According to the 2021 National Health Insurance Service data, Korea has 9,174 beds in 863 intensive care units (ICUs), excluding neonatal ICUs. Approximately half of these ICU beds are located in and around the capital, Seoul. The latest report on the Improvement Study on Assessment of the Appropriateness of Intensive Care Units, commissioned by the Health Insurance Review and Assessment Service (HIRA), revealed significant disparities in the quality of care and services provided depending on the region and hospital. Several hospitals reported substandard clinical outcomes, such as ICU mortality, when compared to other hospitals in Korea and international standards [1]. In 2020, HIRA reported an overall ICU mortality rate of 14.2% in Korea [2], which was higher than the 9% rate in Canada [3] and the 11.3% rate in the US [4].

Numerous factors are associated with ICU mortality, such as age, sex, comorbidities, the utilization of mechanical ventilation, renal replacement therapy (RRT), and vasopressors. From a healthcare system perspective, ICU case volume plays a crucial role in patient outcomes, in conjunction with admission route, the degree of critical care nursing, and a multidisciplinary ICU team. Critically ill patients require specialized and intensive treatment/management, which is provided by professionals with expertise in various relevant fields. Common diagnoses in the ICU, such as sepsis, acute respiratory distress syndrome, and circulatory shock, often demand continuous patient monitoring, mechanical ventilation, continuous RRT, and even extracorporeal membrane oxygenation, sometimes all simultaneously. To optimize the management of critically ill patients, intensivists should be well-trained and familiar with the most recent clinical practice guidelines. In order to maintain a consistent and reliable quality of patient care, an efficiently functioning unit with adequate patient throughput is ideal, which can be assessed as case volume. Numerous studies have examined the impact of hospital case volume for various procedures and settings, including high-risk surgical procedures [5], emergent procedures [6], solid organ transplantation [7,8], and trauma [9]. These studies have consistently reported improved patient outcomes with higher institutional case volumes. The precise mechanism underlying this association remains unclear, but it is thought to result from the cumulative experience of providers and selective referral to institutions with better outcomes. Although the reasons behind the association between ICU patient volume and mortality are not fully understood, the majority of studies have reported a positive correlation between institutional ICU patient volume and patient survival.

A PubMed search was conducted using the terms "intensive care," "critical care," "case volume," "outcome," and "mortality" to identify relevant papers on this topic up until February 13, 2023. Additionally, the reference lists of review papers and published systematic reviews were examined to ensure that no pertinent articles were missed during the electronic search.

Tables 1-3 summarized studies that evaluated the impact of ICU case volume on patient outcome in general critically ill patients (Table 1) [10-13], sepsis and septic shock patients (Table 2) [14-19], and mechanically ventilated patients (Table 3) [20-26].

Most studies on ICU patients have suggested that an increase in case volume positively impacts mortality outcomes for patients who are generally ill in the ICU, including those with sepsis, septic shock, and those requiring mechanical ventilation. Several studies have proposed a threshold volume at which the lowest mortality is achieved; however, both below and above this threshold, mortality begins to increase, demonstrating a U-shaped relationship between case volume and mortality. Interestingly, a few countries with different healthcare systems, such as the UK, Australia, and Canada, have not shown similar impacts of volume on mortality improvement.

Traditionally, the volume-mortality effect has been attributed to the accumulation of clinical experience and selective referral. High-volume institutions are more likely to employ experienced personnel who can readily detect patient deterioration and have access to the various resources needed to manage critically ill patients. The presence of trained intensivists and high-intensity ICU staffing has already been associated with improved mortality rates and reduced ICU and hospital lengths of stay [30]. In 2015, Korea implemented a mandate requiring doctor presence and reimbursement incentives for intensivist presence in ICUs at tertiary centers. Prior to this implementation, fewer than one-third of ICUs had intensive care specialists working a 5-day week [31]. However, by 2020, this number increased to 46.7% [2]. A study examining the implementation of intensivists in 441 Korean ICUs found significant decreases in both ICU and 90-day mortality rates (11.7% to 6.3%, respectively, P=0.047; 18.6% to 10.3%, respectively, P=0.012) and a hazard ratio for 90-day mortality of 0.39 (95% CI, 0.23–0.67; P=0.001) [32]. Despite the improved mortality rates following the implementation of intensivist-led teams, the proportion of ICUs with full-time intensivists remains significantly lower than international standards. A 2011 study of ICUs in Asia found that 66% of the 150 ICUs surveyed had 24-hour intensivist coverage [33], while only 26.9% of Korean ICUs in 2020 were fully staffed by intensivists [2]. As a result, discrepancies in clinical outcomes by region and hospital persist in Korea. Depending on hospital volume, tertiary, general, and primary hospitals in Korea have demonstrated 5-year mortality rates for mechanically ventilated patients of 37%, 55%, and 82%, respectively. Hospitals with the presence of intensivists or those located in the capital tend to exhibit lower mortality rates [2].

Selective referral is a phenomenon where patients are more likely to be referred to high-volume centers, particularly for complex surgical procedures, pediatric surgery, and trauma. However, it has been suggested that more complex organizational factors may be associated with improved outcomes in critically ill patients. Factors such as multidisciplinary rounds, ICU nurse staffing and education, the presence of a clinical pharmacist, protocols for weaning and sedation, and a culture of teamwork and communication may all contribute to enhancing ICU patient outcomes [34,35]. Clinical pharmacists can offer valuable consultation in developing plans for drug dosing and monitoring, as well as assisting in medication reconciliation upon patient discharge [36-38]. In a database review of 199,082 Medicare patients treated in 961 hospitals with vancomycin and aminoglycosides, hospitals where pharmacists monitored the use of these drugs demonstrated lower mortality rates, shorter lengths of stay, and fewer adverse events [39].

The design of Korea's healthcare system does not provide adequate reimbursements for managing critically ill patients. When comparing results from different healthcare systems, caution should be exercised, as the impact of relevant factors may be biased or obscured. One prime example is ICU case volume, as the suggested cut-off may differ between densely populated regions or countries (such as Korea) and sparsely populated countries (such as Australia). Measures proven to improve patient outcomes, such as dedicated intensivists, higher nurse-to-patient ratios, and multidisciplinary care teams that include nutritionists, pharmacists, and physical therapists, are often overlooked. The discrepancy in the results of the Australian/New Zealand data [21] from previous studies of volume-mortality association may stem from differences in health system jurisdictions, which was also identified in results from countries in similar situations, such as Canada [24] and the UK [25]. The authors of the 2021 UK study suggest that the unique nature of the US healthcare system, which has complex funding sources compared to the publicly funded healthcare systems of Canada, Finland, or the UK, may contribute to disparities in access to care and thereby lead to weak or nonexistent associations between sepsis patient volume and clinical outcomes [17]. Private-sector US hospitals are primarily funded through fee-for-service and prospective payment charges, which are reimbursed by private insurance companies, Medicare, state Medicaid programs, and other government funds. This contrasts with Canada and other countries that have publicly funded single-payer healthcare insurance. After adjusting for clustering and the procedure or condition studied, the relative odds of Canada showing a significant volume-outcome association compared to the US were substantially low (OR, 0.23; 95% CI, 0.07–0.76; P=0.01) [40].

The observed center volume effect on specific types of surgeries and patient populations, which demonstrates improved patient mortality, has prompted discussions about whether these procedures should be limited to high-volume centers with favorable outcomes [41,42]. The significant volume effect on mortality in solid organ transplantation and trauma centers has led to policy changes requiring minimum case volumes for center accreditation. In some countries, an enhanced patient referral system that clusters low- and high-volume centers in close proximity is already being implemented [16]. However, it remains debatable whether a similar volume effect exists in critically ill patients, as volume effects have shown inconsistent results. A possible explanation for this inconsistency may be that institutions with higher patient volumes are more likely to adopt newly accredited best practices. For example, the previously mentioned study regarding ICU-admitted AECOPD patients [22] showed that units with higher volumes were more likely to implement NIV, which was associated with improved survival rates. When institutions with higher volumes demonstrate better outcomes by implementing new evidence-based practices, it becomes more likely that these practices will be adopted across all hospitals, leading to an attenuation of the volume and outcome association [20,43].

There is considerable diversity among patients admitted to the ICU. The current inconsistencies in the relationship between ICU case volume and outcomes necessitate a multi-layered study focusing on ICU patients with specific diagnoses. Another aspect that warrants a thorough analysis is the intensivist-to-patient ratio. A recent US study on intensivist-to-patient ratios examined the outcomes of 51,656 patients under the care of 246 intensivists. The average caseload was 11.8 patients per day, with no evident association between the intensivist-to-patient ratio and mortality (hazard ratio for each additional patient: 0.987; 95% CI, 0.97–1.01; P=0.2) [44]. In Korea, the average caseload per intensivist was 44.7 patients per day in 2014, which steadily decreased to 22.2 patients per day in 2021 [1]. Further prospective studies are needed to assess the impact of the intensivist-to-patient ratio.

In conclusion, despite a few inconsistencies, a higher ICU case volume has been demonstrated to positively impact mortality rates. Given the significant effect on patient outcomes, institutional ICU case volume should be taken into account when developing healthcare policies related to this area.