key: cord-1055714-e3t1f0rt
authors: Khalili, Malahat; Karamouzian, Mohammad; Nasiri, Naser; Javadi, Sara; Mirzazadeh, Ali; Sharifi, Hamid
title: Epidemiological Characteristics of COVID-19: A Systemic Review and Meta-Analysis
date: 2020-04-06
journal: nan
DOI: 10.1101/2020.04.01.20050138
sha: 4a2cf4eeb236cbaa2885c9940554afc57cb4c9d5
doc_id: 1055714
cord_uid: e3t1f0rt

Background: Our understanding of the corona virus disease 2019 (COVID-19) continues to evolve. However, there are many unknowns about its epidemiology. Purpose: To synthesize the number of deaths from confirmed COVID-19 cases, incubation period, as well as time from onset of COVID-19 symptoms to first medical visit, ICU admission, recovery and death of COVID-19. Data Sources: MEDLINE, Embase, and Google Scholar from December 01, 2019 through to March 11, 2020 without language restrictions as well as bibliographies of relevant articles. Study Selection: Quantitative studies that recruited people living with or died due to COVID-19. Data Extraction: Two independent reviewers extracted the data. Conflicts were resolved through discussion with a senior author. Data Synthesis: Out of 1675 non-duplicate studies identified, 57 were included. Pooled mean incubation period was 5.84 (99% CI: 4.83, 6.85) days. Pooled mean number of days from the onset of COVID-19 symptoms to first clinical visit was 4.82 (95% CI: 3.48, 6.15), ICU admission was 10.48 (95% CI: 9.80, 11.16), recovery was 17.76 (95% CI: 12.64, 22.87), and until death was 15.93 (95% CI: 13.07, 18.79). Pooled probability of COVID-19-related death was 0.02 (95% CI: 0.02, 0.03). Limitations: Studies are observational and findings are mainly based on studies that recruited patient from clinics and hospitals and so may be biased toward more severe cases. Conclusion: We found that the incubation period and lag between the onset of symptoms and diagnosis of COVID-19 is longer than other respiratory viral infections including MERS and SARS; however, the current policy of 14 days of mandatory quarantine for everyone might be too conservative. Longer quarantine periods might be more justified for extreme cases.

related outcomes which are indeed of utmost importance when it comes to health system 84 preparedness (12, 13) . For example, mean number of days of incubation for COVID-19 varies 85 greatly across the existing literature ranging from 2.5 (14) to 12.1 (15) days. Our knowledge 86 of time from contracting the disease to recovery or death are even more limited. In this 87 systematic review, we tried to identify the studies that recruited patients diagnosed with 88 COVID-19 and calculate polled estimates for several epidemiological and clinical outcomes to 89 help provide an overall picture of the characteristics of COVID-19. Findings of this study could 90 help inform the ongoing public health and public policy practices across the world. Data were extracted independently by the two authors (SJ and NN), and discrepancies were 130 resolved through discussion or by arbitration with a senior co-author (AM). Data were 131 extracted on study type (e.g., descriptive, case-series, mathematical modeling), publication 132

year, location, sample size, patients' age, gender, exposure history, X-ray and computed 133 tomography (18) scan findings, symptoms, and underlying conditions if reported. We also 134 extracted data on the main outcomes of interests including the number of deaths from 135 confirmed COVID-19 cases, incubation period, as well as time from onset of COVID-19 136 symptoms to first medical visit, to ICU admission, to recovery and to death. 137 138

For studies that did not provided enough data to be included in the meta-analysis, we 140 performed a qualitative data synthesis. Case-reports with a sample size of one were also 141 removed from the meta-analysis as they did not provide any dispersion estimate. Meta-142 analysis was performed using STATA's (V.15.1) metan (for numerical variables) and metaprop 143 (for binary variables) commands. The 95% confidence intervals (CI) for binary variables were 144 computed using the exact binomial method. Heterogeneity between studies was assessed 145 using both the I 2 statistic with a cutoff of ≥ 50% and the Chi-square test with P-value <0.10 to ICU admission, recovery and death. For mean incubation period, we estimated 99% CI. We 150 also conducted a random-effects meta-regression using STATA's metareg command to 151 identify the sources of heterogeneity and explore the effect of study-level covariates where 152 data were available. A two-sided P-value <0.05 was considered as statistically significant 153 effect. 154

Participants and study characteristics 157

We found a total of 1675 non-duplicate studies, 57 of which were included in our qualitative 158 synthesis and 43 were considered for meta-analysis ( Figure 1) . A description of the main 159 characteristics of the included studies is provided in Table 1 . The 57 studies included 27 cross-160 sectional, one case-control, one retrospective cohort, and 28 case series/case report studies 161 with sample sizes of observational studies ranging greatly from one to 58182 for a study in 162 the Hubei province (20) . Inclusion criteria varied greatly across the studies but most 163 participants were hospitalized patients living or traveling from various provinces in China. 164

Median (range) age of the participants was 46.2 (17 days to 78.5 years) and about 60% were 165 men. Most studies were conducted between January and February 2020. Clinical and 166 epidemiological characteristics of the patients included in the study are presented in Table 2 . 167

Among studies that reported exposure history among their participants, most patients were 168 directly or indirectly traced back to the city Wuhan (e.g., lived in Wuhan or had recently 169 travelled to Wuhan) and the Huanan seafood market in Hubei province, China. Several cases 170 of contracting COVID-19 through close contacts with family members were also reported unilateral/bilateral pneumonia. Common symptoms reported across the studies include 174 fever, cough, shortness of breath, and fatigue/weakness. Only 15 studies reported some 175 information about the pre-existing conditions of the patients; most of whom had metabolic 176 and cardiovascular underlying conditions. 177

178

The estimated mean incubation period obtained from the included studies as well as resulting 180 pooled mean are presented in Fig. 2 . Out of the 14 studies included in the meta-analysis, 12 181

were conducted in China. The findings of meta-analysis showed that the mean incubation 182 period was 5.84 (95% CI: 4.83, 6.85) days. Heterogeneity testing (I 2 =94.7%) revealed notable 183 differences among the included studies in the meta-analysis. Multivariate meta-regression 184 results ( Table 2) showed no significant differences in incubation period time by country (China 185 vs. others, Adjusted Beta = 1.70; P-value = 0.484), age (Adjusted Beta = -1.16; P-value = 0.239) 186 or percent of male (Adjusted Beta = -13.07; P-value = 0.09) participants. 187 188

The estimated mean number of days from the onset of COVID-19 symptoms to first clinical 190 visit was 4.82 (95% CI: 3.48, 6.15). As shown in Fig. 3 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not peer-reviewed) The copyright holder for this preprint . https: //doi.org/10.1101 //doi.org/10. /2020 Mean time from onset of symptoms to ICU admission 199 The estimated mean number of days from the onset of COVID-19 symptoms to ICU admission 200 was 10.48 (95% CI: 9.80, 11.16), an estimate that was derived from one study in Singapore 201 and one in Wuhan, China (Fig. 4) . The estimated mean number of days from the onset of symptoms to death was reported in 215 three studies with a pooled mean of 15.93 (95% CI: 13.07, 18.79). All of the studies were 216 conducted in China (Fig. 6) . 217 notable differences among the included studies in the meta-analysis. Multivariate meta-222 regression results (Supplement 2) showed a significant difference in death probability by age 223 (Adjusted Beta = 0.058; P-value = 0.016). 224 225

We conducted a systematic review and meta-analysis to provide an overview of the 227 epidemiological characteristics of COVID-19. Our findings suggested that the COVID-19 has an 228 incubation time of more than five days (5.84 days) and a lag of less than five days (4.82 days) 229 from onset of symptoms to first clinical visit. On average, the symptoms lasted more than 230 twenty days (17.76 days) before recovery was achieved and the mortality risk among 231 confirmed cases was 2%, which significantly increased by age. Similar to previous studies (21), 232 fever, dry cough, shortness of breath, and fatigue were common symptoms among the 233 patients in the included studies. As expected, history of direct or indirect exposure to Wuhan 234 was frequent and the most common radiologic findings were bilateral consolidation and 235 pneumonia (22, 23). 236

We found that the average incubation period of COVID-19 infection to be less than 6 days 237 which is broadly consistent with previously reported estimates (24-27). The right tail of the 238 incubation period for COVID-19 even for 99% CI was less than seven days (6.85). This finding 239 is of particular interest as uncertainties continue to exist about the incubation period of 240 COVID-19. For example, while both WHO and United States' CDC suggest an incubation period 241 of 2-14 days, single outlier cases as long as 19 (15), 24 (28) or 27 days (29) have been reported; 242 estimates that are most likely reflecting a double exposure. Our meta-analytic findings are of 243 particular importance for quarantine-related policies and planning and suggest that the 244 current 14-day quarantine period might be rather conservative. Indeed, we found that except 245 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint . https://doi.org/10. 1101 /2020 for one small study from China in Anyang city on a cluster of six patients (Bai 2020 (15)), all 246 other studies reported incubation periods less than 11 days; therefore, a shorter period of 14 247 days would most likely suffice and it is expected that almost all infected people will show 248 symptoms within 11 days of initial exposure. Nonetheless, decisions to modify or keep the 249 existing policies need to weigh the costs of extending active quarantine against the potential 250 or costs of missing a symptomatic case. 251 COVID-19 seems to have a longer incubation period than that of other acute respiratory viral 252 infections (30) such as human coronavirus (3.2 days), influenza A (1.43-1.64 days (31)), 253 parainfluenza (2.6 days), respiratory syncytial virus (4.4 days), and rhinovirus (1.4 days). 254

Furthermore, the median incubation period for SARS has been reported estimated as 4.0 days 255 in 2009 (30), which is considerably lower than what we observed for COVID-19. The longer 256 incubation period of the COVID-19 may be one of the major factors that helps explain its rapid 257 spread in comparison previous respiratory infection viruses. Another factor that contributes 258 to spread of COVID-19 is the lag between the onset of symptoms and first clinical visit and the 259 high number of asymptomatic cases of COVID-19. Our results indicated on average it is less 260 than five days (4.82 days), and particularly increases by age (1.94 days increase per each 10 261 years increase in age). This finding suggests that MERS and SARS progress rapidly to sever 262 symptoms and respiratory failures (32) than most cases infected by . 263

In compare to MERS-COV with a mortality rate of 35.67% (34) and SARS with a mortality rate 264 of 11% (35), we found that COVID-19 has a much lower mortality rate (2%) which significantly 265 increase by age (48% increase for every 1% increase in male participants). Although this 266 estimate is comparable with previous studies (36, 37), it is important to recognize the 267 limitations of calculating mortality rates of COVID-19 while the epidemic is still growing. As 268 most cases of CVOID-19 remain asymptomatic and may recover without seeking medical care, 269 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint . https://doi.org/10. 1101 /2020 it is likely that the true rate of death among people infected with COVID-19 could be even 270 lower. On the other hand, the estimated mortality rates reported in most studies need to be 271 interpreted with caution as they are often based on the cumulative number of deaths relative 272 to the number of confirmed cases, while patients who die on a given day have been infected 273 at a much earlier date and this would bias the denominator of the mortality rate (38). 274

We acknowledge four main limitations of our systematic review. First, our findings are mainly 275 based on studies that recruited patient from clinics and hospitals and so may be biased 276 toward more severe cases. Second, many studies did not report the study outcomes by 277 subgroups such as age or gender and so we could report group-specific outcomes. Third, we 278 used the mean and the standard error of the incubation period assuming a normal 279 distribution which may have led to underestimate the right tail of the distribution. Lastly, 280

given the urgency of topic and the heterogeneity of the studies included in the review, we did 281 not conduct risk of bias and quality assessment of the studies. Inevitably, given the novelty of for healthcare systems' prevention and planning efforts. The incubation period (i.e., <11 days 291 in most studies) and the lag between the onset of symptoms and diagnosis (i.e., <5 days) are 292 longer for COVID-19 compared to other respiratory viral infections including MERS and SARS. 293 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint . https://doi.org/10. 1101 /2020 Nonetheless, current policies of 14 days of mandatory quarantine for everyone, might be too 294 conservative and longer quarantine periods might be more justified for extreme cases. As 295 effective vaccination or treatment for COVID-19 are yet to be developed, practicing the 296 fundamentals of public health and prevention efforts such as social distancing and personal 297 hygiene are critical and need to be emphasized and enforced further to reduce the 298 transmission risk of COVID-19 to vulnerable populations. 299 300 301 302 303 304 305

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint . https: //doi.org/10.1101 //doi.org/10. /2020 a Studies with a sample size less than or equal to 4 patients were labeled as case-reports (88); b A 7-year-old-boy and his parents; c Studies are in press and will be published in future issues of the respective journal 

Arrived from Wuhan

2% Cough; 3.1% Sputum