Artigos

Conteúdo (article):

Spatiotemporal distribution of people hospitalized with cancer and COVID-19 in Rio de Janeiro: two years of pandemic

Lucian da Silva Viana1
orcid.org/0000-0002-4718-1748. E-mail: lucianviana@yahoo.com.br
Gustavo Menezes Silva Damasceno1
orcid.org/0000-0003-4712-7124. E-mail: gustavo.menezesd@gmail.com
Gina Torres Rego Monteiro1
orcid.org/0000-0002-9900-1825. E-mail: gina.monteiro@fiocruz.br
Andrea Sobral1
orcid.org/0000-0003-0552-771X. E-mail: andrea.sobral@fiocruz.br
1Programa de Pós-graduação em Saúde Pública e Meio Ambiente (PPGSPMA), da Escola Nacional de Saúde Pública Sergio Arouca (ENSP), Fundação Oswaldo Cruz (FIOCRUZ), Rio de Janeiro, RJ, Brasil.

ABSTRACT
Objective: to analyze the spatiotemporal distribution of people with cancer and COVID-19 hospitalized in the city of Rio de Janeiro - RJ. Methods: this is an ecological research, focused on the analysis of spatial distribution and exploratory analysis of time of hospitalization, Intensive Care Unit (ICU) and death. In the spatial analysis, the Moran Global and Local Index was calculated using the QGIS software (v.3.22.16) and GeoDa (v.1.22.0.4). In the time analysis, the Mann-Whitney and Kruskal-Wallis tests were used, using the SPSS Statistics® software. Results: the city of Rio de Janeiro had the highest number of cases per 100,000 inhabitants (12.56), with positive spatial autocorrelation between the number of hospitalized patients and their municipalities of origin (Global Moran Index was 0.583). The presence of comorbidities (p-value < 0.01) and the different periods of the pandemic (p-value = 0.03) had an effect on the time to hospitalization. Both the type of tumor (p-value < 0.01 and p-value = 0.02) and the different periods of the pandemic (p-value < 0.01 and p-value = 0.01) had an effect on the time of hospitalization and death, respectively. Conclusion: the need for screening and referral strategies is highlighted, as well as the relevance of personalized care protocols for cancer patients.
Keywords: Coronavirus infections, neoplasms, hospitalization.

INTRODUCTION
COVID-19, a disease caused by the infection of SARS-CoV-2, was defined in February 2020 by the World Health Organization (WHO) as a pandemic1. This was the most severe public health event of recent decades. In two years, about 470 million cases and 6 million deaths were confirmed worldwide2. Brazil stood out negatively during the pandemic due to high incidence rates (13,986.5 cases per 100 thousand inhabitants) and cumulative mortality (310.3 deaths per 100 thousand inhabitants)3.
In two years, in the state of Rio de Janeiro, Brazil, approximately 2 million cases (12,000 per 100,000 inhabitants) and about 72,000 deaths due to the disease (420 per 100,000 inhabitants) were reported. And only in the capital were reported about half of these cases and deaths, with approximately 930 thousand new cases (14 thousand per 100 thousand inhabitants) and about 36 thousand deaths (542 per 100 thousand inhabitants), these indices being considered the highest among Brazilian capitals3.
According to the panel of the State Health Department of Rio de Janeiro (SES-RJ), the occupancy rates of nursing beds and intensive care units (ICU) for COVID-19 have had great variations in recent years, reaching maximum occupancy in peak periods4. In the capital of Rio de Janeiro, there are 15 hospitals for oncological treatment, according to the Cancer Hospital Registry (RHC)5, which are extremely important for the care of patients from the city itself, as well as from the interior of the state of Rio de Janeiro. According to the Hospital Information System - SIH, cancer treatment units had 2,325 confirmed cases of COVID-19 in the two-year pandemic period, which represents 8.5% of the total number of hospitalizations for this reason in the state capital6.
The information obtained so far suggests that, since cancer is a heterogeneous group of diseases, COVID-19 infection can affect them in different ways. Conditions related to hospitalization, such as invasive interventions and nosocomial infections, may also represent an important risk factor for complications and death7. It is believed that, in this group of diseases, in addition to factors such as treatments and other conditions related to oncological disease and COVID-198-10, exposures related to sociodemographic aspects, such as distance between the place of residence and the place of care, the time for hospital admission or even the time of hospitalization in hospital and intensive care beds, may have a direct relationship with an unfavorable outcome. The national and international production that evaluates the effects of the COVID-19 pandemic in the cancer population is growing exponentially, but there is a lack of specific studies on this spatiotemporal distribution.
During the emergence of new infectious diseases, epidemiological surveillance shall use tools to identify spatial and temporal patterns of disease and detect areas that require greater attention from health decision-makers11. In this context, spatiotemporal analysis can help to identify correlations between the incidence of hospitalizations in people with cancer, helping to identify risk factors specific to certain regions. Thus, this study aims to analyze the distribution of people with cancer and COVID-19 diagnosis hospitalized in the city of Rio de Janeiro – RJ, in the first two years of the pandemic, taking into account space and time.

METHODS
This study is an ecological research, focused on the analysis of spatial distribution and exploratory analysis of time of hospitalization, Intensive Care Unit (ICU) and death in patients with cancer due to COVID-19 in Rio de Janeiro. The investigation covers the first two years of the pandemic, between March 2020 and February 2022.
The municipality of Rio de Janeiro has an area of 1,200.329 km2. In 2022, the population consisted of 6,211,223 inhabitants and the population density was 5,175 inhab/km2. It has 257 universal health services facilities, of which 15 are registered as specialized care units in oncology5,12.
Within the scope of the Hospital Information System (SIH), cancer records were identified by applying the International Statistical Classification of Diseases and Health-related Problems – 10th Review (ICD-10). This was done by using the code relevant to group C, referring to neoplasms (chapter II), in the variables associated with primary and secondary diagnoses. As for the patients affected by COVID-19 (ICD-10 B34.2), they were identified in the database of Cases of Acute Respiratory Syndromes of the Influenza Epidemiological Surveillance Information System (SRAG/Sivep-Gripe) through the variable "CLASSI_FIN", which specifies the final diagnosis of the case (code 5 - SRAG for COVID-19), and the variable "EVOLUCAO", responsible for identifying deaths related to COVID-19 (code 2).
The data collection procedure involved the delimitation of the study interval (03/01/2020 to 02/28/2022) and the exclusion of patients under 18 years of age. This decision was based on the observation that, in this age group, the occurrence of severe cases and deaths from COVID-19 was uncommon, with hospitalization rates between 2.5% and 4.1% and need for treatment in intensive care below 1%13.
In order to identify patients hospitalized during the research period who had both cancer and COVID-19 diagnoses, the data from the SIH was linked to the SRAG/Sivep-Gripe database. This process was performed using the Python programming language (version 3.10.12) and the development environment Jupyter Notebook (6.4.8), by the following link variables: Patient name, patient mother’s name and patient birth date. Thus, with the relationship between the databases, there were 2,278 hospitalizations for analysis.
It is noteworthy that, while linking these data, scripts were implemented to extract, transform and combine relevant information from both systems (SIH and Sivep-Gripe). This included the standardization of data formats, the treatment of missing values and the harmonization of unique patient identifiers to ensure consistency and integrity of the data used in the spatiotemporal analysis proposed in this study.
The independent variables chosen are related to the following clinical and epidemiological aspects: 1) the patients’ age; 2) the presence of comorbidities (chronic cardiovascular disease, chronic hematological disease, chronic liver disease, diabetes mellitus, neurological disease, asthma or other chronic pneumopathy – in addition to COVID-19 -, immunodeficiency or immunodepression, chronic kidney disease and obesity); 3) the different tumor subtypes (solid tumor or hematological neoplasm); 4) the place of hospitalization (if in specialized hospitals in oncology or other institutions), taking into account the information extracted from the RHC Integrator System5, which centralizes and consolidates the data from the institutions specialized in Oncology accredited throughout Brazil. And 5) the periods of evolution of the pandemic and/or vaccination initiatives implemented in the city of Rio de Janeiro, as documented4:
● Period 1 (March - July, 2020): beginning of the pandemic and end of the first peak of serious cases and deaths;
● Period 2 (August, 2020 - January, 2021): second peak of serious cases and deaths;
● Period 3 (February - August, 2021): beginning of widespread COVID-19 vaccination - 1st, 2nd, 3rd doses or single dose - in the elderly;
● Period 4 (September, 2021 - February, 2022): beginning of booster dose at the end of the study period.
For the spatial analysis, two data sets were developed. The first database containing the number of patients per municipality of residence was used to draw up a thematic map showing the origin of the cases. The number of cases was divided by the population of the place of residence of the case, according to the estimate for 2021, provided by the Brazilian Institute of Geography and Statistics (IBGE) (https://www.ibge.gov.br/), and multiplied by 100 thousand. During this phase, patients residing in other states (less than 1%) were excluded to ensure a more accurate graphical representation.
In order to describe the spatial correlation between the number of patients with cancer and COVID-19 and their municipality of residence, the Global and Local Moran Index was calculated14. The Global Moran Index indicates if there is spatial dependence in the data set as a whole and the Local Moran Index calculates the spatial autocorrelation for each locality, this allows the identification of areas with significantly similar (high spatial correlation) or different (low spatial autocorrelation) values, thus enabling the detection of clusters with low or high incidence, for example. To verify if the found spatial autocorrelation is significant, a pseudo-significance test was applied, performed from 999 permutations of the values of the attributes associated with the localities. The calculation of Moran Global and Local indices was performed in GeoDa software (v.1.22.0.4), using a contiguity matrix of order 1 type.
The second set of data, containing the number of hospitalizations aggregated based on the health institution where there was care, according to the National Registry of Health Establishments (CNES) available in the SIH database, was employed in the creation of quantitative thematic maps. These maps represent the number of hospitalizations of patients in the city of Rio de Janeiro over the entire period analyzed, as well as in each of the four aforementioned periods, and also by type of cancer (solid or hematological).
The city of Rio de Janeiro was subdivided into administrative regions, using the IBGE administrative divisions. We particularly highlight oncology hospitals due to their significant representativeness, accounting for more than 90% of hospitalizations. The territorial boundaries were obtained through IBGE, and the production of maps was carried out using the QGIS software (v.3.22.16).
The 2,278 aforementioned hospitalizations are from 1,336 cases of cancer patients, 40% of which entered public hospitals in Rio de Janeiro two or more times during the study period. For the time analysis, these cases were considered. Thus, a descriptive analysis was carried out taking into account the following dependent variables: Time of hospitalization, Time of ICU and Time to death.
Frequency and central trend measurements were used for this analysis. The time averages and 95% confidence interval (95%CI) were calculated, generally and for each of the aforementioned independent variables. To evaluate the impact of each category, non-parametric tests were used by Mann-Whitney and Kruskal-Wallis, and a p-value lower than 0.05 was considered statistically significant. All these analyses, together with the elaboration of boxplot-format graphs, were conducted using SPSS Statistics® software.
This study was approved by the Research Ethics Committee (CEP), the National School of Public Health Sergio Arouca, the Oswaldo Cruz Foundation (ENSP/FIOCRUZ) and the CEP of the Rio de Janeiro Municipal Health Department. The SIH and SRAG/Sivep-Gripe banks were made available by the Rio de Janeiro Municipal Health Secretariat (SMS-RJ), in January 2023.

RESULTS
Among the 1,336 cases, 99.56% came from the state of Rio de Janeiro. Patients from other states included those in the Federal District, Maranhão, Minas Gerais, São Paulo, Tocantins and Roraima. Among the 92 municipalities of Rio de Janeiro, 47 had cases of hospitalizations in the capital, as illustrated in Figure 1a. As expected, the city of Rio de Janeiro presented the highest number of cases per 100,000 inhabitants (12.56), followed by neighboring cities of São João de Meriti (11.83) and Nilópolis (11.66). Other municipalities near the capital also contributed to the number of cases per 100,000 inhabitants, including Itaguaí (9.50), Duque de Caxias (9.47), Queimados (9.20), Nova Iguaçu (8.96), Mesquita (8.47) and Japeri (8.46).
The result of the Moran Global Index was 0.583, indicating positive spatial autocorrelation between the number of patients hospitalized with cancer and COVID-19 and their municipalities of origin. In the Moran Local Index, a cluster of nine municipalities that presented high values of hospitalized patients with cancer and COVID-19 per 100,000 inhabitants and were located neighbors to other municipalities also with high values (p-value ≤ 0.01), according to Figure 1b. The observed cluster includes the state capital, Rio de Janeiro, and much of the Baixada Fluminense, including the municipalities of Nilópolis, São João de Meriti, Mesquita, Queimados, Nova Iguaçu, Duque de Caxias, Belford Roxo and Seropédica (p-value ≤ 0.01).
Figure 1: Incidence of COVID-19 cases in cancer patients hospitalized in Rio de Janeiro, according to place of origin (a) and Moran Local Index (b).
Regarding the location of hospitals analyzed in this study, most are located in the central and southern regions of the city of Rio de Janeiro. During the period analyzed, 93.24% of hospitalizations were recorded in 12 hospitals specialized in oncology (oncology hospitals), while the remaining 6.76% occurred in 17 hospitals without specialization in this area (non-oncology hospitals), as you can see in Figure 2.
Figure 2: Quantitative map of hospitalizations in the city of Rio de Janeiro, with emphasis on oncology hospitals.
Oncology hospitals: 1 – Federal Hospital of Andaraí; 2 – General Hospital of Bonsucesso; 3 –Federal Hospital Cardoso Fontes; 4 – University Hospital Clementino Fraga Filho; 5 – University Hospital Gaffree and Guinle; 6 – Cancer Hospital I (INCA); 7 – Cancer Hospital II (INCA); 8 – Cancer Hospital III (INCA); 9 - Federal Hospital of Ipanema; 10 - Federal Hospital of Lagoa; 11 – Hospital Mario Kroeff; 12 - Hospital of the State Workers.
Other hospitals: 13 – Institute Fernandes Figueira (Fiocruz); 14 – National Institute of Infectious Diseases Evandro Chagas (Fiocruz); 15 – National Institute of Traumatology and Orthopedics Jamil Haddad; 16 – Regional Emergency Coordination Leblon; 17 – Municipal Hospital Albert Schweitzer; 18 –Municipal Hospital of Piedade; 19 – Municipal Hospital Evandro Freire; 20 – Municipal Hospital Francisco da Silva Telles; 21 – Municipal Hospital Lourenço Jorge; 22 – Municipal Hospital Miguel Couto; 23 –Municipal Hospital Pedro II; 24 – Municipal Hospital Rocha Faria; 25 – Municipal Hospital Ronaldo Gazolla; 26 – Municipal Hospital Salgado Filho; 27 – Municipal Hospital Souza Aguiar; 28 – Maternity Hospital Fernando Magalhães; and 29 - University Hospital Pedro Ernesto.
The oncology hospital that registered the most hospitalizations during the entire study period was the Cancer Hospital III, of the National Cancer Institute (HC III/INCA) (500 hospitalizations), located in the administrative region (RA) of Vila Isabel; followed by HC I/INCA (442 hospitalizations), located in the RA of the Center and by the HC II/INCA (336 hospitalizations), located in the RA of the Port Area. Other hospitals that also registered a high number of hospitalizations were the Federal Hospital of Andaraí (207 hospitalizations), in the RA of Vila Isabel; the Federal Hospital of Lagoa, in the RA of Lagoa (107 hospitalizations); and Hospital of the State Workers (101 hospitalizations), located in the RA of the Port Area.
The non-oncology hospitals that presented the highest number of hospitalizations were the Pedro Ernesto University Hospital (HUPE/UERJ), with 66 cases, located in the Vila Isabel Regional Government, and the Albert Schweitzer Municipal Hospital, with 18 cases, located in the Realengo Regional Government.
The 12 hospitals with oncology specialization registered 597 hospitalizations from March to July 2020 (Period 1) (Figure 3 a1); 566 hospitalizations from August 2020 to January 2021 (Period 2) (Figure 3 a2); 605 hospitalizations from February to August 2021 (Period 3) (Figure 3 a3); and 356 hospitalizations from September 2021 to February 2022 (Period 4) (Figures 3 to 4).
Figure 3: a) Number of hospitalizations in oncology hospitals during periods 1, 2, 3 and 4. b) Number of hospitalizations in oncology hospitals by type of tumor.
1 – Federal Hospital of Andaraí; 2 – General Hospital of Bonsucesso; 3 – Federal Hospital Cardoso Fontes; 4 – University Hospital Clementino Fraga Filho; 5 – University Hospital Gaffree and Guinle; 6 – Cancer Hospital I (INCA); 7 - Cancer Hospital II (INCA); 8 - Cancer Hospital III (INCA); 9 - Federal Hospital of Ipanema; 10 - Federal Hospital of Lagoa; 11 – Hospital Mario Kroeff; 12 - Hospital of the State Workers.
Hospitals HC I, HC II and HC III presented the highest number of hospitalizations during the four periods analyzed. During periods 1 and 2, hospitals HC I and HC III recorded more than 100 hospitalizations each. In period 3, these same hospitals, along with HC II, also surpassed the mark of 100 hospitalizations each. In period 4, there was a reduction in the number of hospitalizations, but again, HC I, HC II and HC III hospitals led with 62, 62 and 96 hospitalizations, respectively.
Of the total admissions to oncology hospitals, 86% were for patients with solid tumors, while 14% were for patients with hematological tumors. The health facilities that registered the most hospitalizations of patients with solid tumors were HC III, HC II and HC I, with 495, 365 and 315 hospitalizations, respectively (Figure 3 b1). The hospitals with the highest number of hospitalizations of patients with hematological tumors were HC I (127), University Hospital Gaffrée and Guinle (41) and Hospital of the State Workers (38) (Figure 3 b2).
Most patients were aged 60 years or older (61.6%) and had some comorbidity additional to cancer (51.2%), the latter being a prominent condition associated with a higher frequency of admissions to the ICU (62.0%). The most prevalent diagnosis was solid tumors (88.4%), compared to hematological tumors (11.6%). Twenty percent (20%) had at least one admission to an Intensive Care Unit (ICU). Sixty-four percent (64%) of patients died due to COVID-19 (specific lethality for COVID-19). Period 1 (Mar/2020-Jul/2020) recorded the entry of most patients in their first hospital admission (37.0%), with a death rate reaching 42.7% in this period (Table 1).
Table 1: Number and average time (in days) of hospitalization, ICU and death.
*Mann-Whitney test.
**Kruskall-Wallis test.
1, 2, 3 and 4 refer to the four study periods.
In general, the mean time between first symptoms and hospitalization was 3.6 days (CI: 3.2-4.0) and the mean time between admission and hospital discharge (hospital discharges and deaths) was 17.2 days (CI: 15.6-18.8). For patients admitted to the ICU, the mean time between hospitalization and admission to the ICU and the average time of stay in the ICU were 1.8 days (CI: 1.2-2.4) and 15.3 days (CI: 12.7-17.9), respectively. Among patients who died from COVID-19, the time between hospital admission and death was 16.8 days (CI: 14.7-18.9). The mean, confidence interval and p-value distributed by clinical and epidemiological aspects are described in Table 1.
The interval median between first symptoms and hospitalization was two days for patients with comorbidities (1oIIQ: 0/3oIIQ: 6) and one day for those without comorbidities (1oIIQ: 0/3oIIQ: 4), with a higher median (2 days, 1oIIQ: 0/3oIIQ: 5) during the first period of the pandemic. Both the presence of comorbidities (p-value < 0.01) and the different periods of the pandemic (p-value = 0.03) had a significant effect on the time elapsed between the first symptoms and hospitalization (Figure 4).
Figure 4 - Median and percentiles 25 and 75 of time (in days) of hospitalization and death.
*Mann-Whitney test.
**Kruskall-Wallis test.
The interval median between admission and hospital discharge was higher for patients with hematological tumors (13 days, 1oIIQ: 7/3oIIQ: 27) compared to those with solid tumors (9 days, 1oIIQ: 4/3oIIQ: 20). In the third period of the pandemic (Feb/2021-Aug/2021), this interval was 12 days (1oIIQ: 5/3oIIQ: 27). In addition, the median interval between hospital admission and death from COVID-19 showed a similar distribution. Both the type of tumor (p-value < 0.01 and p-value = 0.02) and the different periods of the pandemic (p-value < 0.01 and p-value = 0.01) had a significant effect on the time between admission and hospital discharge and the time between hospital admission and death from COVID-1919 respectively.

DISCUSSION
The spatial distribution of COVID-19 in Brazil was similar to other countries, where the first cases were reported in large cities and then in medium and small urban centers. However, the search for health services was reversed due to the greater supply of high-complexity services, such as ICU beds, in municipalities majors15. Among cancer patients, this search possibly occurred due to the follow-up already carried out in reference centers, which was observed in this study, where approximately 90% of admissions occurred in hospitals specialized in oncology. This corroborates the National Cancer Prevention and Control Policy (Order N. 874, of 16 May 2013)16, which establishes comprehensive cancer care done in licensed health facilities.
There are currently 317 units and centers in Brazil that provide care for the treatment of cancer17. In the city of Rio de Janeiro, 15 hospitals perform oncological treatments. According to the Hospital Cancer Records (RHC), 57,700 visits were registered in these hospitals between 2016 and 2020, which represented about 73% of oncologic visits from all over the state5. The results show this significant distribution of admissions among hospitals specialized in oncology, located mostly in the state capital of Rio de Janeiro, with emphasis on the HC I reference centers, HC II and HC III, and consistently led in number over the four periods under study.
The spatial concentration of health services observed in Brazil is also reflected in municipal and intramunicipal ways. The municipality of Rio de Janeiro has a high health infrastructure centrality18, which attracts individuals from other municipalities, sometimes from other states, in search of greater availability of more complex health services. In this study, almost all hospitalizations were of patients resident in the state of Rio de Janeiro and the capital presented the highest number of cases per 100,000 inhabitants. Despite this, there was a demand for care from other municipalities, mainly neighboring, as the results of this study point out, which may have contributed to increase the pressure on local health services.
Analysis by type of tumor reveals a significant prevalence of solid tumors, representing 86% of the total admissions in oncology hospitals. In this context, HC I, HC II and HC III emerged as the main centers for patients with solid tumors. For patients with hematological tumors, HC I, University Hospital Gaffrée and Guinle and Hospital of the State Workers were the most relevant. This suggests that the specialized hospitals analyzed play a crucial role in the care of cancer patients. The constant leadership of INCA hospitals in the management of cancer patients reinforces the importance of this institution as a pillar in the response to the treatment of cancer patients in different phases of the pandemic. In addition, the variation in the number of hospitalizations between different hospitals and types of tumors highlights the complexity of cancer treatment and emphasizes the need for differentiated approaches to each patient profile.
In the two years after the pandemic began, the state of Rio de Janeiro experienced five waves of high transmission of COVID-19 cases. In the city of Rio de Janeiro, the occupancy rates of nursing beds and ICU due to COVID-19 reached their maximum capacity during the peaks4. In turn, vaccination began in Rio de Janeiro in February 2021, prioritizing the elderly population. By the following year, in February 2022, approximately 99% of the adult population had completed the vaccination scheme, while about 60% had already received the booster dose. In this study, the periods were defined by the emergence of the first two waves (periods 1 and 2) and by the beginning of vaccination and its reinforcement (periods 3 and 4). Period 1 recorded the highest percentage of hospitalizations (37.0%), admission to ICU (30%) and deaths (42.7%). In the subsequent periods, there was a significant reduction in these rates, especially for period 4, attributed by some authors to the effectiveness of vaccination against the disease20,21.
It was also observed that the average time elapsed between the first symptoms and hospitalization was 4 days, while the average total hospitalization period was 17 days, including 15 days of stay in the ICU. Comparatively, a study conducted in Canada22 revealed a median of hospitalization time and shorter ICU, being 11 and 8 days, respectively. In contrast, a survey conducted in the state of São Paulo - SP, Brasil23 showed an average length of hospitalization similar (18 days), but with a shorter stay in the ICU, 5 days. In the city of Belém - PA, Brazil, it was observed that 49% of patients with cancer and COVID-19 hospitalized remained for 10 days in the ICU, while 57% required mechanical ventilation for the same period24. These results provide important insights into the dynamics and outcomes of hospitalizations of cancer patients during the COVID-19 pandemic.
The analysis of time between the first symptoms and hospitalization showed a significant variation, highlighting the influence of the presence of comorbidities and different periods of the pandemic. Patients with comorbidities tended to seek hospital care a little later, possibly due to restrictive changes imposed, associated with a self-perceived risk and increased health measures in this period. In the analysis of time of hospitalization in ICU there was no significant variation (this was 2 days). However, in studies with patients with COVID-19 without cancer diagnosis, conducted in the USA25 and Brazil26, the average time from hospitalization to admission to the ICU was 7 days. However, another study8, which involved patients diagnosed with COVID-19 and cancer, was observed an average of 6 days from hospital admission to the ICU, with a lower mean (3 days) for those with hematological neoplasms. These results suggest that patients with additional cancer diagnosis may be more vulnerable, requiring immediate intensive care to prevent or minimize complications, which are more common in this group of patients.
The average time between first symptoms and death was 21 days, being higher in adults (18 to 59 years: 24 days) and lower in elderly people (60 years or more: 19 days). In a Chinese study, the time to death was similar (20 days), with shorter periods recorded in lung cancer (17 days) and hematology (19 days). These intervals were higher than the average of another study conducted in Pernambuco – PE, Brazil, involving women with cancer27, in which the time between the first symptoms and death was 12 days, with the highest averages (19 days) observed between 50 and 59 years, decreasing to 12 days among the elderly aged 60 to 79 years.
Furthermore, the variation in time between admission and hospital discharge and between hospital admission and death from COVID-19 reveals important nuances, especially with respect to the type of tumor and the periods of the pandemic in which this effect was significant. Patients with hematological tumors had longer periods of hospitalization and death, possibly related to the complexity and severity of these cases, as well as clinical choices directed by different specialties. The increase in time during the third period of the pandemic suggests a variable dynamic over time, reflecting the evolution of COVID-19 understanding and treatment, as well as possible changes in health policies.
Some limiting aspects of this study deserve to be mentioned, among them the lack of detailed clinical information about the patients\' oncological disease, such as the number and type of oncologic treatments previously performed, the hospital where the most recent treatment, progression or remission of the disease was performed, among others. As well as the presence of incomplete or missing data in relation to important dates, such as specific procedure dates (such as the beginning and end of ventilatory support use, date of last surgery, date of last chemotherapy treatment cycle, among others). The inclusion of these data could enrich the study results.
Besides, typing errors may have impacted the accuracy of the analyses performed. Obtaining these data in a more complete and accurate way could improve the quality of the analyses. The lack of a unique and common identification key, such as the Natural Persons Register (CPF), the National Health Card (CNS) and the Authorization of Hospital Admission (AIH), resulted in operational difficulties. This required the use of probabilistic techniques to identify matches between records based on strings, as highlighted by Peng and Mation28.
A strategy to overcome these difficulties would be a unified digital system that integrates such information in the different levels of care and digitally, with national databases, which is not yet a reality in the public health network of Rio de Janeiro. It is worth pointing out that this analysis of the data is based on specific contexts and may vary in different regions and times of the pandemic. The constant evolution of knowledge about COVID-19 and clinical practices can influence outcomes, reinforcing the need for continuous updates in approaches to care for cancer patients during health crises.
This study demonstrates the magnitude of hospital admissions and ICU beds predominantly specialized in oncology in the city of Rio de Janeiro. Although in Brazil, with the Law n 12.732 of 2012, the cancer patient has the legal right to undergo the first treatment in the Unified Health System (SUS) within 60 days29, during the COVID-19 pandemic, there was a need to readjust the routines and health services, the suspension of non-emergency clinical services and delays in diagnosis and treatment due to the overload of health services, conditions that may have negatively impacted the incidence, the severity and mortality of cancer30. Several studies have described delays and cancellations in screening, diagnosis and oncologyco31-32 treatment, this combined with the overload caused by COVID-19 bed occupancy may have had a multidimensional impact on the cancer patient.
The results of this study may guide future research on the interaction between cancer and COVID-19, highlighting specific areas of concern or success in public health strategies adopted; as well as fill gaps in the scientific literature, providing data and analysis specific to the field of Oncology in the context of infectious diseases such as COVID-19. This will contribute to global knowledge about the intersection between pre-existing health conditions and the pandemic.
It is also important to use specific approaches adapted to different types of tumors, highlighting the need for adequate screening and referral strategies for cancer patients. The impact of the variables analyzed, such as comorbidities, type of tumor and pandemic period, underlines the complexity in the management of these patients and highlights the relevance of personalized care protocols. These conclusions highlight the need for a multidisciplinary and individualized approach to ensure the best possible care for cancer patients hospitalized in future pandemics.

Acknowledgement
To the Vice Directorate of Research and Innovation (VDPI/ENSP/Fiocruz). (VDPI/ENSP/Fiocruz).

REFERENCES
1. World Health Organization. Naming the coronavirus disease (COVID-19) and the virus that causes it [Internet]. 2020 [citado 23 de fevereiro de 2024]. Disponível em: https://www.who.int/emergencies/diseases/novel-coronavirus-2019/technical-guidance/naming-the-coronavirus-disease-(covid-2019)-and-the-virus-that-causes-it
2. World Health Organization. COVID-19 cases | WHO COVID-19 dashboard [Internet]. datadot. [citado 23 de fevereiro de 2024]. Disponível em: https://data.who.int/dashboards/covid19/cases
3. Ministério da Saúde. Coronavírus Brasil [Internet]. [citado 23 de fevereiro de 2024]. Disponível em: https://covid.saude.gov.br/
4. Secretaria de Estado de Saude do Rio de Janeiro. Painel de monitoramento Covid-19. [Internet]. Painel Covid19. [citado 23 de fevereiro de 2024]. Disponível em: https://www.saude.rj.gov.br/
5. Instituto Nacional de Câncer. Ministério da Saúde. Integrador RHC [Internet]. [citado 23 de fevereiro de 2024]. Disponível em: https://irhc.inca.gov.br/RHCNet/
6. Ministério da Saúde. Sistema de Informações Hospitalares [Internet]. [citado 23 de fevereiro de 2024]. Disponível em: http://sihd.datasus.gov.br/principal/index.php
7. Savassi LCM, Bedetti AD, Abreu ABJ de, Costa AC, Perdigão RM da C, Ferreira TP. Ensaio acerca das curvas de sobrecarga da COVID-19 sobre a atenção primária. JMPHC J Manag Prim Health Care ISSN 2179-6750. 27 de outubro de 2020;12:1–13. Disponível em: https://doi.org/10.14295/jmphc.v12.1006
8. Dai M, Liu D, Liu M, Zhou F, Li G, Chen Z, et al. Patients with Cancer Appear More Vulnerable to SARS-CoV-2: A Multicenter Study during the COVID-19 Outbreak. Cancer Discov. 2020;10(6):783–91.
9. Yang K, Sheng Y, Huang C, Jin Y, Xiong N, Jiang K, et al. Clinical characteristics, outcomes, and risk factors for mortality in patients with cancer and COVID-19 in Hubei, China: a multicentre, retrospective, cohort study. Lancet Oncol. 1o de julho de 2020;21(7):904–13.
10. Richardson S, Hirsch JS, Narasimhan M, Crawford JM, McGinn T, Davidson KW, et al. Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the New York City Area. JAMA. 26 de maio de 2020;323(20):2052–9.
11. Damasceno GMS, Oliveira BFA de, Matos VP de, Sobral A. Persistence and changes in emerging COVID-19 space-time clusters in Brazilian municipalities in 2020-2021. Hygeia - Rev Bras Geogr Médica E Saúde. 22 de janeiro de 2024;20:e2002–e2002.
12. Instituto Brasileiro de Geografia e Estatística. Panorama do Rio de Janeiro [Internet]. [citado 23 de fevereiro de 2024]. Disponível em: https://cidades.ibge.gov.br/brasil/rj/rio-de-janeiro/panorama
13. Nehab MF. COVID-19 e saúde da criança e do adolescente [Internet]. Instituto Nacional de Saúde da Mulher, da Criança e do Adolescente Fernandes Figueira. Editora Fiocruz, Rio de Janeiro; 2020. Disponível em: http://www.iff.fiocruz.br/pdf/covid19_saude_crianca_adolescente.pdf
14. Anselin L, Syabri I, Kho Y. GeoDa: An introduction to spatial data analysis. Em: Fischer MM, Getis A, organizadores. Handbook of Applied Spatial Analysis: Software Tools, Methods and Applications. 1o ed Berlin ; New York: Springer; 2010.
15. Magalhães M de AFM, Barcellos C, Xavier DR, Saldanha R de F. Pandemia de Covid-19: o processo de interiorização e aceleração da transmissão no país. Em: Covid-19 no Brasil: cenários epidemiológicos e vigilância em saúde [Internet]. Editora Fiocruz, Rio de Janeiro: Freitas, Carlos Machado de; Barcellos, Christovam; Villela, Daniel Antunes Maciel; 2021. p. 85–98. Disponível em: https://doi.org/10.7476/9786557081211.0005
16. Ministério da Saúde. PORTARIA N° 874 DE 16 DE MAIO DE 2013. Institui a Política Nacional para a Prevenção e Controle do Câncer na Rede de Atenção à Saúde das Pessoas com Doenças Crônicas no âmbito do Sistema Único de Saúde (SUS) [Internet]. Disponível em: https://pesquisa.in.gov.br/imprensa/jsp/visualiza/index.jsp?jornal=1&pagina=129&data=17/05/2013
17. Instituto Nacional de Câncer. Ministério da Saúde. Onde tratar pelo SUS [Internet]. [citado 6 de março de 2024]. Disponível em: https://www.gov.br/inca/pt-br/assuntos/cancer/tratamento/onde-tratar-pelo-sus
18. Affonso da Silva S. Pandemia de COVID-19 no Brasil: O acesso e a qualidade dos serviços de saúde como determinante social. Revista Contexto Geográfico. 6(11):56–76.
19. Secretaria Municipal do Rio de Janeiro. Painel Rio COVID-19 [Internet]. [citado 3 de julho de 2024]. Disponível em: https://experience.arcgis.com/experience/38efc69787a346959c931568bd9e2cc4
20. Mohammed I, Nauman A, Paul P, Ganesan S, Chen KH, Jalil SMS, et al. The efficacy and effectiveness of the COVID-19 vaccines in reducing infection, severity, hospitalization, and mortality: a systematic review. Hum Vaccines Immunother. 31 de dezembro de 2022;18(1):2027160.
21. Patel MD, Rosenstrom E, Ivy JS, Mayorga ME, Keskinocak P, Boyce RM, et al. Association of Simulated COVID-19 Vaccination and Nonpharmaceutical Interventions With Infections, Hospitalizations, and Mortality. JAMA Netw Open. 1o de junho de 2021;4(6):e2110782.
22. Elkrief A, Desilets A, Papneja N, Cvetkovic L, Groleau C, Lakehal YA, et al. High mortality among hospital-acquired COVID-19 infection in patients with cancer: A multicentre observational cohort study. Eur J Cancer. 2020;181–7.
23. Barros LA do R, Magalhães Filho MAF, Alves R de B, Rebouças CV de, Rodrigues CM, Viu MM, et al. Alta mortalidade entre pacientes com câncer e infecção por COVID-19: a experiência de um centro oncológico brasileiro. Einstein São Paulo. 5 de novembro de 2021;19:eAO6254.
24. Oliveira AKB de, Araújo MS de, Alves SFL, Filho LEC de S, Silva ML da, Cunha K da C. Perfil Clínico dos Pacientes Críticos Oncológicos com Covid-19. Rev Bras Cancerol. 31 de janeiro de 2023;69(1):e-133060.
25. Gupta S, Hayek SS, Wang W, Chan L, Mathews KS, Melamed ML, et al. Factors Associated With Death in Critically Ill Patients With Coronavirus Disease 2019 in the US. JAMA Intern Med. 1o de novembro de 2020;180(11):1436–47.
26. Ranzani OT, Bastos LSL, Gelli JGM, Marchesi JF, Baião F, Hamacher S, et al. Characterisation of the first 250,000 hospital admissions for COVID-19 in Brazil: a retrospective analysis of nationwide data. Lancet Respir Med. abril de 2021;9(4):407–18.
27. Galindo RJ da SC, Andrade LB de, Sena GR, Nogueira LRM, Lima TPF de, Lima JTO, et al. Women with cancer and COVID-19: an analysis of lethality and clinical aspects in Pernambuco. Rev Bras Saúde Materno Infant. 24 de fevereiro de 2021;21:157–65.
28. Peng Y, Mation LF. O desafio do pareamento de grandes bases de dados: Mapeamento de métodos de record linkage probabilístico e diagnóstico de sua viabilidade empírica [Internet]. Texto para Discussão; 2018 [citado 1o de março de 2024]. Report No.: 2420. Disponível em: https://www.econstor.eu/handle/10419/211367
29. Congresso Nacional do Brasil. LEI No 12.732, DE 22 DE NOVEMBRO DE 2012. Dispõe sobre o primeiro tratamento de paciente com neoplasia maligna comprovada e estabelece prazo para seu início. [Internet]. [citado 6 de março de 2024]. Disponível em: https://www.planalto.gov.br/ccivil_03/_ato2011-2014/2012/lei/l12732.htm
30. Sharpless NE. COVID-19 and cancer. Science. 19 de junho de 2020;368(6497):1290.
31. Teglia F, Angelini M, Casolari G, Astolfi L, Boffetta P. Global Association of COVID-19 Pandemic Measures with Cancer Treatment: A Systematic Review and Meta-Analysis. Cancers. 8 de novembro de 2022;14(22):5490.
32. Nikolopoulos M, Maheshwari MK, Doumouchtsis SK. Impact of COVID-19 in gynaecological oncology care: a systematic rapid review. Arch Gynecol Obstet. 2022;305(3):555–65.