Conditional logistic regression.

The clinical features of COVID-19 are vary widely, ranging from asymptomatic states or mild upper respiratory tract infections to severe pneumonia. Previous studies have shown that 20.0% of COVID-19 patients are hospitalized, out of which 10.0–20.0% are admitted to the Intensive Care Unit. The present study aims to assess predictors associated with COVID-19 leading to Intensive Care Unit admission among reverse transcriptase- polymerase chain reaction (RT-PCR) positive patients in Sukraraj Tropical and infectious disease hospital, Nepal. A case-control study was conducted from June 2022 to July 2022 among patients admitted to Sukraraj Tropical and Infectious Disease Hospital. A hospital-based age (± 2 years) and sex-matched case-control study design were adopted in which ICU admitted (case group, n = 33) and general ward admitted (control group, n = 66) were included. Data were collected using a structured questionnaire comprising of socio-demographic, clinical, and preventive predictors. Data were analyzed using the Statistical Package for Social Science version 11.5. The Chi-square test and conditional logistic regression to determine the predictors associated with ICU admission. High blood pressure, high C-reactive protein and poor application of preventive practices were found to be the predictors of ICU admission. Conditional logistics regression analyses revealed that independent risk factors associated with ICU admission were elevated blood pressure (AOR = 2.22; 95% CI 1.05–4.71, p = 0.015) and abnormal C-Reactive Protein (AOR = 2.92; 95% CI 1.24–6.84, p = 0.012) at the time of hospital admission were more likely to get admitted to ICU. Likewise, patients with poor preventive practice (AOR = 3.34; 95% CI 1.19–9.31, p = 0.02) more likely to get admitted to ICU than patient with good preventive practices.These research findings hold potential significance for facilitating early triage and risk assessment in COVID-19 patients.新型冠状病毒肺炎（COVID-19）的临床特征差异显著，可表现为无症状感染或轻度上呼吸道感染，乃至重症肺炎。既往研究表明，20.0%的COVID-19患者需住院治疗，其中10.0%~20.0%需收入重症监护病房（Intensive Care Unit，ICU）。本研究旨在评估尼泊尔苏克拉拉杰热带与传染病医院内，逆转录聚合酶链反应（Reverse Transcriptase-Polymerase Chain Reaction，RT-PCR）检测阳性的COVID-19患者中，与病情进展至需收入ICU相关的预测因素。本研究于2022年6月至7月在苏克拉拉杰热带与传染病医院开展，为病例对照研究。本研究采用基于医院的年龄（±2岁）与性别匹配的病例对照研究设计，纳入收入ICU的患者作为病例组（n=33），以及收入普通病房的患者作为对照组（n=66）。研究人员通过结构化问卷收集数据，内容涵盖社会人口学特征、临床指标及预防措施相关预测因素。采用社会科学统计软件包11.5版（Statistical Package for the Social Sciences，SPSS 11.5）进行数据分析，通过卡方检验与条件logistic回归分析确定与ICU收治相关的预测因素。研究结果显示，高血压、高C反应蛋白水平以及预防措施依从性不佳，为患者需收入ICU的预测因素。条件logistic回归分析进一步表明，入院时合并高血压（校正优势比AOR=2.22；95%置信区间CI：1.05~4.71，p=0.015）及C反应蛋白异常（AOR=2.92；95%CI：1.24~6.84，p=0.012）的患者，收入ICU的风险显著升高。同样，预防措施依从性差的患者相较于依从性良好的患者，收入ICU的风险更高（AOR=3.34；95%CI：1.19~9.31，p=0.02）。本研究结果对于推动COVID-19患者的早期分诊与风险评估具有潜在的临床应用价值。