RESUMO
BACKGROUND: Quality assessments are being introduced in many countries to improve the quality of care and maintain acceptable quality levels. In South Korea, various quality assessments are being conducted to improve the quality of care, but there is insufficient evidence on intensive care units (ICUs). This study aims to evaluate the impact of ICU quality assessments on the structural indicators in medical institutions and the resulting in-hospital mortality of patients. METHODS: This study used data collected in the 2nd and 3rd ICU quality assessments in 2017 and 2019. A total of 72,879 patients admitted to ICUs were included during this period, with 265 institutions that received both assessments. As for structural indicators, changes in medical personnel and equipment were assessed, and in-hospital deaths were evaluated as patient outcomes. To evaluate the association between medical staff and in-hospital mortality, a generalized estimating equation model was performed considering both hospital and patient variables. RESULTS: Compared to the second quality evaluation, the number of intensivist physicians and experienced nurses increased in the third quality evaluation; however, there was still a gap in the workforce depending on the type of medical institution. Among all ICU patients admitted during the evaluation period, 12.0% of patients died in the hospital. In-hospital mortality decreased at the 3rd assessment, and hospitals employing intensivist physicians were associated with reduced in-hospital deaths. In addition, an increase in the number of experienced nurses was associated with a decrease in in-hospital mortality, while an increase in the nurse-to-bed ratio increased mortality. CONCLUSIONS: ICU quality assessments improved overall structural indicators, but the gap between medical institutions has not improved and interventions are required to bridge this gap. In addition, it is important to maintain skilled medical personnel to bring about better results for patients, and various efforts should be considered. This requires continuous monitoring and further research on long-term effects.