Survival in adult inpatients with COVID-19

We conducted a nationwide and retrospective cohort study to assess the survival experience and determining factors in adult inpatients with laboratory- confirmed COVID-19. Data from 5,393 individuals were analyzed using the Kaplan-Meier method and a multivariate Cox proportional hazard regression model was fitted. The 7-day survival was 0.822 and went to 0.482, 0.280, and 0.145 on days 15, 21, and 30 of hospital stay, respectively. In the multiple anal- ysis, factors associated with an increased risk of dying were: male gender, age, longer disease evolution before hospital entry, exposure to mechanical ventila- tor support, and personal history of chronic noncommunicable diseases (namely obesity, type-2 diabetes mellitus, and chronic kidney disease). To the best of our knowledge, this is the first study analyzing the survival probability in a large subset of Latin-American adults with COVID-19 and our results contribute to achieving a better understanding of disease evolution.

COVID-19; Inpatients; Cohort Studies; Survival; Proportional Hazards Models. ∗Corresponding author Email address: [email protected] (Carlos Hern´ andez-Su´ arez) Preprint submitted to Journal of LATEX Templates May 22, 2020 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.

Worldwide, the coronavirus disease 2019 (COVID-19) by severe acute res- piratory syndrome coronavirus 2 (SARS-COV-2) pandemic represents unprece- dented health and social crisis. The clinical spectrum of SARS-COV-2 infection is wide and includes asymptomatic contagion, mild upper and unspecific res-5 piratory tract symptoms, and severe viral pneumonia [1]. Most of COVID-19 cases have a good prognosis but a subset of patients develop a critical condition and even die [2]. On May 21, 2020, the observed COVID-19 mortality in Mexico has been high and over 6.5 thousand deaths were registered [3] and, among Latin-American10 countries, is only overcome by Brazil (nearly 18 thousand deaths) [4]. Published data regarding the clinical course of COVID-19 inpatients is scarce. The com- puted 14-day survival rate in a study that took place in the city of New York (U.S.), and where 2,773 inpatients were analyzed, was around 50% [5]. The evaluation of clinical outcomes in hospitalized patients with SARS-15 COV-2 infection may help clinicians and epidemiologists better appreciate the disease evolution, and lead to a more efficient allocation of healthcare resources [6]. This study aimed to assess the survival experience and associated fac- tors in a large cohort of hospitalized adult inpatients with laboratory-confirmed COVID-19.20

Study design We conducted a nationwide and retrospective dynamic cohort study focus- ing on the survival of hospitalized adult patients with laboratory-confirmed (reverse transcription polymerase chain reaction, qRT-PCR) COVID-19. Eligi-25 ble subjects were identified from the nominal records of a normative and web- based system for the epidemiological surveillance of viral respiratory diseases, which belongs to the Mexican Institute of Social Security ( IMSS, the Spanish acronym). 2 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint Population30 Individuals aged 18 years or above at acute illness onset and with conclusive evidence of COVID-19 by SARS-COV-2 were potentially eligible. Children and teenagers were not enrolled since current data suggest that severe illness is a rare event among them [7]. Subjects with hospital admission date later than May 5, 2020, were excluded, as well as those with missing clinical or epidemi-35 ological data of interest. A total of 341 inpatients were excluded (voluntary hospital discharge, 6.5%; aged under 18 years, 6.7%; referred to another health institution, 33.1%; missing information, 53.7%). Data collection Clinical and epidemiological data of interest were collected from the au-40 dited database and included demographic characteristics, illness severity (mild- moderate/severe) [8] at hospital admission, the personal history of chronic non- communicable diseases (no/yes; obesity, arterial hypertension, type 2 diabetes mellitus, asthma, chronic obstructive pulmonary disease, and chronic kidney disease). Dates from illness onset, hospital admission, and discharge (if appli-45 cable), as well as the exposure to invasive mechanical ventilation during stay (no/yes), were also obtained from the analyzed surveillance system. The ana- lyzed variables are summarized in Table 1. Medical files from the patients and death certificates represent the primary data source of the surveillance system which data base was employed.50 Outcome We analyzed the survival time of hospitalized COVID-19 adult patients mea- sured as the time elapsed from the date of hospital entry (starting event) to the date of in-hospital death (final event). The censored variable was defined as the patients who did not present the interest event (did not die) during the follow-55 up period and the date of hospital discharge was used to compute the time-at risk. 3 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint Laboratory methods Nasopharyngeal and deep nasal swabs were collected from all analyzed pa- tients in order to perform qRT-PCR (SuperScript™ III Platinum™ One-Step60 qRT-PCR Kits) analysis. Statistical analysis Summary statistics were computed. The Kaplan–Meier method [9] was em- ployed to estimate the probability of survival from the date of hospital entry. We fitted a Cox proportional hazard regression model to evaluate factors associated65 with the risk in-hospital death. The assumption of proportional hazard was ver- ified by using a Schoenfeld residual-based test. All analyses were performed by using the Stata software (StataCorp. 2017. Stata Statistical Software: Release 15. College Station, TX: StataCorp LLC.). Ethical considerations70 This study was approved by the Local Ethics in Health Research Committee (601) of the IMSS (R-2020-601-015).

Data from 5,393 participants (admitted to hospital in a period of 62 days from March 4, to May 5, 2020) were analyzed for a total follow-up of 48,56875 person-days. The overall COVID-19 in-hospital lethality rate ( n= 1,735) was 35.7 per 1,000 person-days. The mean hospital stay (± standard deviation) was 8.4 ± 6.4 vs. 9.3 ± 4.0 days in cases with fatal and nonfatal outcome, respectively (p< 0.001). Table 1 shows the characteristics of participants for selected variables. Most80 of them were male (63.6%) and 3 out of 4 were aged 45 years or above at hospital admission. Severe illness at entry was documented in 80.5% of participants. In general and as is also shown in Table 1, enrolled patients had a high prevalence of analyzed chronic noncommunicable illnesses. 4 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint The Kaplan-Meier survival estimators are presented in Figure 1. A total of85 153 deaths were registered within the first day of stay. The survival probabilities of COVID-19 adult inpatients at different periods (1, 3, 7, 15, 21, and 30 days) from hospital admission are summarized in Table 2. The 7-day survival rate was 0.808 (95% CI 0.791-0.824). After 2 weeks from admission, the survival was below 50% (0.482, 95% CI 0.450-0.513).90 In the multiple model (Table 3), male gender (HR= 1.26, 95% Ci 1-14- 1.40) and growing age were associated with an increased risk of in-hospital death. When compared with younger participants (18-29 years), subjects aged 45-59 and 60/above years old, had a 2-fold increase in the risk of dying (45-59 years, HR= 1.99, 95% CI 1.31-3.02; 60 years or above, HR= 2.57, 95% CI 1.69-95 3.92). Subjects with longer waiting time between symptoms onset and hospital admission also had a lower survival probability ([reference: ¡1 day] 1-3 days, HR= 1.59, 95% CI 1.38-1.82; ≥ 4, HR= 1.68, 95% CI 1.51-1.87), as wells as those with severe manifestations at entry (HR= 1.32, 95% 1.15-1.52). COVID-19 inpatients requiring ventilatory mechanical support during the100 stay was also associated with the risk of dying (HR= 1.91, 95%, CI 1.70-2.15). High-risk comorbidities included obesity, type-2 diabetes mellitus, and chronic kidney disease (Table 3).

The results of this study describe the survival experience of hospitalized105 adults with COVID-19 and several factors associated with disease outcomes were evaluated. To the best of our knowledge, this is the first study evaluating illness outcomes in a large subset of Latin-American COVID-19 inpatients. The related burden of SARS-COV-2 in Mexico has been high and obesity and chronic noncommunicable diseases (mainly type-2 diabetes mellitus), both110 of them showing epidemic characteristics in Mexican adults, may play a role in the observed scenario. Public policy focusing on the prevention of these illnesses has failed and growing trends have been documented [10, 11]. 5 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint The prevalence of type-2 diabetes mellitus and arterial hypertension in our study sample was significantly higher than national means (diabetes, 31.1%115 vs. 10.3%, p< 0.001; hypertension, 36.6% vs. 18.4%, p< 0.001) [12]. These findings were secondary to the inclusion of cases requiring hospitalization; per- sonal history of chronic illness has been associated with a greater risk of severe COVID-19 manifestations and of hospital entry [13]. Gender-related differences have been documented in the severity of SARS-120 COV-2 symptomatic infection and diseases outcomes. In our study, a shorter survival was observed in males (log-rank test, p< 0.001) and, for example, the Kaplan-Meier estimator after one week of hospitalization was 0.840 (95% 0.823-0.856) and 0.810 (95% 0.797-0.824) in women and men, respectively. A protective role of estrogen signaling seems to be involved [14].125 Elderly has been consistently associated with death risk among COVID-19 patients and this association is independent from gender and other diseases which frequency also increases with age. In our study, the adjusted HR per additional year of age was 1.019 (95% CI 1.015-1.022). Factors determining the age-related risk have not been elucidates but recently published data suggest a130 role of angiotensin-converting enzyme 2 overexpression together with antibody- dependent enhancement [15]. In our study, longer waiting time between symptoms onset and admission was also associated with survival; participants with longer delay (≥4 days), and when compared with those with recent symptoms (<1 day from disease onset to135 admission), had a 70% increase in the risk of dying (HR= 1.68, 95% 1.51-1.87). Similar findings were described in Hubei, China [16], however the mean elapsed days in our study sample was lower (3.1 vs. 5.7). Patients requiring mechanical ventilator support during stay had a nearly 2-fold (HR= 1.96, 95%, CI 1.75-2.21) in death risk. This seems to be an effect140 of the illness severity rather than a cause, since ventilator support was needed in 10.4% vs.4.5% (p < 0.001) of severe and mild-moderate cases, respectively. However, and despite the use of these mechanical devices, COVID-19 patients commonly complicate with organ failure or shock [17]. In addition, bacterial co- 6 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint infections related to invasive therapeutic procedures may play an undetermined145 role in disease outcomes [18]. The inclusion of only laboratory-positive cases, together with the large sam- ple size and national representativeness, are major strengths of this study. How- ever, potential limitations must be cited. First, we were unable to assess a gra- dient between body mass and survival functions, since anthropometric registers150 are not collected by the audited epidemiological surveillance system. Instead, obesity data is collected as a dichotomous variable. And second, no biomarkers data were available and which may have improved the accuracy of built models. Among others, a prognostic value of B-type natriuretic peptide and creatine kinase-MB has been documented recently [19].155

The COVID-19 pandemic-related mortality in Mexico has been high. The survival experience of hospitalized adults was documented in this nation-wide study and factors determining the illness outcome were assessed. Since obese and type 2 diabetes mellitus patients had a poor prognosis, our results highlight160 the major relevance of public health policies and interventions focusing on their prevention in the analyzed population. Conflict of interest None to declare. Data Availability Statement165 The data that support the findings of this study are available on request from the corresponding author. 7 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint

[1] F. Zhou, T. Yu, R. Du, G. Fan, Y. Liu, Z. Liu, J. Xiang, Y. Wang, B. Song, X. Gu, L. Guan, Y. Wei, H. Li, X. Wu, J. Xu, S. Tu, Y. Zhang,170 H. Chen, B. Cao, Clinical course and risk factors for mortality of adult inpa- tients with covid-19 in wuhan, china: a retrospective cohort study., Lancet (London, England) 395 (2020) 1054–1062. doi:10.1016/S0140-6736(20) 30566-3. [2] C.-C. Lai, T.-P. Shih, W.-C. Ko, H.-J. Tang, P.-R. Hsueh, Severe acute175 respiratory syndrome coronavirus 2 (sars-cov-2) and coronavirus disease- 2019 (covid-19): The epidemic and the challenges., International journal of antimicrobial agents 55 (2020) 105924. doi:10.1016/j.ijantimicag. 2020.105924. [3] Direcci´ on General de Epidemiolog´ ıa, Covid-19 en M´ exico: Informaci´ on gen-180 eral, accessed on May 21, 2020 (2020). URL https://coronavirus.gob.mx/datos [4] World Health Organization, Coronavirus disease (COVID-19) dashboard, accessed on May 21, 2020 (2020). URL https://covid19.who.int/185 [5] I. Paranjpe, V. Fuster, A. Lala, A. Russak, B. S. Glicksberg, M. A. Levin, A. W. Charney, J. Narula, Z. A. Fayad, E. Bagiella, et al., Association of treatment dose anticoagulation with in-hospital survival among hospital- ized patients with covid-19, Journal of the American College of Cardiology (2020). doi:10.1016/j.jacc.2020.05.001.190 [6] W.-H. Liang, W.-J. Guan, C.-C. Li, Y.-M. Li, H.-R. Liang, Y. Zhao, X.-Q. Liu, L. Sang, R.-C. Chen, C.-L. Tang, T. Wang, W. Wang, Q.-H. He, Z.-S. Chen, S.-S. Wong, M. Zanin, J. Liu, X. Xu, J. Huang, J.-F. Li, L.-M. Ou, B. Cheng, S. Xiong, Z.-H. Xie, Z.-Y. Ni, Y. Hu, L. Liu, H. Shan, C.-L. Lei, Y.-X. Peng, L. Wei, Y. Liu, Y.-H. Hu, P. Peng, J.-M. Wang, J.-Y.195 8 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint Liu, Z. Chen, G. Li, Z.-J. Zheng, S.-Q. Qiu, J. Luo, C.-J. Ye, S.-Y. Zhu, L.-L. Cheng, F. Ye, S.-Y. Li, J.-P. Zheng, N.-F. Zhang, N.-S. Zhong, J.- X. He, Clinical characteristics and outcomes of hospitalised patients with covid-19 treated in hubei (epicenter) and outside hubei (non-epicenter): A nationwide analysis of china., The European respiratory journal (Apr.200 2020). doi:10.1183/13993003.00562-2020. [7] J. F. Ludvigsson, Systematic review of covid-19 in children shows milder cases and a better prognosis than adults, Acta Paediatrica 109 (6) (2020) 1088–1095. doi:10.1111/apa.15270. [8] M. Cascella, M. Rajnik, A. Cuomo, S. C. Dulebohn, R. Di Napoli, Features,205 evaluation and treatment coronavirus (covid-19), in: Statpearls [internet], StatPearls Publishing, 2020. [9] M. K. Goel, P. Khanna, J. Kishore, Understanding survival analysis: Kaplan-meier estimate, International journal of Ayurveda research 1 (4) (2010) 274. doi:10.4103/0974-7788.76794.210 [10] T. Shamah-Levy, M. Romero-Mart´ ınez, L. Cuevas-Nasu, I. M´ endez G´ omez- Humaran, M. Antonio Avila-Arcos, J. A. Rivera-Dommarco, The mex- ican national health and nutrition survey as a basis for public policy planning: Overweight and obesity, Nutrients 11 (8) (2019) 1727. doi: 10.3390/nu11081727.215 [11] C. A. D´ avila-Cervantes, M. Agudelo-Botero, Sex disparities in the epidemic of type 2 diabetes in mexico: national and state level results based on the global burden of disease study, 1990–2017, Diabetes, metabolic syndrome and obesity: targets and therapy 12 (2019) 1023. doi:10.2147/DMSO. S205198.220 [12] Secretar´ ıa de Salud, Instituto Nacional de Salud P´ ublica, Instituto Na- cional de Estad´ ıstica y Geograf´ ıa, Encuesta Nacional de Salud y Nutrici´ on (ENSANUT) 2018: Presentaci´ on de resultados, accessed on May 21, 2020 9 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint (2019). URL https://ensanut.insp.mx/encuestas/ensanut2018/doctos/225 informes/ensanut_2018_presentacion_resultados.pdf [13] R.-H. Du, L.-R. Liang, C.-Q. Yang, W. Wang, T.-Z. Cao, M. Li, G.- Y. Guo, J. Du, C.-L. Zheng, Q. Zhu, et al., Predictors of mortality for patients with covid-19 pneumonia caused by sars-cov-2: a prospec- tive cohort study, European Respiratory Journal 55 (5) (2020). doi:230 10.1183/13993003.00524-2020. [14] R. Channappanavar, C. Fett, M. Mack, P. P. Ten Eyck, D. K. Meyerholz, S. Perlman, Sex-based differences in susceptibility to severe acute respira- tory syndrome coronavirus infection, The Journal of Immunology 198 (10) (2017) 4046–4053. doi:10.4049/jimmunol.1601896.235 [15] J. Schatzmann Peron, H. Nakaya, Susceptibility of the elderly to sars-cov-2 infection: Ace-2 overexpression, Shedding and Antibody-dependent En- hancement (ADE) (2020). doi:10.6061/clinics/2020/e1912. [16] W.-h. Liang, W.-j. Guan, C.-c. Li, Y.-m. Li, H.-r. Liang, Y. Zhao, X.- q. Liu, L. Sang, R.-c. Chen, C.-l. Tang, et al., Clinical characteristics and240 outcomes of hospitalised patients with covid-19 treated in hubei (epicenter) and outside hubei (non-epicenter): A nationwide analysis of china, Euro- pean Respiratory Journal (2020). doi:10.1183/13993003.00562-2020. [17] N. Chen, M. Zhou, X. Dong, J. Qu, F. Gong, Y. Han, Y. Qiu, J. Wang, Y. Liu, Y. Wei, et al., Epidemiological and clinical characteristics of 99 cases245 of 2019 novel coronavirus pneumonia in wuhan, china: a descriptive study, The Lancet 395 (10223) (2020) 507–513. doi:10.1016/S0140-6736(20) 30211-7. [18] M. J. Cox, N. Loman, D. Bogaert, J. O’grady, Co-infections: potentially lethal and unexplored in covid-19, The Lancet Microbe (2020). doi:10.250 1016/S2666-5247(20)30009-4. 10 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint [19] M. Aboughdir, T. Kirwin, A. Abdul Khader, B. Wang, Prognostic value of cardiovascular biomarkers in covid-19: A review, Viruses 12 (5) (2020) 527. doi:10.3390/v12050527. 11 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint Tables and Figures255 Figure 1: Survival estimators and 95% confidence intervals (CI) in 5,393 adult inpatients with laboratory-confirmed COVID-19, Mexico 2020 12 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint Table 1. Characteristics of study sample, Mexico 2020 Died Total Follow-up n = 1,735 n = 5,393 (person-days) Gender Female 577 (33.3) 1,961 (36.4) 17,678 Male 1,158 (66.7) 3,432 (63.6) 30,890 Age group (years) 18-29 23 (1.3) 231 (4.3) 1,744 30-44 212 (12.2) 1,113 (20.6) 9,397 45-59 651 (37.6) 2,082 (38.6) 18,999 60 or more 849 (48.9) 1,967 (36.5) 18,428 Days from symptoms onset to hospitalization <1 681 (39.3) 2,277 (45.2) 23,421 1 to 3 299 (17.2) 909 (16.9) 7,373 ≥ 4 755 (43.5) 2,207 (40.9) 17,774 Disease severity Mild-moderate 234 (13.5) 1,052 (19.5) 9,451 Severe 1,501 (86.5) 4,341 (80.5) 39,117 Invasive mechanical ventilation No 1,371 (79.0) 4,895 (90.8) 43,773 Yes 364 (21.0) 498 (9.2) 4,795 Hospital stay (days) 3 or less 427 (24.6) 639 (11.9) 1,283 4-6 372 (21.4) 641 (11.9) 3,215 7-15 720 (41.5) 3,691 (68.4) 35,147 16-30 201 (11.6) 394 (7.3) 7,917 31 or more 15 (0.9) 28 (0.5) 1,006 Personal history of: 13 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint Table 1 continued from previous page Table 1.Characteristics of study sample, Mexico 2020 Died Total Follow-up n = 1,735 n = 5,393 (person-days) Obesity (BMI 30 or higher) No 1,277 (73.6) 4,196 (77.8) 37,757 Yes 458 (26.4) 1,197 (22.2) 10,811 Arterial hypertension No 927 (53.4) 3,420 (63.4) 31,099 Yes 808 (45.6) 1,973 (36.6) 17,469 Type-2 diabetes mellitus No 1,033 (59.5) 3,716 (68.9) 34,056 Yes 702 (40.5) 1,677 (31.1) 14,512 Asthma No 1,690 (97.4) 5,247 (97.3) 47,263 Yes 45 (2.6) 146 (2.7) 1,305 COPD No 1,612 (92.9) 5,120 (94.9) 46,072 Yes 123 (7.1) 273 (5.1) 2,496 Chronic kidney disease No 1,562 (90.0) 5,094 (94.5) 46,135 Yes 173 (10.0) 299 (5.5) 2,433 Abbreviations: BMI, body mass index; COPD, Chronic obstructive pulmonary disease Note: The absolute and relative (%) frequencies are presented 14 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint Table 2. Kaplan Meier survival estimates in adult inpatients with COVID-19, Mexico 2020 Day Begin Deaths Survival 95% CI 1 5,393 153 0.972 (0.967-0.976) 3 4,982 140 0.920 (0.912-0.927) 7 4,113 122 0.822 (0.811-0.832) 15 510 45 0.482 (0.456-0.507) 21 183 17 0.280 (0.250-0.309) 30 33 2 0.145 (0.114-0.180) Abbreviations: CI, Confidence interval. 15 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint T able 3. Hazard ratio of dying in COVID-19 adult inpatients, Mexico 2020 HR (95% CI), p Unadjusted Adjusted Male gender 1.14 (1.03-1.26) 0.011 1.26 (1.14-1.39) < 0.001 Age group, years (Ref. 18-29) 30-44 1.74 (1.13-2.67) 0.012 1.47 (0.96-2.27) 0.079 45-59 2.59 (1.71-3.93) < 0.001 1.99 (1.31-3.02) 0.001 60 + 3.48 (2.30-5.27) < 0.001 2.57 (1.69-3.92) < 0.001 Days from symptoms onset to hospitalization (Ref. < 1) 1 to 3 1.59 (1.39-1.83) < 0.001 1.59 (1.38-1.82) < 0.001 ≥ 4 1.63 (1.47-1.81) < 0.001 1.68 (1.51-1.87) < 0.001 Illness severity at admission (Ref. Mild-moderate) Severe 1.51 (1.31-1.73) < 0.001 1.32 (1.15-1.52) < 0.001 Invasive mechanical ventilation (yes) 2.20 (1.95-2.47) < 0.001 1.91 (1.70-2.15) < 0.001 Personal history of: Obesity (BMI 30 or higher), yes 1.21 (1.09-1.35) < 0.001 1.28 (1.15-1.43) < 0.001 Arterial hypertension, yes 1.56 (1.42-1.72) < 0.001 1.10 (0.99-1.22) 0.086 16 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint T able 2 continued from previous page Table 3. Hazard ratio of dying in COVID-19 adult inpatients, Mexico 2020 HR (95% CI), p Unadjusted Adjusted T ype-2 diabetes mellitus, yes 1.67 (1.52-1.84) < 0.001 1.41 (1.27-1.56) < 0.001 Asthma, yes 0.95 (0.71-1.28) 0.741 0.92 (0.68-1.25) 0.601 COPD, yes 1.42 (1.18-1.70) < 0.001 1.11 (0.92-1.34) 0.276 Chronic kidney disease, yes 2.16 (1.84-2.52) < 0.001 1.78 (1.51-2.09) < 0.001 Abbreviations: CO VID-19, Coronavirus disease 2019; HR, Hazard ratio; CI, Confidence interval; Ref., Reference; BMI, Body mass index; COPD; Chronic pulmonary obstructive disease Notes: 1) Cox proportional hazards regression models were used to compute HR and 95% CI; 2) Variables listed in the table were used to compute adjusted HR. 17 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 26, 2020. ; https://doi.org/10.1101/2020.05.25.20110684doi: medRxiv preprint