Coronavirus infection (COVID-19), caused by the SARS-CoV-2 virus, declared a pandemic by the World Health Organization (WHO), continues to be a major health concern. Around 676 million cases and 6.8 million deaths have been reported as of August 2023. ¹ The population deprived of liberty (PDL) is classified as vulnerable to COVID-19 infection, as they live in environments susceptible to the spread of infection. ^2,3

Preventing COVID-19 in prison units (PUs) is a challenge due to population density, difficulties in accessing facilities and hygiene supplies as well as limited space for isolation and quarantine to ensure the PDL safety. ^4,5 In an attempt to control the spread of infection in the PDL, Brazil adopted measures recommended by the WHO. ⁶

Due to the complex situation of COVID-19, PDL presented a high burden of the disease. ⁷ In a systematic review carried out in 2022, a prevalence of 24% was reported in PDL. ⁸ In a prevalence study carried out in the state of Espírito Santo, 31.6% of COVID-19 cases were reported in PDL. ⁹ In this context, prison health and security teams had to reorganize work processes to deal with the disease as well as carry out actions in conjunction with the health surveillance service to minimize the impact of COVID-19 transmission among the PDL.

Intersectoral collaboration between prison teams and health surveillance are essential for early detection and control of disease transmission in prisons. Understanding the aspects of COVID-19 in prisons and its real impact on healthcare services’ routine and prison security is essential to predict improvements in the disease prevention process and propose appropriate actions. Therefore, this study aimed to describe the COVID-19 outbreak in prisons and analyze health surveillance actions for its control.

This is a mixed study of a quantitative and qualitative nature to address the aspects of the COVID-19 outbreak that occurred between July and September 2020 in a Remand Center (RC) located in the municipality of São Mateus, in the north of the state of Espírito Santo.

The study was divided into two stages. In the first stage, a descriptive study of the COVID-19 outbreak in the RC was carried out using secondary data from the compulsory notification information system for diseases in the municipality of São Mateus. All notification forms that were made between July and September 2020 and had the RC of São Mateus as the notifying health facility were included.

To describe the outbreak, we used the variables from the COVID-19 suspected case notification form, which were: age; race/skin color (black, yellow, white, indigenous, and unknown); clinical signs and symptoms (fever, difficulty breathing, cough, nasal congestion, runny nose, sore throat, diarrhea, headache, weakness, loss of smell, loss of taste); comorbidities and risk factors (chronic lung disease, chronic cardiovascular disease, diabetes mellitus, and HIV infection); results of specific tests for COVID-19 (positive and negative for rapid serological immunochromatographic test (RT) and reverse transcription-polymerase chain reaction (RT-PCRT); and outcome of suspected case and evolution (cure, death, discarded).

For the second stage, a qualitative study was carried out with professionals who worked in health surveillance at the Superintendence of the Northern Region of Health and professionals from epidemiological surveillance in the municipality of São Mateus. To participate in the study, workers had to have worked on actions during the COVID-19 outbreak at RC. The interviews were audio-recorded with a voice recorder and deleted after data transcription, which were stored in a Microsoft Word^® “.doc” file.

A semi-structured interview script was used, consisting of two parts: the first contained sociodemographic information, and the second was represented by guiding questions organized into chunks, namely: understanding of the process of requesting to act within the RC; understanding of the process of acting within the RC; coping, control and prevention actions; intervening factors for acting within the RC.

The interviews were conducted individually in a private room and lasted a maximum of 45 minutes. No criteria were established for closing the interviews. Participant anonymity and confidentiality were preserved by coding statements, using the letters “HP” for healthcare professionals, followed by a sequential Arabic number for each interview. All health workers interviewed signed the Informed Consent Form.

Quantitative data analyses were performed using the software Epi Info™ statistical version 7.4.2. Relative and absolute frequencies were calculated for qualitative variables, and mean and standard deviation were calculated for quantitative variables. A histogram graph was created based on the date of onset of symptoms or the date of test collection for asymptomatic cases. The graph was created using Microsoft Office Excel^® 2019.

For qualitative data analysis, the software Interface de R pour les Analyses Multidimensionnelles de Textes et de Questionnaires (IRAMUTEQ) version 0.7 alpha 2.3.3.1 was used. This software performs analysis on text corpora through the Descending Hierarchical Classification (DHC) method, which grouped the segments of workers’ speeches into classes by similar excerpts. ¹⁰ Based on the DHC, thematic content analysis was performed, including material exploration, which aims to transform the raw data to reach the core of understanding of the text, and treatment of results, in which the author makes inferences and interpretations.

The study was approved by the ethics committee of Centro Universitário Norte do Espírito Santo, Universidade Federal do Espírito Santo, under Opinion 5.738.429/2022.

We found a high rate of SARS-CoV-2 infection among PDL with rapid transmission of the virus within the prison system. In Brazil, a cross-sectional study conducted in four prisons found a seroprevalence for SARS-CoV-2 of 81.1%. ¹¹ In the state of Espírito Santo, the prevalence among PDL was 31.64%, with emphasis on the North region, which presented 43.7% prevalence. ⁹ The differences found in the studies may be related to the structure of the PU healthcare services in the different regions, which may affect the early detection of symptoms and delay in diagnosis, affecting the spread of the disease.

Monitoring people with respiratory symptoms is essential in the health surveillance process at the PU. Early identification of symptomatic people triggers preventive actions, such as the use of masks and isolation in specific cells. It was from monitoring notifications of suspected cases of COVID-19 that the epidemiological surveillance service raised the hypothesis of a possible outbreak among the PDL, triggering actions to control the disease. The PDL presented several symptoms, the most prevalent being fever, headache, runny nose, cough and sore throat. The symptoms are inherent to the fact that the respiratory system is the most commonly affected by COVID-19. ^12-15 However, a quantity of PDL greater than the PU’s capacity may provide incipient monitoring, creating a favorable scenario for the spread of COVID-19.

A mass testing strategy used in 16 PUs in the United States reported a prevalence of 86.8% of SARS-CoV-2 among the PDL, with a reduction in the emergence of new cases in the PU. ⁴ We observed that the strategy of testing all PDL patients rapidly increased the positivity rate, but then presented a decrease in positive cases, remaining at zero, suggesting a good strategy for controlling the disease. Considering that one of the main objectives of health surveillance is disease control actions, this action was effective in identifying new cases and acting to contain the spread of COVID-19 in the PU.

Not only testing everyone, but also testing the PDL made it possible to reduce cases and control the outbreak in the PU. Measures such as screening for respiratory symptoms, dividing the PDL into symptomatic or asymptomatic blocks, isolation cells, suspension of visits, among others, were essential for controlling COVID-19. The pandemic required changes to PU’s routine. ^16,17 The pandemic scenario was challenging for PU professionals, given the need to deal with something new, with the creation of new protocols without compromising the PDL safety and health,¹⁶ with interprofessional interaction between safety professionals and health surveillance professionals being the key point for controlling COVID-19 in the PU. ¹⁸

Prison conditions are conducive to the spread of SARS-CoV-2. ¹⁹ Although these conditions are favorable for the introduction and spread of respiratory diseases, there are few reports on the transmission dynamics and impact of COVID-19 in prisons. ^20-21 In an investigation of a COVID-19 outbreak in a prison in the Brazilian capital, no statistical association was found between overcrowding or sleeping location and the presence of SARS-CoV-2 antibodies. ¹¹

There are several possibilities for SARS-CoV-2 to enter the PU. ^22,23 One of the ways of entering is through the external environment, through visitors and professionals who work inside the PU.^23-24 With the introduction of SARS-CoV-2 inside the PU, its transmission is extremely fast, which responds to a peak in cases. ²⁵ A systematic approach study assessed the risk of SARS-CoV-2 transmission in different environments, describing that a detainee infected with SARS-CoV-2 has a 60% chance of infecting another within the same cell, which may justify the speed of COVID-19 transmission in the PU. ²⁶

Professionals were concerned about the worsening of PDL with a diagnosis of COVID-19, which may require hospitalization and risk of death. At the beginning of the COVID-19 pandemic, little was known about clinical management and factors associated with death and hospitalization, so there was greater concern for the most vulnerable populations. Studies have already described a higher risk of hospitalization and death from COVID-19 in PDL when compared to the general population. ^27-29 However, in our study, no people were found in the PDL who required hospitalization and had a fatal outcome, which may be a result of early management and monitoring, and interprofessional partnership with health surveillance.

This study highlights the interprofessional actions listed to contain a COVID-19 outbreak in a public health unit that contribute to the process of preventing the disease and proposing appropriate actions for possible new outbreaks and pandemics. However, it is necessary to highlight its limitations. Since this is a secondary data analysis, the lack of complete information in the notification forms or incorrect filling of information in the information system may have influenced the results. However, since this was a major COVID-19 outbreak, the data were revisited by the health surveillance team, validating the information.

In conclusion, the data led to the understanding that the PDL is vulnerable to rapid transmission of respiratory diseases due to the context in which these people live. Health surveillance strategies for the rapid identification of introduction of pathogens into the PU and blocking transmission are essential factors to prevent the massive spread of the communicable disease. Structuring health surveillance services and qualifying their organizational process is essential for the prevention and control of future outbreaks.