key: cord-0917018-x4ilyrkr
authors: Ranzani, Otavio T.; Bozza, Fernando A.
title: ‘Vaccine effectiveness of ChAdOx1 nCoV-19 against COVID-19 in a socially vulnerable community in Rio de Janeiro, Brazil' – Author’s reply
date: 2022-04-06
journal: Clin Microbiol Infect
DOI: 10.1016/j.cmi.2022.03.037
sha: 40fa28d2fbcdb116371ba7681482f67d2743c1f3
doc_id: 917018
cord_uid: x4ilyrkr

nan

To the editor We appreciate the interest in our publication [1, 2] . First, we would like to highlight that our study followed a pre-specified protocol, which was based on other protocols for evaluation of COVID-19 vaccine effectiveness applying case-control, test-negative study design conducted in Brazil and in accordance with the World Health Organization (WHO) recommendations [3, 4] .

Regarding the decrease in effectiveness following the first dose, the change overtime is shown in Table 1 for the first dose. We did not evaluate change overtime for the second dose because we had not sample size/power to conduct this analysis. As stated in our discussion, the observed decrease overtime for the first dose can be attributed to some reasons. For instance, waning, which is the natural decline of protection, as well as the surge of Delta variant. Because we do not have data on sequencing, we cannot disentangle this issue further, as well as we cannot obtain an estimate for protection against Gamma and Delta variants.

We agree that evaluation of effectiveness by age is important [3] , but we pre-specified that we would conduct this analysis by the median age. Subgroup analysis could lead to spurious findings if not pre-specified. We have very few elderlies recruited in the study, because of the age distribution of the analysed population in "Complexo da Maré". Finally, our study population comprises only adults, and children were not even receiving COVID-19 vaccines at the time of study, therefore it would be impossible to estimate direct vaccine effectiveness in children.

Regarding potential confounding factors, we excluded previous infected individuals (eFigure 3, supplementary material). We agree that personal behaviour could affect the risk of infection, and if this behaviour is different between vaccinated individuals compared to unvaccinated individuals, we could have residual confounding, as acknowledged in the article [2] . However, as also discussed in the manuscript, our indicator bias did not indicate any remarkable bias, under the assumption of a case-control, test-negative design [5] . We agree cross-protection could influence on vaccine effectiveness estimates if those tested negative individuals were symptomatic by some virus with cross-protection, however this is still a hypothesis for COVID-19 vaccines.

We would like to highlight that our vaccine estimates were in accordance to previous literature and robust to several sensitivity analyses, as shown in the supplementary material [2] .

Otavio T Ranzani¹ ,2 , Fernando A Bozza 3,4 , on behalf of all co-authors

Vaccine effectiveness of ChAdOx1 nCoV-19 against COVID-19 in a socially vulnerable community in Rio de Janeiro, Brazil: a test-negative design study

Vaccine effectiveness of ChAdOx1 nCoV-19 against COVID-19 in a socially vulnerable community in Rio de Janeiro, Brazil: a test-negative design study

Effectiveness of the CoronaVac vaccine in older adults during a gamma variant associated epidemic of covid-19 in Brazil: test negative case-control study

Evaluation of post-introduction COVID-19 vaccine effectiveness: Summary of interim guidance of the World Health Organization

Use of recently vaccinated individuals to detect bias in test-negative case-control studies of COVID-19 vaccine effectiveness. Epidemiology

The authors thank the Redes da Mare for all their support and the efficient strategies on community engagement and commu-nication during the pandemic, as well as the Unidade de Apoio ao Diagno_stico da Covid-19 for their support on the testing diagnosis. OTR acknowledges support from the Spanish Ministry of Science and Innovation and State Research Agency through the "Centro de Excelencia Severo Ochoa 2019e2023" Program (CEX2018-000806-S) and from the Generalitat de Catalunya through the CERCA Program. In addition, the authors thank the Center for Healthcare Operations and Intelligence research group for their discussions and collaborative production of scientific analyses of the COVID-19 pandemic in Brazil.