key: cord-0936064-6st0b9km
authors: Ferraz, F. R.; Venancio, I.; Lopes, L. N.; Azeredo, C.; Conde, w.
title: COVID-19 infection and hospitalization according to the burden of chronic noncommunicable diseases in Brazil
date: 2021-05-05
journal: nan
DOI: 10.1101/2021.05.03.21256532
sha: 6907caf13bc8e177e6489640d20eb95ba57e450a
doc_id: 936064
cord_uid: 6st0b9km

Chronic diseases, worse socioeconomic conditions and old age can increase infection and hospitalization rate due to Coronavirus disease (COVID-19). We assessed the association between the burden of NCDs and the occurrence of infections and hospitalizations of COVID-19 in Brazil in a large COVID-19 national survey data. We analyzed only data collected between July and November 2020 (n = 1,071,782). The frequencies of positive COVID-19 diagnosis and NCD burden were estimated according to age, sex, socioeconomic strata and skin color categories. We estimated hazard ratios and 95% confidence intervals using Cox regression models. There is a non-linear dose-response inverse association between per capita income and the rates of infection and hospitalization due to COVID-19. The presence of NCDs was associated with a higher incidence of COVID-19 infection (HR1NCD = 1.34; 95% CI: 1.26; 1.43; HR2 or more NCD= 1.54 95% CI: 1.39; 1.71) and incidence of hospitalization (HR1NCD = 3.08 95% CI: 2.26; 4.19; HR 2or more NCD= 6.81 95% CI: 4.88; 9.49).The difference between the risks of infection or hospitalization of COVID-19 attributable to the burden of NCDs is non-linearly associated with the income.

hospitalization due to COVID-19. The presence of NCDs was associated with a higher incidence 22 of COVID-19 infection (HR1NCD = 1.34; 95% CI: 1.26; 1.43; HR2 or more NCD= 1.54 95% CI: 1.39; 1.71) 23 and incidence of hospitalization (HR1NCD = 3.08 95% CI: 2.26; 4.19; HR 2or more NCD= 6.81 95% CI: 4.88; 24 9.49).The difference between the risks of infection or hospitalization of COVID-19 attributable to 25 the burden of NCDs is non-linearly associated with the income. hypertension are the comorbidities leading towards the worst prognosis for COVID-19 (4,5). 37 In Brazil, considering the prevalence of NCDs, nationwide estimates for adults at risk of 38 severe COVID-19 range from 34.0% (53 million) to 54.5% (86 million) (6). To date we have no 39 evidence that the presence of comorbidities increases the risk of infection or if there is a gradient 40 between the number of comorbidities and the risk of hospitalizations due to COVID-19 (7). 41 Therefore, the aim of this study was to analyze the association between the burden of NCDs and 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 this version posted May 5, 2021. between 20 and 59 years of age, collected between July and November 2020 (n=2.650.459).

PNAD COVID-19 collected information about exams, results of COVID-19, income and 57 sociodemographic characteristics of residents. It was also asked about previous diagnosis of 58 diabetes mellitus, high blood pressure, chronic respiratory diseases (asthma, bronchitis, 59 emphysema and others), heart diseases (infarction, angina, heart failure, arrhythmia), depression 60 and cancer. Three questions related to the tested specimen were also used to diagnose COVID-19 61 infection, without discriminating the test performed (molecular biology, antigen or antibody).

The oronasal swab test uses the RT-PCR technique to identify viral RNA, usually up to the 8th 63 day of infection (9,10). In the venous blood test it is possible to detect antigen (current infection) 64 or antibody (current or past infection), and it is usually applied from the 8th day after the 65 beginning of the symptoms . The capillary blood test, also available for diagnosis, allows 66 verification of results after 15 minutes of the exam (11). We used three diagnostic outcomes:

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 certified by peer review)

The copyright holder for this preprint this version posted May 5, 2021. ; https://doi.org/10.1101/2021.05.03.21256532 doi: medRxiv preprint positive diagnosis in oronasal swab, positive diagnosis in venous blood, and positive diagnosis in 68 capillary blood.

Sociodemographic information was also analyzed, such as sex (male and female), age (in 70 years), skin color ("white"/"yellow" and "black"/"brown"/"indigenous"), information on per The frequencies of positive COVID-19 diagnosis and NCD burden were estimated according to 85 age, geographic and socioeconomic strata and skin color categories. We calculated hazard ratios 86 (HR) and 95% confidence intervals (CI) for the association between NCD and the risk of 87 infection and hospitalization for COVID-19 using Cox regression models via maximum 88 likelihood to account for repeated measures, as each person could contribute with one or more 89 interviews to the study. Findings at p < 0.05 were considered statistically significant. Initially, 90 . 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 this version posted May 5, 2021. ; https://doi.org/10.1101/2021.05.03.21256532 doi: medRxiv preprint we assessed the number of chronic noncommunicable diseases in separate models without 91 including any covariates.. Secondly, we included a priori selected demographic variables to 92 assess potential confounding: sex, age, race/skin color, education, household income and 93 rural/urban residency. The adjusted prevalence of COVID-19 was predicted according to NCD 

The prevalence of COVID-19 was higher among individuals aged 30 to 50 years (58%), 101 while the prevalence of hospitalizations was higher among individuals aged 40 or over more 102 (59%). Around 60% of the people in the age group of 40 years or over had NCDs. The 103 prevalence of 2 or more NCDs among the richest was 1.3 times the prevalence among the 104 poorest, however, the prevalence of hospitalizations was 1.2 times more frequent in the poorest 105 group.

The COVID-19 infection rate as a function of the NCD burden was not statistically 107 associated with sex. Table 2 shows the association between NCD burden and outcomes of 108 COVID-19. The presence of NCDs was associated with a higher incidence of COVID-19 109 infection for 1 NCD (HR = 1.34; 95% CI: 1.26; 1.43) and for 2 or more (HR = 1.54 95% CI: 110 1.39; 1.71). Also, the presence of NCDs was associated with higher incidence of hospitalization 111 for 1 NCD (HR = 3.08 95% CI: 2.26; 4.19) and for two or more NCDs (HR = 6.81 95% CI: 4.88; 112 9.49).

. 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. In Figure 1 , we observed that the risk of hospitalization among patients without NCD did 120 not vary according to income strata, even after adjusting for sex, age, education, income and skin 121 color. Among patients with NCDs infected with COVID-19, the chance of hospitalization was 122 higher among the richest than the poorest. Worldwide, approximately one in three adults suffers from multiple chronic diseases 131 (12). In Brazil, approximately 52% of the adult population has at least one chronic disease, 132 which is responsible for 72% of the causes of death in the country (6,13).

The COVID-19 pandemic highlights the role of NCDs as risk factors for infection, 134 hospitalization and death. Global data show that 1.7 billion people, 22% of the world population, 135 have at least one underlying condition that puts them at increased risk for COVID-19 infection 136 . 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 this version posted May 5, 2021. ; https://doi.org/10.1101/2021.05.03.21256532 doi: medRxiv preprint (14). In Brazil, from one third (53 million) to half (86 million) of adults have at least one risk 137 factor for COVID-19 (7).

The association between NCDs, diagnosis and hospitalization for COVID-19, observed in 139 our study, is in line with results from other countries (15-19) . A meta-analysis carried out with 140 Chinese data showed that the most prevalent comorbidities among patients with COVID-19 were 141 hypertension (21%), diabetes (10%), cardiovascular disease (8%) and respiratory system disease 142 (2%). The combined OR for presenting a severe case among those with hypertension, Our findings revealed that among NCDs patients infected with COVID-19, the chance of 171 hospitalization was higher for the richest relative to the poorest. Income determinants seem to be 172 associated with less access to health and intensive treatment of the disease. In our interpretation, 173 poorer individuals, even though they are more vulnerable to the contagion of the disease, are less 174 likely to obtain access to treatment and hospitalization for COVID-19. 175 As a limitation of the study, the classic limitations of observational studies can be cited, The burden of NCDs increases the risk of infection and hospitalization by This is troublesome in a context of increasing social and health inequalities as inBrazil. 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 this version posted May 5, 2021. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2021. 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 certified by peer review)

The copyright holder for this preprint this version posted May 5, 2021. . 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 certified by peer review)

The copyright holder for this preprint this version posted May 5, 2021. 

Oxygen sources and distribution for COVID-19 treatment centres

Pandemia por COVID-19 no Brasil: análise da demanda e da oferta de leitos hospitalares

e equipamentos de ventilação assistida segundo diferentes cenários. Cad Saúde Pública

Characteristics of and Important Lessons From the Coronavirus 210 Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases 211 From the Chinese Center for Disease Control and Prevention

Probabilidade de morte prematura por doenças crônicas não 214 transmissíveis, Brasil e regiões, projeções para 2025

Pesquisa Nacional de Saúde 2019: 216 Percepção do estado de saúde, estilos de vida, doenças crônicas e saúde bucal

RJ -Brasil

Adults 219 at high-risk of severe coronavirus disease-2019 (Covid-19) in Brazil

Rio de Janeiro: IBGE

. 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. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2021. ; https://doi.org/10.1101/2021.05.03.21256532 doi: medRxiv 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 this version posted May 5, 2021. ; https://doi.org/10.1101/2021.05.03.21256532 doi: medRxiv preprint