Evaluation of the Candida score as a mortality predictor in hospitalized patients with Candida spp. colonization: A retrospective cohort study

Invasive candidiasis has risen alarmingly over the last decades, with diagnosis often complicated by colonization at multiple body sites. The Candida score offers a simplified tool to help identify patients at risk of invasive candidiasis, particularly in intensive care unit (ICU) settings; however, its applicability in patients with broader characteristics remains uncertain. Moreover, research on Candida colonization and the utility of such tools in Indonesia is still limited. This study aimed to evaluate the association between clinical and laboratory characteristics and patient outcomes, while also assessing the predictive value of the Candida score in both ICU and non-ICU patients at a tertiary academic hospital in Yogyakarta, Indonesia. This is an observational retrospective cohort study. During the period of January 2019 to December 2021, recorded laboratory data and clinical characteristics of patients (ICU and non-ICU patients) whose specimens yielded Candida spp. were collected through electronic health records. The correlation of laboratory profile and clinical characteristics with the outcome was analyzed comparatively using bivariate and multivariate analysis. Ward status, length of stay, number of comorbid, history of steroids, Candida score of ≥3, and mechanical ventilation were significantly correlated with the non-survivor group. Multivariate analysis showed that intensive care status and a Candida score ≥3 were found to have the strongest and independent correlation with the non-survivor group (p ≤ 0.001; OR 11.8 [95% CI, 4.368-31.962] and p = 0.004; OR, 3.9 [95% CI, 1.562-9.065], respectively). Being in the ICU and having a Candida score of ≥3 was independently correlated with poorer clinical outcomes in hospitalized patients with Candida spp. colonization.