible to assume that the patterns encountered in other health systems and countries present even more heterogeneity and, thus, opportunities for future studies and quality improvement strategies,” Bastos and colleagues wrote.

Reference:

• High-performance ICUs reduce mortality rates during pandemics and other health crises. https://en.idor.org/high-performance-icus-reduce-mortality-rates-during-pandemics-and-other-health-crises/. Published Oct. 19, 2023. Accessed Oct. 19, 2023.