ABSTRACT
BACKGROUND: Meta-analysis is a statistical method to synthesize evidence from a number of independent studies, including those from clinical studies with binary outcomes. In practice, when there are zero events in one or both groups, it may cause statistical problems in the subsequent analysis. METHODS: In this paper, by considering the relative risk as the effect size, we conduct a comparative study that consists of four continuity correction methods and another state-of-the-art method without the continuity correction, namely the generalized linear mixed models (GLMMs). To further advance the literature, we also introduce a new method of the continuity correction for estimating the relative risk. RESULTS: From the simulation studies, the new method performs well in terms of mean squared error when there are few studies. In contrast, the generalized linear mixed model performs the best when the number of studies is large. In addition, by reanalyzing recent coronavirus disease 2019 (COVID-19) data, it is evident that the double-zero-event studies impact the estimate of the mean effect size. CONCLUSIONS: We recommend the new method to handle the zero-event studies when there are few studies in a meta-analysis, or instead use the GLMM when the number of studies is large. The double-zero-event studies may be informative, and so we suggest not excluding them.
Subject(s)
COVID-19 , Data Analysis , Meta-Analysis as Topic , Research Design/trends , Humans , Linear ModelsABSTRACT
OBJECTIVE: COVID-19 started spreading widely in China in January 2020. Outpatient fever clinics (FCs), instituted during the SARS epidemic in 2003, were upgraded to serve for COVID-19 screening and prevention of disease transmission in large tertiary hospitals in China. FCs were hoped to relieve some of the healthcare burden from emergency departments (EDs). We aimed to evaluate the effect of upgrading the FC system on rates of nosocomial COVID-19 infection and ED patient attendance at Peking Union Medical College Hospital (PUMCH). DESIGN: A retrospective cohort study. PARTICIPANTS: A total of 6365 patients were screened in the FC. METHODS: The FC of PUMCH was upgraded on 20 January 2020. We performed a retrospective study of patients presenting to the FC between 12 December 2019 and 29 February 2020. The date when COVID-19 was declared an outbreak in Beijing was 20 January 2020. Two groups of data were collected and subsequently compared with each other: the first group of data was collected within 40 days before 20 January 2020; the second group of data was collected within 40 days after 20 January 2020. All necessary data, including patient baseline information, diagnosis, follow-up conditions and the transfer records between the FC and ED, were collected and analysed. RESULTS: 6365 patients were screened in the FC, among whom 2912 patients were screened before 21 January 2020, while 3453 were screened afterward. Screening results showed that upper respiratory infection was the major disease associated with fever. After the outbreak of COVID-19, the number of patients who were transferred from the FC to the ED decreased significantly (39.21% vs 15.75%, p<0.001), and patients generally spent more time in the FC (55 vs 203 min, p<0.001), compared with before the outbreak. For critically ill patients waiting for their screening results, the total length of stay in the FC was 22 min before the outbreak, compared with 442 min after the outbreak (p<0.001). The number of in-hospital deaths of critically ill patients in the FC was 9 out of 29 patients before the outbreak and 21 out of 38 after the outbreak (p<0.05). Nineteen cases of COVID-19 were confirmed in the FC during the period of this study. However, no other patients nor any healthcare providers were cross-infected. CONCLUSION: The workload of the FC increased significantly after the COVID-19 outbreak. New protocols regarding the use of FC likely helped prevent the spread of COVID-19 within the hospital. The upgraded FC also reduced the burden on the ED.