If so, the final CFR associated with the MERS outbreak in South Korea during 2015 might be <22%. However, the proportion of patients who died (18%–19%) has been fairly stable since June 27 ( Figure 2), which might indicate an asymptotic approach toward the final outbreak-specified CFR ( 5 – 7). Because 19 (10%) of 186 case-patients reported remain hospitalized, the final CFR of the outbreak might be higher than our current estimate. On the basis of case-patients who had known outcomes through July 15, the ongoing MERS outbreak in South Korea had an estimated CFR (22%) that was half the CFR (44%) for all known case-patients with MERS in Saudi Arabia ( 3), but a CFR similar to that calculated for patients with only nonsporadic illness (21%) ( 4). Total cumulative cases over time were calculated by date. Cumulative proportion of case-patients with Middle East respiratory syndrome who were hospitalized, recovered, and died, South Korea, as of July 15, 2015. Furthermore, for every 1-year increase in age, odds of dying increased by 12% (odds ratio 1.12, 95% CI 1.07–1.17).įigure 2. The model estimated that odds of dying were 7 times higher for persons with concurrent health conditions than for persons without these conditions (odds ratio 7.14, 95% CI 2.27–22.41). Univariate logistic regression models for each risk factor showed that older age and having a concurrent health condition were associated with death (both p<0.001) both variables remained significant after we adjusted for all 5 variables in a multivariate logistic regression model ( Table). Median time from diagnosis to death and from diagnosis to discharge were 13 (IQR 17–25.3) and 22 (IQR 9–16.5) days, respectively.Īs of July 15, a total of 35/159 cases analyzed were considered fatal, which yielded an estimated case-fatality rate (CFR) of 22%. Time from onset to diagnosis was positively skewed: median 4 days (interquartile range 2–7 days). All deaths occurred in patients >48 years of age. Age was normally distributed (range 16–87 years, mean 55 years). Per WHO definitions, 25 (16%) had concurrent health conditions and 22 (14%) were health care workers. Of the 159 case-patients analyzed, 94 (59%) were men. We tested the Cox proportional hazards assumption by using Schoenfeld residuals and included an interaction term for predictor and follow-up time. For time-to-event analyses, patients were categorized into outbreak weeks by date of onset. Five potential covariates were analyzed: sex, age, concurrent health condition status, health care worker status, and time from onset to diagnosis. We used this subset to describe the patient population, evaluate risk factors for death by using logistic regression models, and assess predictors of time from onset to diagnosis and onset to discharge by using Cox proportional hazards models. The MOHW daily MERS reports provided real-time outcome information.Īs of July 15, outcomes and covariates were publicly available for 159 of 186 case-patients, all of whom became ill during weeks 2–7 of the outbreak. The WHO line list included additional risk factor data, which were cross-validated against meta data from the MOHW.
Matching between the 2 data sources was conducted by using age, sex, and date of reporting. Case identification numbers were matched between the June 26, 2015, World Health Organization (WHO) line list ( 2) and daily text-based MERS reports from the South Korean MOHW ( 1).
0 kommentar(er)
