Trends in COVID-19 case-fatality rates in Brazilian public hospitals: A longitudinal cohort of 398,063 hospital admissions from 1st March to 3rd October 2020.

BACKGROUND: Almost 200,000 deaths from COVID-19 were reported in Brazil in 2020. The case fatality rate of a new infectious disease can vary by different risk factors and over time. We analysed the trends and associated factors of COVID-19 case fatality rates in Brazilian public hospital admissions during the first wave of the pandemic.
METHODS: A retrospective cohort of all COVID-19-related admissions between epidemiological weeks 10-40 in the Brazilian Public Health System (SUS) was delimited from available reimbursement records. Smoothing time series and survival analyses were conducted to evaluate the trends of hospital case fatality rates (CFR) and the probability of death according to factors such as sex, age, ethnicity, comorbidities, length of stay and ICU use.
RESULTS: With 398,063 admissions and 86,452 (21.7%) deaths, the overall age-standardized hospital CFR trend decreased throughout the period, varying from 31.8% (95%CI: 31.2 to 32.5%) in week 10 to 18.2% (95%CI: 17.6 to 18.8%) in week 40. This decreasing trend was observed in all sex, age, ethnic groups, length of stay and ICU admissions. Consistently, later admission (from July to September) was an independent protective factor. Patients 80+ year old had a hazard ratio of 8.18 (95% CI: 7.51 to 8.91). Ethnicity, comorbidities, and ICU need were also associated with the death risk. Although also decreasing, the CFR was always around 40-50% in people who needed an ICU admission.
CONCLUSIONS: The overall hospital CFR of COVID-19 has decreased in Brazilian public hospitals during the first wave of the pandemic in 2020. Nevertheless, during the entire period, the CFR was still very high, suggesting the need for improving COVID-19 hospital care in Brazil.

Introduction

Novel coronavirus disease (COVID-19) is the major global public health threat today. As of 26 March 2021, there were more than 124,535,520 confirmed cases and 2,738,876 deaths reported worldwide (https://covid19.who.int). In Brazil, almost 200,000 deaths from COVID-19 have been reported only in 2020. The infection fatality rate of COVID-19 across countries was estimated to be 0.68% (0.53%–0.82%) based on a systematic review and meta-analysis of published studies until June 16, 2020 [1]. Another systematic review and meta-analysis assessed the case-fatality rate (CFR) of patients with confirmed [2] COVID-19 in intensive care units, showing a CFR of 41.6% (34.0–49.7%). The authors also suggested that the reported mortality rates declined from above 50% in March 2020 to close to 40% in May 2020. However, both meta-analyses showed considerable heterogeneity, which may mean that observed differences in results from the included studies are not comparable.

During the first wave of infections, COVID-19 CFR appeared to fall at the hospital level as the pandemic progressed [3]. In England, the hospital CFR declined from 6.0% on April 2 to 1.5% on June 15, 2020 [4]. A national cohort study in England has also indicated that this trend remained after adjustment for patient demographics and comorbidities [5]. In Germany, COVID-19 CFR have reduced across all age groups. A larger decrease was observed in the ages 60–79, with an average close to 9% in March/April falling to 2% in July/August 2020 [6]. In the USA, adjusted mortality dropped from 25.6% (23.2–28.1) in March to 7.6% (2.5–17.8) in August 2020 in New York City [7]. Similar results seem to be observed in Singapore and the Netherlands [8].

Empirical evidence on hospital CFR over time is scarce and skewed towards high-income countries. Therefore, it is critical to also gather evidence based on routinely collected health data from upper middle-income countries. In this context, Brazil provides a unique opportunity to study trends in hospital CFR over time. First, Brazil has a universal health system, in which 75% of the population (158 million Brazilians) receives health care exclusively through the public system [9]. Second, there is large regional disparity in access to healthcare services and health outcomes, which likely worsens with the austerity economic policies recently introduced [10]. Finally, there are important public datasets available covering a large sample of the affected population, such as the hospitalization authorizations (AIH) database, covering the individual reimbursement records of all hospital admissions in the public health system (http://sihd.datasus.gov.br).

Although it is not clear when the end of the first wave of COVID-19 infections in Brazil has happened, an overlap between ongoing first wave and second wave is likely to exist due to its heterogeneous geography [11]. Nevertheless, after reaching its peak, there was a sustained trend of reduction in the number of new infections between epidemiological weeks 30 and 45, which converges to call it the first wave of COVID-19 in Brazil (https://covid.saude.gov.br). Thus, we aimed to investigate the trends in COVID-19 hospital case-fatality rate (CFR) in Brazilian public hospitals and fatality risk factors, such as sex, age, ethnicity, comorbidities, and COVID-19 severity, during the first epidemic period in 2020.

Methods

Study design

This is a retrospective cohort study based on reimbursement records of hospital admissions in the Brazilian Public Health System (Sistema Único de Saúde, SUS). The present report follows the RECORD (Reporting of Studies Conducted using Observational Routinely collected Data) statement, an extension of the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines [12].

Study setting

SUS provides health care, free of charge at the point of service, to the entire Brazilian population, covering both ambulatory and hospital care. The reimbursement of hospitalizations by SUS budget is done through the hospital admission authorizations (Autorização de Internação Hospitalar, AIH), document that identify the patient and the services performed during the hospital stay. The AIH is generated at the time o of admission to public or private hospitals that provide services for SUS, and are sent monthly to the Ministry of Health to enable billing of services delivered. The AIH are grouped and managed through the Hospital Information System (Sistema de Informações Hospitalares do SUS, SIHSUS), an administrative system that supports planning, regulation and control. Besides, SIHSUS allows the assessment of hospital morbidity and mortality profile and the quality of health care offered to the population, providing elements to improve health policies.

Participants

The study population was based on available hospital admission records. For this purpose, any admission including ICD-10 (“U071", "U072", "B972" or "B342") or medical procedure codes ("0802010296", "0802010300", "0802010318" or "0303010223") related to COVID-19 (code descriptions available at http://sigtap.datasus.gov.br) in the diagnosis, causes of death and treatment fields of the reimbursement records was identified and classified as “COVID-19 related” hospital admission, becoming part of our study population. Our data cover the admissions that occurred between the 10th and 40th epidemiological weeks (from March 1 to October 3, 2020) according to the Brazilian epidemiological calendar (S1 File).

Data sources, access and cleaning methods

All analyses were based on hospitalization authorization (AIH), which is public data available at the SIHSUS repository (htftp://ftp.datasus.gov.br/dissemin/publicos/SIHSUS/200801_/Dados) until the end of January 2021. In the AIH database, each hospitalization receives a unique key called the AIH number. If necessary, duplicated AIH numbers were filtered, considering only the main hospitalization record. The available data were fully anonymized before we accessed them.

In addition to cleaning and manipulation process in R language, the data was accessed with microdatasus package [13]. The programming code, dictionary and deidentified admissions data used in this study can be found at a public repository (https://github.com/ivanzricardo/covid19_lethality).

Variables

We considered each unique hospitalization-level data on the variables corresponding to patient characteristics (sex, age, ethnicity, and comorbidities), clinical severity (length of stay, use of ICU and occurrence of death), geographical location and epidemiological week of admission (complete description available in S1 Table in S1 File). Ethnicity was based on patient self-declaration of race/color at the time of admission, which could be classified as: white, black, brown, yellow, native Brazilian or not informed. The death outcome was based on the discharge information field (“discharge due to death”) available in the hospitalization records, thus covering only the in-hospital deaths. The selected comorbidities were obesity, bacterial infection, cancer, diabetes, cardiovascular disease, kidney failure, HIV and symptoms and signs involving the circulatory and respiratory systems (ICD-10 codes R00-R09).

Statistical analysis

Hospital case-fatality rate

Considering the date of admission as a reference point, the hospital CFR was estimated for each epidemiological week based on the proportion between the number of COVID-19 related admission that evolved to death and the total number of COVID-19 admission in that week. In order to make group comparisons, the weekly hospital CRF estimates were standardized through direct method assuming the age pattern of the total number of hospital admissions in the entire period as the reference population. Then, smoothing time series plots were created to analyse the trends of the hospital case-fatality rate and its variation according to sex, age, ethnicity, length of hospital stay and ICU use. Smoothing was based on local polynomial regression fitting (loess) [14] method available in geom_smooth() R function, which fits a polynomial surface determined by one or more numerical predictors using local fitting (fit is made using points in each point neighbourhood, weighted by their distance). Finally, we also planned to run additional Spearman correlation analysis to evaluate the effect of local caseload against the hospital CFR.

Survival analyses

A survival analysis approach was used to estimate the probability of death during the study period and to assess the relative significance of its associated factors. A descriptive analysis was performed to describe the frequency distribution and mean time by selected variables. For the survival analysis, the dependent variable was time, in days, defined as the difference between the date of death during follow-up and the date of admission.

The selection of the independent sociodemographic variables was based on the information available in the literature and the findings obtained in the stratified hospital case-fatality rate analysis. Univariate and multivariate analyses with Cox regression models were performed to estimate each independent factor hazard ratio (HR) and its 95% confidence interval.

All analyses were performed with R software, version 4.0.0 (RStudio Team. RStudio: Integrated Development Environment for R. Boston: RStudio, PBC; 2020) and its interface RStudio, version 1.3.959 (RStudio Team. RStudio: Integrated Development Environment for R. Boston: RStudio, PBC; 2020) and the Stata software program, version 14 (StataCorp. 2019. Stata Statistical Software: Release 16. College Station, TX: StataCorp LLC.).

Results

Based on our selection criteria, between March 1 (start of epidemiological week 10) and October 3 (end of epidemiological week 40), the SIHSUS received 398,063 reimbursement authorizations classified as COVID-19-related hospital admissions. Of these, 86,452 (21.7%) had death as the outcome (Fig 1).

Fig 1

Participant selection flowchart from the hospital admission records, Brazil, 2020.

Hospitalizations

Among the 398,063 hospital admissions, the number of men hospitalized surpassed that of women during the entire period, corresponding to a proportion of 55.5% of hospitalizations in the entire study period (Table 1). During the entire period, 39.5% of admissions were of people between 60 and 79 years old, 31.1% of people between 40 and 59 years old, and 13.4% between people aged 80 and over. People aged between 20 and 39 years represented 12.5% of hospitalizations, and people aged 19 years or less accounted for 3.5% of hospitalizations. Black people represented 40.6% of hospitalizations in the entire period (brown 35.8%, black 4.8%), white 29.8%, Asian 4%. Native Brazilians (indigenous) represented only 0.3% of hospitalizations in the entire period, but approximately 25% of the observations did not contain information about the patient’s ethnicity. Most hospitalizations lasted less than seven days (62.53%), followed by hospitalizations lasting between seven and 14 days (23.1%) and stays longer than 14 days (14.37%). Additionally, 26.07% of hospitalizations used the ICU.

Table 1

Characterization of 398,063 COVID-19-related hospital admissions, Brazil, 1st March to 3rd October 2020.

Variable	Death outcome		Total n (%)
	Yes n (%)	No n (%)		p-value*
Sex				< 0.001
Male	50,137 (22.70%)	170,692 (77.30%)	220,836 (55.48%)
Female	36,315 (20.49%)	140,919 (79.51%)	177,239 (44.52%)
Age				< 0.001
≤ 19 years	588 (4.19%)	13,441 (95.81%)	14,029 (3.52%)
20–39 year	3,757 (7.58%)	45,826 (92.42%)	49,586 (12.46%)
40–59 years	17,710 (14.3%)	106,178 (85.70%)	123,89 (31.12%)
60–79 years	43,323 (27.58%)	113,784 (72.42%)	157,113 (39.47%)
80 + years	21,074 (39.42%)	32,382 (60.58%)	53,457 (13.43%)
Ethnicity				< 0.001
White	25,206 (21.83%)	90,276 (78.17%)	115,482 (29.01%)
Brown	30,946 (21.69%)	111,760 (78.31%)	142,706 (35.85%)
Black	4,976 (26.19%)	14,024 (73.81%)	19,000 (4.77%)
Asian descent	3,027 (18.87%)	13,014 (81.13%)	16,041 (4.03%)
Native Brazilians (indigenous)	195 (18.82%)	841 (81.18%)	1,036 (0.26%)
Not declared	22,103 (21.29%)	81,707 (78.71%)	103,810(26.08%)
Main diagnosis				< 0.001
Other disease	5,472 (17.98%)	24,967 (82.02%)	30,439 (7.65%)
COVID-19	80,981 (22.03%)	286,655 (77.97%)	367,636 (92.35%)
High complexity admission				0,22
No	86,243 (21.72%)	310,791 (78.28%)	397,034 (99.74%)
Yes	210 (20.17%)	831 (79.83%)	1,041 (0.26%)
ICU utilization				< 0.001
No	36,378 (12.36%)	257,932 (87.64%)	294,310 (73.93%)
Yes	50,075 (48.26%)	53,690 (51.74%)	103,765 (26.07%)
Time of admission				< 0.001
March	424 (27.64%)	1,110 (72.36%)	1,534 (0.39%)
April	7,547 (29.38%)	18,144 (70.62%)	25,691(6.45%)
May	17,794 (25.37%)	52,335 (74.63%)	70,129 (17.62%)
June	17,562 (21.70%)	63,376 (78.30%)	80,938 (20.33%)
July	18,804 (20.35%)	73,607 (79.65%)	92,411 (23.21%)
August	14,304 (19.40%)	59,416 (80.60%)	73,720 (18.52%)
September	9,178 (18.56%)	40,280 (81.44%)	49,458 (12.42%)
October	840 (20.03%)	3,354 (79.97%)	4,194 (1.05%)
Comorbidity				< 0.001
Not reported	70,841 (20.08%)	281,999 (79.92%)	352,840 (88.64%)
Cardiovascular disease	7,368 (33.96%)	14,326 (66.04%)	21,964 (5.45%)
ICD R000-R099 **	2,979 (27.77%)	7,750 (72.23%)	10,729 (2.70%)
Diabetes	3,299 (33.06%)	6,681 (66.94%)	9,980 (2.51%)
Bacterial infection	3,498 (65.67%)	1,829 (34.33%)	5,327 (1.34%)
Respiratory disease	958 (22.45%)	3,309 (77.55%)	4,267 (1.07%)
Kidney failure	2,680 (67.03%)	1,318 (32.97%)	3,998 (1.00%)
Obesity	851 (29.87%)	1,998 (70.13%)	2,849 (0.72%)
Cancer	636 (41.95%)	880 (58.05%)	1,516 (0.38%)
HIV	123 (26.34%)	344 (73.66%)	467 (0.12%)
Total	86,453 (21.72%)	311,622 (78.28%)	398,063 (100%)

Notes

* Pearson’s Chi-squared test

** Signs and symptoms relating to the circulatory and respiratory systems.

Between epidemiological weeks 27 and 30, there was the highest number of hospitalizations: more than 20 thousand per week, with its peak being reached at the 28th week, when 21,461 hospitalizations were registered (Fig 2). As of 7/8/2020 (3,369 hospitalizations), a decreasing trend in the number of hospital admissions was observed.

Fig 2

Timeline of 398,063 COVID-19-related hospital admissions stratified by sex, age, comorbidities, ethnicity, length of stay and ICU need during epidemiological weeks 10 to 40, Brazil, 2020.

Regarding geographic region, it was possible to observe the different times of the epidemics according to the local number of hospitalizations (S1 Fig). The Southeast region had the highest number of hospitalizations (172,084 admissions; 1.93 admissions per 1,000 inhabitants) and the highest peak, reaching 9,051 admissions during the 28th week, but the Northern region had the highest rate (40,027 admissions; 2.14 admissions per 1,000 inhabitants). The Northern region was also the first to reach its admission peak during the 21st epidemiologic week with 2,654 admissions. The Midwestern region had the lowest number of hospitalizations and peak, with a total of 32,929 admissions (2.00 admissions per 1,000 inhabitants) and not surpassing 2,500 admissions during the entire analysed period, but the South region had the lowest rate (45,269 admissions; 1.50 admissions per 1,000 inhabitants). The Northeast region had a total of 107,754 admissions (a rate of 1.88 admissions per 1,000 inhabitants)

Hospital case-fatality rate (CFR)

Globally, after an initial growth trend until the 15th epidemiological week, hospital CFR trend decreased over time, varying from 31.8% (95%CI: 31.2 to 32.5%) in the 10th week to 18.2% (95%CI: 17.6 to 18.8%) in 40th week (Fig 3). This reduction was observed in both sexes, all age and ethnic groups. Proportionally, men died more than women, and age was directly proportional to the hospital CFR during the time. Black people data showed a higher hospital CFR and took longer to confirm reductions. In the 40th week, the age-standardized hospital CFR trend estimates were 22.53% (95%CI: 20.62 to 24.45%) in black, 18.06% (95%CI: 17.06 to 19.05%) in white, 17.78% (95%CI: 16.97 to 18.58%) in brown, 13.07% (95%CI: 09.27 to 16.86%) in native Brazilians and 14.69% (95%CI: 13.12 to 16.24%) in Asian people (complete hospital CFR data available in S2 Table in S1 File). However, as shown previously in Table 1, 26.1% of the ethnicity data were not available.

Fig 3

Timeline of 398,063 COVID-19-related hospital case-fatality rates of all admissions stratified by sex, age, comorbidities, ethnicity, length of stay and ICU need during epidemiological weeks 10 to 40, Brazil, 2020.

According to different lengths of hospitalization, it was also possible to see a hospital CFR decrease, but among those with up to seven days in duration. In this group, hospital CFR stands out, with a peak observed in the 18th epidemiological week. In addition, trends in age-standardized hospital CFR were clearly higher among people who were admitted to the ICU, reaching a peak in the 20th week of 41.03%. In the 40th epidemiological week, age-standardized hospital CFR trends were 38.40% (95%CI: 37.43 to 40.36%) and 10.91% (95%CI: 10.26 to 11.56%) in people who needed and who did not need an ICU admission, respectively.

The decreasing trend in hospital CFR was present in each of the 27 Brazilian states and it was clear in the North, Northeast and Southeast regions during all periods (S1 Fig). Nevertheless, this trend was not clear in the midwestern and southern regions, where the age-standardized hospital CFR started to decrease only after the 27th and 31st epidemiologic weeks, respectively. A complete pattern of the age-standardized hospital CFR during the epidemiological weeks in all of the 27 Brazilian states can be found in S2 Fig. In addition, as presented in S3 Fig, it was observed a strong negative correlation between the number of admissions per week and the age-standardized hospital CFR in the South (Spearman coefficient: -0.78; p-value < 0.001) and Southeast regions (Spearman coefficient: -0.79; p-value < 0.001).

Survival analyses

Table 2 shows the results of the Cox regression model for fatality in Brazilian COVID-19-related hospital admissions. The adjusted model takes into account personal characteristics, as well as the region. Age was by far the most important individual hazard factor among those analysed, where those 60+ and 80+ years old presented a 4.7 and 8.1 increased likelihood of death, respectively. With a small but significant effect, women were less susceptible to die from COVID-19. The results indicate that compared to whites, any other ethnic group is more likely to die, except for individuals of Asian descent. Some comorbidities, such as obesity, diabetes and respiratory diseases, did not have a significant effect on fatality from the new coronavirus. The presence of other diseases and infections increased the probability of hospital case-fatality, with HIV showing the higher coefficient associated with death (HR: 1.36; 95% CI: 1.134–1.631).

Table 2

Hazard ratios for COVID-19 hospital mortality adjusted for exposure factors in the multivariate Cox regression, Brazil, 1st March to 3rd October 2020.

Variable/Category	Crude values		Adjusted values
	HR	95% CI	HR	95% CI
Sex
Male	1.000		1.000
Female	0.929**	0.916–0.942	0.923**	0.910–0.936
Age
≤ 19 years	1.000		1.000
20–39 year	1.887**	1.725–2.064	1.978**	1.807–2.165
40–59 years	2.889**	2.653–3.145	2.939**	2.697–3.202
60–79 years	4.765**	4.380–5.185	4.787**	4.397–5.213
80 + years	7.658**	7.034–8.338	8.178**	7.505–8.911
Ethnicity
White	1.000		1.000
Brown	1.076**	1.058–1.094	1.045**	1.023–1.067
Black	1.140**	1.106–1.176	1.093**	1.057–1.130
Asian descent	0.986	0.949–1.024	0.975	0.935–1.017
Native Brazilians (indigenous)	1.126	0.970–1.307	1.247**	1.072–1.449
Not declared	1.093**	1.073–1.113	1.107**	1.083–1.132
Main diagnosis
Other disease	1.000		1.000
COVID-19	1.190**	1.157–1.224	1.080**	1.046–1.115
High complexity admission
No	1.000		1.000
Yes	0.535**	0.468–0.612	0.557**	0.485–0.639
ICU utilization
No	1.000		1.000
Yes	1.985**	1.958–2.014	2.077**	2.046–2.109
Time of admission
March	1.000		1.000
April	1.180**	1.064–1.310	1.131*	1.013–1.263
May	1.166**	1.052–1.292	1.073	0.963–1.197
June	1.004	0.906–1.112	0.912	0.818–1.017
July	0.961	0.868–1.065	0.881*	0.791–0.983
August	0.934	0.842–1.034	0.851**	0.763–0.949
September	0.913	0.823–1.012	0.831**	0.744–0.927
October	0.970	0.859–1.095	0.884	0.779–1.004
Comorbidity
Not reported	1.000		1.000
Cardiovascular disease	1.155**	1.127–1.183	1.048**	1.017–1.081
ICD R000-R099 *	1.210**	1.167–1.255	1.178**	1.131–1.227
Diabetes	1.145**	1.106–1.187	1.040	0.998–1.084
Bacterial infection	1.646**	1.590–1.704	1.521**	1.462–1.582
Respiratory disease	0.989	0.929–1.053	0.961	0.899–1.027
Kidney failure	1.457**	1.401–1.516	1.178**	1.128–1.231
Obesity	0.929*	0.869–0.994	1.011	0.942–1.086
Cancer	1.310**	1.210–1.418	1.398**	1.278–1.529
HIV	0.757**	0.632–0.906	1.360**	1.134–1.631
Total number of observations	398,063

* p<0.05

** p<0.01

*** Signs and symptoms relating to the circulatory and respiratory systems.

Patients treated with procedures classified as high complexity were less likely to die (HR: 0.557; 95% CI: 0.485–0.639). On the other hand, patients who needed an ICU during admission had a twofold risk of a fatal outcome. Consistent with our previous timeline hospital CFR analysis, mortality also decreased from March to September (reaching a significant drop from July to September) in the survival analysis. There was a small and not significant increase in October.

Discussion

This study has shown that after an early growth trend, the overall hospital case-fatality rate (CFR) of COVID-19-related admissions in Brazilian public hospitals decreased during the first wave period. This trend was associated with several factors, including age, sex, ethnicity, need for ICU and geographic region.

The downward trend we detected is in line with what most of the literature has been indicating recently. In a cohort of over 1,600 hospitalized patients in Spain admitted between March and September 2020, mortality decreased from 11.6% to 1.4% in the last month [15]. A similar period was analysed in the United States and indicated a drop in adjusted mortality from 25.6% in March to 7.6% in August [7]. In Germany, COVID-19 fatality rates have reduced across all age groups. A larger decrease was observed in the ages 60–79, with an average close to 9% in March/April falling to 2% in July/August 2020 [6]. The same pattern has been observed in England, where the hospital fatality ratio fell from 6% in early April to 1.5% in mid-June 2020 [4]. In France, time analysis of hospital CFR also showed a decrease over time, globally and in almost all districts [3]. Although the reasons for this reduction in hospital CFR are unknown since there is no specific treatment for COVID-19, some studies have raised potential reasons, such as a higher proportion of younger patients with fewer comorbidities over time; health workers have become more skilled at the management of severe patients during the epidemic; and early use of remdesivir to patients not requiring mechanical ventilation and dexamethasone to those requiring supplemental oxygen or on mechanical ventilation [3, 8, 15, 16]. Nevertheless, as much as Brazil has shown a decreasing trend as other countries, it is concerning that despite presenting an important relative drop (approximately 33%), it remained at a high level at week 40 –an age-standardized rate around 18%. It is worth noting that case fatality rates are not directly comparable among countries due to high heterogeneity in terms of countries’ health information systems (in-hospital versus 30-day after discharge), quality of registry (mandatory versus voluntary report), disease classification (ICD-10 versus ICD-9 or other earlier versions), and completeness of the information.

In survival analysis, sociodemographic variables and some comorbidities were identified as associated with COVID-19 hospital mortality. Like what has been identified in other studies, in Brazil and other countries, males, increasing age, black and brown ethnic group, displayed higher adjusted hazard ratios of death. Blacks, browns, and native Brazilians (indigenous) people were more likely to die during hospitalization in all estimated models. Previous studies that used data from the Hospital Information System (SIM) [17] and Epidemiological Surveillance Information System (SIVEP-Gripe) [18] to study hospital mortality related to COVID-19 in Brazil in the first semester of 2020 showed that being black or brown was an important risk factor for hospital mortality. In other countries, higher in-hospital mortality for blacks in comparison to whites has also been observed [17-20]. In Brazil, such findings may be more a reflection of socioeconomic vulnerabilities to which blacks, browns and native Brazilians are exposed, which influence living conditions, lifestyles, access to healthcare, and ultimately have implications for their health, including COVID-19 outcomes [21]. In other countries, such as England and the United States, sociodemographic inequalities are also correlated with higher mortality among blacks and Hispanics when compared to whites [19, 22].

Different from other reports in the literature that looked at obesity, respiratory diseases, and diabetes [23, 24], our results were not significant in adjusted models on these comorbidities. This may be explained by the level of reporting of these variables in an administrative data source. Higher risks of mortality were observed for comorbidities such as cancer, bacterial infections, heart disease, other symptoms involving the circulatory and respiratory systems and kidney failure. A systematic review and meta-analysis conducted by Yang and colleagues (2021) [25] identified that patients with a cancer diagnosis were more susceptible to COVID-19 and were at increased odds of dying from COVID-19. Similar findings were observed in other studies, including meta-analyses for coronary heart disease and COVID-19 [26, 27]. A systematic review and meta-analysis carried out to evaluate the significance of demographics and comorbidities in COVID-19 demonstrated that metabolic diseases, comprising CVD, diabetes, hypertension and respiratory diseases (COPD and others), were the most common comorbidities associated with a severely poor prognosis and severe outcomes [28]. Between March and September 2020, a cross-sectional study carried out in 25 hospitals in the South and Southeast regions of Brazil concluded that the high risk of hospital mortality was associated with having hypertension, being male, ages over 69 years, having kidney disease and for patients who were admitted in the ICU, mortality was 47.6% [29]. For HIV, there is emerging evidence suggesting a moderately increased risk of COVID-19 mortality among people living with HIV (PLWH) [30].

Strengths and weaknesses of this study

Using a rich dataset covering 398,063 hospital admissions for COVID-19 over a 7-month period, we provided estimates of the COVID-19 hospital CFR trends by epidemiological week at public hospitals in Brazil (stratified by sex, age, and ethnicity) and risk factors related to COVID-19 mortality (controlled for sex, age, ethnicity, comorbidities, month of hospital admission, type of hospital and impatient stay). To the best of our knowledge, this is the first study to investigate trends in COVID-19 hospital CFR in an upper middle-income country in a real-world reimbursement dataset of that size.

Some limitations of our study should also be acknowledged. First, in addition to a high level of missing information on ethnicity, epidemiological information on pre-existing comorbidities, have low quality and completeness in administrative datasets since this information is not mandatory for reimbursement. Based on this, our estimates from these variables must be interpreted with caution due to potential underreporting bias. Second, approximately 25% of the Brazilian population has private health insurance, and our findings may not reflect trends and risk factors related to COVID-19 in private hospitals. Third, although the reduction in the COVID-19 hospital death rate trends over time may be explained by improvements in clinical practice, we were not able to examine this causality effect due to a lack of clinical practice records at secondary information systems. Third, the COVID-19 outbreak was asymmetrically distributed across the country and over time, particularly in large cities; thus, all global trends should be interpreted with caution. Finally, our findings reflect the hospital CFR trends and mortality risk factors related to COVID-19 admissions in the first wave of the pandemic. Finally, we highlight that our data only covers in-hospital deaths reimbursed by SUS and should not be generalized to hospitalizations in private health insurance networks, paid out-of-pocket or deaths that occurred in other settings such as home deaths.

Implications for clinical practice and health policy

Since the pandemic emerged, the main concern of health authorities worldwide has been the collapse of healthcare systems and the lack of hospital beds for patients with moderate and severe COVID-19. Our findings suggest that the response of the Brazilian public health system (SUS) to the COVID-19 pandemic from March to October 2020 was able to achieve a sustained reduction in hospital CFR over time. However, there is a long way to go in terms of achieving stability, since the crude hospital CFR for patients with COVID-19 remained high in our data (approximately 20% in October 2020). There has been a large effort to provide ICUs and respirators to public hospitals but less attention to the training of health workers to support clinical practice and the management of ventilated patients [18], which directly impacts clinical outcomes. There are also concerns about geographical access to hospital beds and ICUs across Brazilian municipalities. A study suggested that the average distance travelled by patients from 464 Brazilian municipalities (8% of the total) was more than 240 km to obtain the ICU [31].

Our results also highlighted the population groups at higher risk of death due to COVID-19 at the hospital level: elderly people; native Brazilians (indígenas), patients with comorbidities, and hospitalized in the ICU. These population groups were prioritized by the National Immunization Programme in Brazil (except for patients hospitalized in the ICU) since vaccination against COVID-19 started in mid-January 2021. Finally, since December 2020, there has been a strong resurgence of new cases and deaths in Brazil (https://covid.saude.gov.br), as well as the emergence of a new SARS-CoV-2 variant [32], and it is important to state that our results could not reflect this new reality.

Timeline of 398,063 COVID-19-related hospital admissions (left) and age-standardized hospital case-fatality rates (right) stratified by geographic region during epidemiological weeks 10 to 40, Brazil, 2020.

(TIF)

Click here for additional data file.

Hospital case-fatality rates stratified by Brazilian states during epidemiological weeks 10 to 40, Brazil, 2020.

(TIF)

Click here for additional data file.

Correlation analysis between age-standardized hospital case-fatality rates and the number of hospital admissions per week stratified by A) All regions, B) North, C) Northeast, D) South, E) Southeast and F) Midwest region during epidemiological weeks 10 to 40, Brazil, 2020.

(TIF)

Click here for additional data file.

(DOCX)

Click here for additional data file.

Transfer Alert

This paper was transferred from another journal. As a result, its full editorial history (including decision letters, peer reviews and author responses) may not be present.

14 May 2021

PONE-D-21-11421

Trends in COVID-19 case-fatality rates in Brazilian public hospitals: an analysis based on 398,063 hospital admissions records from 1st March to 3rd October 2020

PLOS ONE

Dear Dr. Zimmermann,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jun 28 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Aleksandar R. Zivkovic

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1) Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2) Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

3) In your ethics statement in the Methods section and in the online submission form, please clarify whether all data were fully anonymized before you accessed them.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer #1: This is an interesting paper, and useful as a resource to examine the changing nature of in-hospital fatality for COVID-19 in Brazil. I hope that the authors edit the paper, as it would be a shame to see it go unpublished.

There are several major issues with the publication as it currently stands:

1. The paper reports adhering to the STROBE guidelines for observational research, but does not. Some of these issues are minor, such as the title which should properly identify the study as a longitudinal cohort according to STROBE, but some are more major. In particular, the covariates used in the primary analysis are not well described and are very hard to understand. How is ethnicity defined in hospital collections in Brazil? How is this data accessed? What comorbidities were included, how were they identified (ICD codes?), does this leave room for error and which errors if so? I would suggest the authors carefully go through not just the STROBE checklist, but the accompanying papers to ensure that they are indeed meeting the guidelines for reporting.

2. There is no information on how deaths were garnered from the records, which is a major weakness of the research. Were these in-hospital deaths? Does Brazil have a linked hospital/death reporting system? If these are in-hospital deaths, reporting lags may be a large issue - if not, death reporting across regions should be considered as a potential weakness.

3. One hypothesis for the changing hospital CFR in many places is the nature of the pandemic itself. In the UK, hospital CFRs fell from the peak until the second wave, and then increased again. This is potentially due to an overwhelming effect, whereby a large number of COVID-19 cases changes the type of patient who is admitted to hospital, and thus changes the denominator for the in-hospital CFR. While it may be impossible to fully examine the impact of such changes, it is clear from the trends in CFR when stratified by age and ethnicity that there is some impact. It might be useful for this study to look at the hospital CFR for each region over time against the current caseload of that area, although this is an addition that would take some extra work.

4. There is insufficient information in the text about missing data. While routine hospital data is incredibly useful, it is also usually filled with missing fields. While the dataset appears to be impressive, there should be a detailed discussion in the manuscript of how missing data was managed.

5. The statistical analysis is currently not fully described. The method of obtaining smoothed curves should be elucidated.

6. A somewhat minor point, but to me the tables are extremely hard to read. I would suggest having more columns and fewer rows, perhaps breaking each table down by age group. Similarly, the regression outputs are hard to read, especially given that the reference categories have been excluded.

7. While the introduction and discussion are good, I would ask for more information particularly in the introduction about the Brazilian hospital system. For international readers, there is scant detail on how it works - an additional paragraph would be very helpful to understand the context.

8. Currently, the link the authors have provided to their data/code goes to github's main webpage. Probably a typo.

Reviewer #2: This article describes the evolution of hospital lethality due to COVID-19 in the public hospital care system in Brazil. The article uses patient health data collected by the Brazilian public social insurance system, which covers more than 300,000 patients, with the indication of the duration of hospitalization and the outcome of the hospitalization. This large dataset allows a precise analysis of the evolution of hospital lethality and the article is of undeniable interest.

Some general remarks:

1. Brazil is a very big country and have a very heterogeneous geography. The authors could present in more detail the differences between the different States of Brazil (instead regions).

2. It would be interesting to have indication on the duration of the hospitalization (statistical distribution, relationship with the outcome of the disease, etc.).

3. To study the dynamics and evolution of CFR and to compare rates over time, if we want to exclude known factors related to death (mainly age), we have to standardize on age between weeks: therefore, we have to take into account the evolution of the age structure of patients over time.

4. In several countries a strong correlation between morbidity and case fatality rates has been observed. It would be interesting to have this analysis also for Brazil. And in particular to do it by state, because the differences between regions in the evolution of the disease are sensitive (but as mentioned before, when comparing geographical units, it is necessary to standardize on age).

5. Finally, it should be noted that it is always difficult to compare hospital case fatality rates between countries, because even within areas with comparable health systems (e.g., the EU), these hospital case fatality rates show differences that cannot be explained solely by differences in patient management, but first by the difference between countries in definition, declaration and reporting systems for morbidity and mortality

6. The quality of illustration can be improved. Maps are welcome.

Some minor remarks:

1. Why calculate the lethality rate per week, when the calculation can be done per day and then smoothed per week?

2. Line 58, 59. Hospital (or inpatient, or intensive care unit inpatient) fatality rate (not mortality rate)

3. Line 63-71: need to adjust the terminology (death rates, hospital fatality rate, adjusted mortality…). I think that « inpatient case-fatality rate » or « hospital case-fatality rate » is appropriate.

4. Line 142: better “proportion between the number of COVID-19 related admission that evolved to death and the total number of COVID-19 admission in that week”.

5. Line 170-179: all these results by category (age, ethnicity, comorb.) in the hospitalized COVID-19 population must be compared with the proportion of the same category in the global population, and the authors must indicate if the differences are statistically significant. Table 1 must present these results.

6. Line 182-184: Table 1 is mixing different information. I think it must be splitled in various tables. For example, Comorbidity analysis with CFR differences will be very interesting.

7. Line 195-201: Must present morbidity rates and not only morbidity.

8. Line 208-212: comparison of CFR between ethnic group is interesting only if major cause of death (age) in excluded: data must be standardized on age before comparison. Also, size of groups is different: it is needed to present confidence intervals.

9. Line 223-226: Is there a relationship between morbidity and CFR, as observed in other countries? Globally? By region? It is possible that changes in CFR are directly related to morbidity, so the study should be refined by analyzing data on regions (or better on states).

10. Line 235-239: this is very strange. Perhaps the explanations given in the discussion (data quality) could be the subject of an earlier paragraph in the data and methods section.

11. Line 246: there is no indication of the source of the data on the health care system or the criteria used to characterize it ("well-equipped and staffed hospital" is not enough).

12. Line 266: reference needed.

13. Line 336: remark: the same level of CFR was observed in France in second and third wave.

Reviewer #3: This is a straightforward epidemiological study of a cohort of nationwide COVID-admissions in Brazil from March to October 2020 analyzing in-hospital mortality and its associated factors.

The study has been well planned and executed. The sheer size of the cohort gives the opportunity to examine factors influencing the CFR with a high power.

I would wish the authors to clarify only a couple of points:

The decrease in CFR is highest and most pronounced in the short term hospitalized group. This raises the question, whether the populations hospitalized in different categories (short term, non-ICU vs. ICU) changed over time. Examining these questions could be important for the interpretation of the time trends shown.

In addition, if there are changes in the populations an additional analysis stratified for variables with clear changes over time should be done (or reported if already done).

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Gideon Meyerowitz-Katz

Reviewer #2: No

Reviewer #3: Yes: Bernd Salzberger, MD

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

29 Jun 2021

Dear Editor,

We carefully considered each of the reviewers’ suggestions. As a result, it is our opinion that the reviewers made an exceptionally good contribution to the clarity and the quality of our paper.

Please find below all the reviewers’ comments with our responses and respective indications of changes in the manuscript:

No. Reviewer’s #1 Comments Authors’ Answers

-- This is an interesting paper, and useful as a resource to examine the changing nature of in-hospital fatality for COVID-19 in Brazil. I hope that the authors edit the paper, as it would be a shame to see it go unpublished ANSWER: Thank you very much. We carefully considered each of the suggestions to be able to publish the paper.

1 The paper reports adhering to the STROBE guidelines for observational research, but does not. Some of these issues are minor, such as the title which should properly identify the study as a longitudinal cohort according to STROBE, but some are more major. ANSWER: We have revised our manuscript in order to be consistent with RECORD statements checklist (http://record-statement.org), including our new title suggestion as:

Trends in COVID-19 case-fatality rates in Brazilian public hospitals: a longitudinal cohort of 398,063 hospital admissions from 1st March to 3rd October 2020

We would like to emphasize that even RECORD extension including specific statements for studies using routinely-collected health data (such as health administrative data), like statement 6.1 “The methods of study population selection (such as codes or algorithms used to identify subjects) should be listed in detail.”, some statements are not applicable to all settings, like “Explain how the study size was arrived at”, as we have collected all the available data of the considered period of time, not a sample.

Please, if any other important RECORD statement was not met, we will be glad to revise.

1.1 In particular, the covariates used in the primary analysis are not well described and are very hard to understand. ANSWER: We have added a better explanation to the new version of the manuscript providing details on the covariates included.

Please, if any details about the covariates is still not clear, we will be glad to revise.

1.2 How is ethnicity defined in hospital collections in Brazil? ANSWER: We have revised the manuscript methods, including a better explanation of how ethnicity was defined in our Study:

Ethnicity was based on patient self-declaration of race/color at the time of admission, which could be classified as: white, black, brown, yellow, native Brazilian or not informed

1.3 How is this data accessed? ANSWER: As stated in methods section, the data is available at the SIHSUS repository (ftp://ftp.datasus.gov.br/dissemin/publicos/SIHSUS/200801_/Dados). This repository was accessed with R microdatasus package. Thus, we have revised the text in order to make clear how the data was accessed:

All analyses were based on hospitalization authorization (AIH), which is public data available at the SIHSUS repository (ftp://ftp.datasus.gov.br/dissemin/publicos/SIHSUS/200801_/Dados) until the end of January 2021. In the AIH database, each hospitalization receives a unique key called the AIH number. If necessary, duplicated AIH numbers were filtered, considering only the main hospitalization record.

In addition to cleaning and manipulation process in R language, the data was accessed with microdatasus package13. The programming code, dictionary and deidentified admissions data used in this study can be found at a public repository (https://github.com).

1.4 What comorbidities were included, how were they identified (ICD codes?), does this leave room for error and which errors if so? ANSWER: Comorbidities were included based on associated ICD-10 codes, which were fulfilled by hospitals in order to require reimbursement from the Ministry of Health. As comorbidities refer to secondary-diagnosis, this information is not mandatory (just primary-diagnosis is mandatory for means of reimbursement in the public health system in Brazil). On this basis, comorbidities tend to be under reported. All ICD-10 codes are available in Table S1 in the supplementary material.

1.5 I would suggest the authors carefully go through not just the STROBE checklist, but the accompanying papers to ensure that they are indeed meeting the guidelines for reporting. ANSWER: We have revised our manuscript in order to be consistent with RECORD statements checklist (http://record-statement.org). Please, if any other important RECORD statement was not met, we will be glad to review.

2 There is no information on how deaths were garnered from the records, which is a major weakness of the research. Were these in-hospital deaths? Does Brazil have a linked hospital/death reporting system? If these are in-hospital deaths, reporting lags may be a large issue - if not, death reporting across regions should be considered as a potential weakness. ANSWER: There is no public linked data with National Death Information System (SIM), but there is a particular field in hospitalization records stating if it was a discharge due to death or not. Thus, the deaths here reported regard only to in-hospital deaths. As we are only dealing with in-hospital deaths based on discharge data from the year 2020, we believe that reporting lag is not a major issue. Nevertheless, we agree with reviewer’s points and the revised manuscript states clear how deaths were garnered from the records and also emphasize the limitation of only dealing with in-hospital deaths in discussion section.

The death outcome was based on the discharge information field (discharge due to death) available in the hospitalization records, thus covering only the in-hospital deaths.

[…]

Finally, we highlight that our data only covers in-hospital deaths reimbursed by SUS and should not be generalized to hospitalizations or deaths in other settings as home deaths.

3 One hypothesis for the changing hospital CFR in many places is the nature of the pandemic itself. In the UK, hospital CFRs fell from the peak until the second wave, and then increased again. This is potentially due to an overwhelming effect, whereby a large number of COVID-19 cases changes the type of patient who is admitted to hospital, and thus changes the denominator for the in-hospital CFR. While it may be impossible to fully examine the impact of such changes, it is clear from the trends in CFR when stratified by age and ethnicity that there is some impact. It might be useful for this study to look at the hospital CFR for each region over time against the current caseload of that area, although this is an addition that would take some extra work. ANSWER: We agree with the comment. All our trend estimates are presented now in an age-standardized fashion, which deals with potential changes in the age structure of the population affected during the wave. A brief explanation was included in methods section:

In order to make group comparisons, the weekly hospital CRF estimates were standardized through direct method assuming the age pattern of the total number of hospital admissions in the entire period as the reference population.

In addition, we have added a correlation analysis of the age-standardized CRF against the week caseload of each region. These results were included in the manuscript results in the Figure S3. A brief explanation was also included in methods section:

Finally, we also planned to run additional Spearman correlation analysis to evaluate the effect of local caseload against the hospital CFR.

Thanks to the comment, we think we have more robust estimates now.

4 There is insufficient information in the text about missing data. While routine hospital data is incredibly useful, it is also usually filled with missing fields. While the dataset appears to be impressive, there should be a detailed discussion in the manuscript of how missing data was managed. ANSWER: All the major fields included in the analyses are obligatory fields. Nevertheless, we identified potential missing information in ethnicity (which was describe in results section, where about 20% of the data was stated as “not informed”) and comorbidities field, which could suffer from underreporting.

Because of the limited amplitude of the number of variables impacted and their profile (ethnicity and comorbidities), we considered that imputation or other missing values procedures were not applicable.

Still, we have revised our discussion section and highlighted the statement about underreporting:

First, in addition to a high level of missing information on ethnicity, epidemiological information on pre-existing comorbidities, have low quality and completeness in administrative datasets since this information is not mandatory for reimbursement. Based on this, our estimates from these variables must be interpreted with caution due to potential underreporting bias.

5 The statistical analysis is currently not fully described. The method of obtaining smoothed curves should be elucidated. ANSWER: We have revised the methods section and a better description of the method of obtaining smoothed curves is now available:

Then, smoothing time series plots were created to analyse the trends of the hospital case-fatality rate and its variation according to sex, age, ethnicity, length of hospital stay and ICU use. Smoothing was based on local polynomial regression fitting (loess)14 method available in geom_smooth() R function, which fits a polynomial surface determined by one or more numerical predictors using local fitting (fit is made using points in each point neighbourhood weighted by their distance). Finally, we also planned to run additional Spearman correlation analysis to evaluate the effect of local caseload against the hospital CFR.

6 A somewhat minor point, but to me the tables are extremely hard to read. ANSWER: Thanks. We have revised all the tables to make them clearer.

6.1 I would suggest having more columns and fewer rows, perhaps breaking each table down by age group. ANSWER: Thanks. Although not adding more columns and fewer rows, we have revised all the tables to make them clearer.

6.2 Similarly, the regression outputs are hard to read, especially given that the reference categories have been excluded. ANSWER: Thanks. We have revised all the tables to make them clearer, including the reference categories.

7 While the introduction and discussion are good, I would ask for more information particularly in the introduction about the Brazilian hospital system. For international readers, there is scant detail on how it works - an additional paragraph would be very helpful to understand the context. ANSWER: We have expanded the description of the SIHSUS in the "Study Setting" section:

"SUS provides health care, free of charge at the point of service, to the entire Brazilian population, covering both ambulatory and hospital care. The reimbursement of hospitalizations by SUS budget is done through the hospital admission authorizations (Autorização de Internação Hospitalar, AIH), document that identify the patient and the services performed during the hospital stay. The AIH is generated at the time o of admission to public or private hospitals that provide services for SUS, and are sent monthly to the Ministry of Health to enable billing of services delivered. The AIH are grouped and managed through the Hospital Information System (Sistema de Informações Hospitalares do SUS, SIHSUS), an administrative system that supports planning, regulation and control. Besides, SIHSUS allows the assessment of hospital morbidity and mortality profile and the quality of health care offered to the population, providing elements to improve health policies."

8 Currently, the link the authors have provided to their data/code goes to github's main webpage. Probably a typo. ANSWER: We have not stated the specific github repository in the text for anonymity purpose during the peer review process, but we will be glad to share the complete code if wanted or we can already describe the full link if allowed by editors.

No. Reviewer’s #2 Comment Authors’ Answer

-- This article describes the evolution of hospital lethality due to COVID-19 in the public hospital care system in Brazil. The article uses patient health data collected by the Brazilian public social insurance system, which covers more than 300,000 patients, with the indication of the duration of hospitalization and the outcome of the hospitalization. This large dataset allows a precise analysis of the evolution of hospital lethality and the article is of undeniable interest ANSWER: Thank you very much. We carefully considered each of the suggestions in order to enhance the clarity and the quality of our paper.

1 Brazil is a very big country and have a very heterogeneous geography. The authors could present in more detail the differences between the different States of Brazil (instead regions) ANSWER: We agree with this comment. Although not being able in the scope of this paper to run all the analyses separately for each Brazilian state, we have now included a supplementary analysis of the age-standardized hospital CFR timeline on each of the 27 Brazilian states. This can be found in Figure S3.

2 It would be interesting to have indication on the duration of the hospitalization (statistical distribution, relationship with the outcome of the disease, etc.). ANSWER: Thanks for the comment. We have revised the entire structure of Table 1, including this information.

In addition, we believe that important relationships have been addressed in the CFR according to the length of stay and, icu need in the survival analysis model.

3 To study the dynamics and evolution of CFR and to compare rates over time, if we want to exclude known factors related to death (mainly age), we have to standardize on age between weeks: therefore, we have to take into account the evolution of the age structure of patients over time. ANSWER: Thanks for the comment. All our trend estimates are presented now in an age-standardized fashion, which deals with potential changes in the age structure of the population affected during the wave. A brief explanation was included in methods section:

In order to make group comparisons, the weekly hospital CRF estimates were standardized through direct method assuming the age pattern of the total number of hospital admissions in the entire period as the reference population.

We think we have more robust estimates now.

4 In several countries a strong correlation between morbidity and case fatality rates has been observed. It would be interesting to have this analysis also for Brazil. And in particular to do it by state, because the differences between regions in the evolution of the disease are sensitive (but as mentioned before, when comparing geographical units, it is necessary to standardize on age). ANSWER: Thanks for the comment. Following the suggestion, we sought to analyze the direct impact of morbidity on CFR through the correlation between the number of admissions per week and the age-standardized CFR for each region. So, we have added a correlation analysis of the age-standardized CRF against the week caseload of each region. A brief explanation was included in methods section:

Finally, we also planned to run additional Spearman correlation analysis to evaluate the effect of local caseload against the hospital CFR.

We identified a strong correlation in two of the 5 Brazilian regions. We have included this data in the revised manuscript and in the supplementary material Figure S3. Although not being able in the scope of this paper to run all the analyses separately for each of the 27 Brazilian states, we have also included a supplementary analysis of the age-standardized hospital CFR timeline on each state. This can be found in Figure S2

Thanks to the comment, we think we have more robust estimates now.

5 Finally, it should be noted that it is always difficult to compare hospital case fatality rates between countries, because even within areas with comparable health systems (e.g., the EU), these hospital case fatality rates show differences that cannot be explained solely by differences in patient management, but first by the difference between countries in definition, declaration and reporting systems for morbidity and mortality ANSWER: We totally agree with the reviewer’s point of view. When we brought evidence from other countries, our idea was to highlight the “trend” and not the “value of case-fatalily rates" per se. To make it clear, we insert new sentences in the discussion.

It is worth noting that case fatality rates are not directly comparable among countries due to high heterogeneity in terms of countries’ health information systems (in-hospital versus 30-day after discharge), quality of registry (mandatory versus voluntary report), disease classification (ICD-10 versus ICD-9 or other earlier versions), and completeness of the information.

We totally agree with the reviewer.

6 The quality of illustration can be improved. Maps are welcome. ANSWER: Thanks. We have revised our Figures and included more illustrations in the supplementary material

We have all the Figures in high quality vectorial files (SVG), but the PLOS One system doesn’t recognize it as a “Figure”. Thus, in addition to the .tif files, we have uploaded a compressed file with all figures in high quality vectorial format (SVG).

MR1 Why calculate the lethality rate per week, when the calculation can be done per day and then smoothed per week? ANSWER: When dealing with daily hospitalizations instead of weekly there is a high potential for outliers (days with high or low rates), mainly in the first epidemic weeks. The epidemic week is standardized method that performs better and allows the comparison of data year after year. Similar to previous studies and consistent with WHO recommendations on dissemination on epidemiological information on cases and outbreaks of diseases under the International Health Regulations, we think that the epidemic week fashion is more stable.

MR2 Line 58, 59. Hospital (or inpatient, or intensive care unit inpatient) fatality rate (not mortality rate) ANSWER: Ok, revised

MR3 Line 63-71: need to adjust the terminology (death rates, hospital fatality rate, adjusted mortality…). I think that « inpatient case-fatality rate » or « hospital case-fatality rate » is appropriate. ANSWER: Ok, the term hospital case-fatality rate was adopted

MR4 Line 142: better “proportion between the number of COVID-19 related admission that evolved to death and the total number of COVID-19 admission in that week” ANSWER: Ok, revised

MR5 Line 170-179: all these results by category (age, ethnicity, comorb.) in the hospitalized COVID-19 population must be compared with the proportion of the same category in the global population, and the authors must indicate if the differences are statistically significant. Table 1 must present these results. ANSWER: Thanks for the comment. Nevertheless, we would like to emphasize that the entire sample was hospitalized, and it is not possible to calculate the mentioned proportion. Though, we worked on a clearer version of the data description, including the proportion of hospitalizations according to each category and the occurrence or not of death in Table 1.

MR6 Line 182-184: Table 1 is mixing different information. I think it must be splitled in various tables. For example, Comorbidity analysis with CFR differences will be very interesting. ANSWER: We have worked on a clearer version of the data description in Table 1 including the proportion of hospitalizations according to each category and the occurrence or not of death.

Nevertheless, a better analysis of the CRF according to the presence of comorbidities and other factors is also presented in Table 2 with Cox regression results.

MR7 Line 195-201: Must present morbidity rates and not only morbidity. ANSWER: We agree and now we have presented morbidity rates as well in the text.

MR8 Line 208-212: comparison of CFR between ethnic group is interesting only if major cause of death (age) in excluded: data must be standardized on age before comparison. Also, size of groups is different: it is needed to present confidence intervals. ANSWER: Thanks for the comment. We have revised our trend estimates in an age-standardized fashion, which deals with potential changes in the age structure of the population affected during the wave.

We think we have more robust estimates now. In addition, confidence intervals for the cited estimates were also included in the text.

MR9 Line 223-226: Is there a relationship between morbidity and CFR, as observed in other countries? Globally? By region? It is possible that changes in CFR are directly related to morbidity, so the study should be refined by analyzing data on regions (or better on states). ANSWER: Thanks for the comment. Following the suggestion, we sought to analyze the direct impact of morbidity on CFR through the correlation between the number of admissions per week and the age-standardized CFR for each region. We identified a strong correlation in two of the 5 Brazilian regions. We have included this data in the revised manuscript and a figure about it in the supplementary material Figure S3.

Although not being able in the scope of this paper to run all the analyses separately for each Brazilian state, we have also included a supplementary analysis of the age-standardized hospital CFR timeline on each of the 27 Brazilian states. This can be found in Figure S3

MR10 Line 235-239: this is very strange. Perhaps the explanations given in the discussion (data quality) could be the subject of an earlier paragraph in the data and methods section. ANSWER: Exactly. As mentioned in the discussion section, we believe that this is a potential limitation. Our database is very valid in some points such as the presence of the discharge outcome, however, it is still deficient in other aspects such as the description of comorbidities.

MR11 Line 246: there is no indication of the source of the data on the health care system or the criteria used to characterize it (“well-equipped and staffed hospital” is not enough). ANSWER: Thanks. We have revised this definition. This field is classified by the Ministry of Health according to the set of procedures reimbursed. The high complexity is the set of procedures that, in the context of SUS, involve high technology and high cost.

We have added this definition in the TableS1 in Supplementary information.

MR12 Line 266: reference needed. ANSWER: Thanks. I’m afraid that the all the references were included at the end of the whole sentence. Please, if there is a better way of citing this statement, we will be glad to revise.

MR13 Line 336: remark: the same level of CFR was observed in France in second and third wave. ANSWER: Thanks for the comment

Notes: MR: Minor Remark

No. Reviewer’s Comments Authors’ Answers

Reviewer #3

1 This is a straightforward epidemiological study of a cohort of nationwide COVID-admissions in Brazil from March to October 2020 analyzing in-hospital mortality and its associated factors. The study has been well planned and executed. The sheer size of the cohort gives the opportunity to examine factors influencing the CFR with a high power. ANSWER: Thank you very much. We considered carefully each of the suggestions in order to enhance the clarity and the quality of our paper.

2 The decrease in CFR is highest and most pronounced in the short term hospitalized group. This raises the question, whether the populations hospitalized in different categories (short term, non-ICU vs. ICU) changed over time. Examining these questions could be important for the interpretation of the time trends shown. ANSWER: We would like to thank the reviewer for the comment. Indeed, the population structure could have changed overtime and this could be observed in crude estimates.

Thus, we have revised our trend estimates and now they are presented in an age-standardized fashion, which deals with potential changes in the age structure of the population affected during the wave. A brief explanation was included in methods section:

In order to make group comparisons, the weekly hospital CRF estimates were standardized through direct method assuming the age pattern of the total number of hospital admissions in the entire period as the reference population.

We think we have more robust estimates now.

3 In addition, if there are changes in the populations an additional analysis stratified for variables with clear changes over time should be done (or reported if already done). ANSWER: Thanks for the comment. We have revised our trend estimates in an age-standardized fashion, which deals with potential changes in the age structure of the population affected during the wave.

We think we have more robust estimates now.

Submitted filename: Response_to_reviewers.docx

Click here for additional data file.

1 Jul 2021

Trends in COVID-19 case-fatality rates in Brazilian public hospitals: a longitudinal cohort of 398,063 hospital admissions from 1st March to 3rd October 2020

PONE-D-21-11421R1

Dear Dr. Zimmermann,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Aleksandar R. Zivkovic

Academic Editor

PLOS ONE

6 Jul 2021

PONE-D-21-11421R1

Trends in COVID-19 case-fatality rates in Brazilian public hospitals: a longitudinal cohort of 398,063 hospital admissions from 1st March to 3rd October 2020

Dear Dr. Zimmermann:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Aleksandar R. Zivkovic

Academic Editor

PLOS ONE