Professional Documents
Culture Documents
Andrea Brizzi1∗ , Charles Whittaker2∗ , Luciana M. S. Servo3∗ , Iwona Hawryluk2∗ , Carlos A. Prete Jr4∗ , William M.
de Souza5∗ , Renato S. Aguiar6 , Leonardo J. T. Araujo7 , Leonardo S. Bastos8 , Alexandra Blenkinsop1 , Lewis F. Buss9 ,
Darlan Candido10 , Marcia C. Castro11 , Silvia F. Costa9 , Julio Croda12 , Andreza Aruska de Souza Santos 13 , Christo‐
pher Dye10 , Seth Flaxman14 , Paula L. C. Fonseca6 , Victor E. V. Geddes6 , Bernardo Gutierrez10 , Philippe Lemey15 ,
Anna S. Levin16 , Thomas Mellan2 , Diego M. Bonfim6 , Xenia Miscouridou1 , Swapnil Mishra2,17 , Mélodie Monod1 ,
Filipe R. R. Moreira18 , Bruce Nelson 19 , Rafael H. M. Pereira 3 , Otavio Ranzani20 , Ricardo P. Schnekenberg21 , Eliza‐
veta Semenova1 , Raphael Sonnabend2 , Renan P. Souza 6 , Xiaoyue Xi1 , Ester C. Sabino9∗∗ , Nuno R. Faria2,10,16,22∗∗ ,
Samir Bhatt2∗∗ , Oliver Ratmann1∗,∗∗
Affiliations
1
Department of Mathematics, Imperial College London, London, United Kingdom
2
MRC Centre for Global Infectious Disease Analysis, Jameel Institute, School of Public Health, Imperial College
London
3
Institute for Applied Economic Research ‐ IPEA, Brasília, Brazil
4
Departamento de Engenharia de Sistemas Eletrônicos, Escola Politécnica da Universidade de São Paulo, São
Paulo, Brazil
5
World Reference Center for Emerging Viruses and Arboviruses and Department of Microbiology and Immunol‐
ogy, University of Texas Medical Branch, Galveston, TX, USA
6
Departamento de Genética, Ecologia e Evolução, Instituto de Ciências Biológicas, Universidade Federal de Minas
Gerais, Belo Horizonte, Brazil
7
Laboratory of Quantitative Pathology, Center of Pathology, Adolfo Lutz Institute, São Paulo, Brazil
8
Fundação Oswaldo Cruz, Rio de Janeiro, Brazil
9
Departamento de Moléstias Infecciosas e Parasitárias e Instituto de Medicina Tropical da Faculdade de Medicina
da Universidade de São Paulo, São Paulo, Brasil
10
Department of Zoology, University of Oxford, Oxford, United Kingdom
11
Department of Global Health and Population, Harvard T.H. Chan School of Public Health, Boston, United States
12
Department of Epidemiology of Microbial Diseases, Yale School of Public Health, New Haven, United States 13
Latin American Centre, University of Oxford, Oxford, United Kingdom
14
Department of Computer Science, Oxford, United Kingdom
15
Department of Microbiology, Immunology and Transplantation, KU Leuven – University of Leuven, Leuven, Bel‐
gium
16
Department of Infectious Disease Epidemiology, Imperial College London, London, United Kingdom
17
Section of Epidemiology, School of Public Health, University of Copenhagen.
18
Departamento de Genética, Instituto de Biologia, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil
19
Environmental Dynamics, INPA, National Institute for Amazon Research, Bairro Petropolis, Brazil
20
Barcelona Institute for Global Health, ISGlobal, Barcelona, Spain
21
Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom
22
Instituto de Medicina Tropical, Faculdade de Medicina da Universidade de São Paulo, São Paulo, Brazil
∗
Contributed equally, ∗∗ joint senior corresponding authors.
†
Corresponding authors: Oliver Ratmann, oratmann@gmail.com; Samir Bhatt, bhattsamir@gmail.com; Nuno R.
Faria, n.faria@imperial.ac.uk; Ester C. Sabino, sabinoec@usp.br.
SUGGESTED CITATION
A Brizzi, C Whittaker, LMS Servo et al. Factors driving extensive spatial and temporal fluctuations in
COVID-19 fatality rates in Brazilian hospitals - FIGURES AND TABLES. Imperial College London (06-10-2021),
doi https://doi.org/10.25561/91875.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives
4.0 International License.
with PCR confirmed or clinically diagnosed COVID‐19 infection, or severe respiratory infection with no
other confirmed cause. Data capture admissions and deaths until 26 July 2021.
‡ To reported counts were added expected fatal outcomes in COVID‐19 attributable hospital admissions in
physicians and critical care specialists, i. e. intensive care physicians), and reported equipment (critical care beds, ICU beds,
ventilators) from Brazil’s National Register of Health Facilities (Cadastro Nacional de Estabelecimentos de Saúde ‐ CNES);
see Supplementary Text, page 13. Values in brackets give the percent increase from March 2020 to July 2021. CC stands
for critical care.
Location Longest observed duration of COVID‐19 in‐hospital fatality rates above 50%
(in weeks, by age group shown in years)
0‐15 16‐29 30‐39 40‐49 50‐59 60‐69 70‐74 75‐79 80‐84 85‐89 90+
Belo Horizonte 0 0 0 0 0 0 0 0 0 0 0
Curitiba 0 0 0 0 0 0 3 2 2 4 9
Florianópolis 0 0 0 0 0 0 1 7 7 7 21
Goiânia 0 0 0 0 0 0 7 12 11 22 23
João Pessoa 0 0 0 0 0 3 8 19 39 25 19
Macapá 0 7 0 8 9 11 15 20 23 53 31
Manaus 0 0 0 1 4 9 9 13 51 26 50
Natal 0 0 0 0 0 23 15 25 30 48 45
Porto Alegre 0 0 0 0 0 6 9 10 13 56 47
Porto Velho 0 0 0 0 0 8 16 30 25 48 64
Rio de Janeiro 0 0 0 0 0 0 23 72 72 72 72
Salvador 0 0 0 0 3 0 10 11 15 38 29
São Luís 1 2 3 0 0 2 2 13 11 16 14
São Paulo 0 1 0 0 0 4 7 2 8 20 19
Table S3: Longest observed duration of COVID‐19 in‐hospital fatality rates above 50%.
0% 0% 0% 0% 0%
Macapá Manaus Natal Porto Alegre Porto Velho
100% 100% 100% 100% 100%
0% 0% 0% 0% 0%
20 0− 5− 1
20 0− 7− 1
20 0− 9− 1
2021− 1− 1
20 1− 1− 1
20 1− 3− 1
21 05 01
7− 1
01
Rio de Janeiro Salvador São Luís São Paulo
2 0 0
2 0 0
2 0 0
1 0
2 0 0
2 0 0
−0 −0
20 0− 3−
100% 100%
2 0
20 0−
2
20
75% 75% 75% 75%
0% 0% 0% 0%
20 0− 5− 1
20 0− 7− 1
20 0− 9− 1
2021− 1− 1
20 1− 1− 1
20 1− 3− 1
21 05 01
7− 1
01
20 0− 5− 1
20 0− 7− 1
20 0− 9− 1
2021− 1− 1
20 1− 1− 1
20 1− 3− 1
21 05 01
7− 1
01
20 0− 5− 1
20 0− 7− 1
20 0− 9− 1
2021− 1− 1
20 1− 1− 1
20 1− 3− 1
21 05 01
7− 1
01
20 0− 5− 1
20 0− 7− 1
20 0− 9− 1
2021− 1− 1
20 1− 1− 1
20 1− 3− 1
21 05 01
7− 1
01
2 0 0
2 0 0
2 0 0
1 0
2 0 0
2 0 0
−0 −0
2 0 0
2 0 0
2 0 0
1 0
2 0 0
2 0 0
−0 −0
2 0 0
2 0 0
2 0 0
1 0
2 0 0
2 0 0
−0 −0
2 0 0
2 0 0
2 0 0
1 0
2 0 0
2 0 0
−0 −0
20 0− 3−
20 0− 3−
20 0− 3−
20 0− 3−
2 0
2 0
2 0
2 0
20 0−
20 0−
20 0−
20 0−
2
2
20
20
20
20
Figure S1: Spatiotemporal expansion of the SARS‐CoV‐2 Gamma variant across Brazil. SARS‐CoV‐2 genome se‐
quences were obtained from the GISAID repository [1] along with confirmed lineage assignments. The frequency
of the Gamma variant (dots) in weekly SARS‐CoV‐2 genome sequence counts (size of dots) is shown along with
posterior median estimates of Gamma’s variant frequencies (black line) under the Bayesian multi‐strain fatality
model (Supplementary Text page 43) and 95% credible intervals (grey ribbon).
0
200
400
600
0
100
200
300
0
500
1000
0
200
400
600
0
500
1000
20 20
06 October 2021
Goiânia
Manaus
20 −02 1 21 −0 21 −0 20 −02 1 2 1 −0
São Luís
Porto Velho
21 −0 20 −03 1 20 −03 1 21 −0 20 −03 1
Belo Horizonte
20 −04 1 21 −0 21 −0 20 −04 1 2 1 −0
21 −0 20 −05 1 20 −05 1 21 −0 20 −05 1
20 −06 1 21 −0 21 −0 20 −06 1 2 1 −0
21 −0 1 20 −07 1 21 −0 20 −07 1
20 −07
20 −08 1 21 −0 21 −0 20 −08 1 2 1 −0
21 −0 −0 1 −0 1 21 −0 −0 1
−1 1 9− 9− −1 1 9−
0− 01 01 0− 01
01 01
DOI: https://doi.org/10.25561/91875
20 20
Outcome
20 20 20 20 20
0
1000
2000
3000
4000
5000
0
500
1000
1500
2000
0
100
200
300
0
100
200
300
400
0
200
400
600
800
20 20 20
20 −01 20 −01
20 −0 20 −02 20 −02 20 −02 2 0 −0
20 −0 20 −0 20 −0
20 −03 1 20 −03 1
20 −0 20 −04 1 20 −04 1 20 −04 1 2 0 −0
20 −0 20 −0 20 −0
20 −05 1 1 20 −05 1
20 −0 20 −06 1 20 −06 1 20 −06 2 0 −0
20 −0 20 −0 20 −0
20 −07 1 1 20 −07 1
20 −0 20 −08 2 0 −0
Death Covid
20 −08 1 20 −08 1
20 −0 20 −0 20 −0
20 −09 1 1 20 −09 1
20 −0 20 −10 1 20 −10 1 20 −10 2 0 −0
20 −0 20 −0 20 −0
20 −11 1 1 1 − 1 20 −11 1
21 −0 20 −12 20 −12 20 12 2 1 −0
20 −01 1 21 −0 21 −0 21 −0 20 −01 1
Natal
weeks
Curitiba
São Paulo
21 −0 21 −0 21 −0
João Pessoa
20 −03 1 20 −03 1
Rio de Janeiro
21 −0 20 −04 1 20 −04 1 20 −04 1 2 1 −0
Discharged
20 −05 1 21 −0 21 −0 21 −0 20 −05 1
21 −0 1 2 1 −0
20 −06 1 20 −06 1 20 −06
20 −07 1 21 −0 21 −0 21 −0 20 −07 1
21 −0 1 2 1 −0
20 −08 1 20 −08 1 20 −08
−0 1 21 −0 21 −0 21 −0 −0 1
9− −1 1 −1 1 −1 1 9−
in ICU
01 0− 0− 0− 01
01 01 01
20 20
20 20 20 20
0
250
500
750
0
200
400
600
0
25
50
75
100
0
50
100
150
200
20 20
20 −01 20 −01 −0
20 −0 20 −0 20 4− 20 −02
20 −03 1 20 −03 1 20 01 2 0 −0
−0
Unknown
20 −0 20 −0 20 6− 20 −04 1
20 −05 1 20 −05 1 20 01 2 0 −0
20 −0 20 −0 −0 20 −06 1
20 8− 2 0 −0
20 −07 1 20 −07 1 20 01
20 −0 20 −0 −1 20 −08 1
20 −09 1 20 −09 1 20 0− 2 0 −0
20 −0 20 −0 20 01 20 −10 1
outcomes. The date of Gamma’s first detection in each city is shown as a vertical line.
−1 2 0 −0
20 −11 1 20 −11 1 20 2− 1
21 −0 21 −0 21 01 20 −12
20 −01 1 20 −01 1 −0 2 1 −0
20 2−
Macapá
21 −0 21 −0 20 −02 1
Salvador
21 01
2 1 −0
Porto Alegre
−0
Florianópolis
20 −03 1 20 −03 1
21 −0 21 −0 20 4− 20 −04 1
21 01 2 1 −0
20 −05 1 20 −05 1 −0
21 −0 21 −0 20 6− 20 −06 1
20 −07 1 20 −07 1 21 01 2 1 −0
21 −0 21 −0 −0
20 8− 20 −08 1
−0 1 −0 1 21 01 2 1 −0
9− 9− −1 −1 1
01 01 0− 0−
01 01
Figure S2: COVID‐19 attributable hospital admissions among residents without evidence of vaccination. Data are
from the SIVEP‐Gripe platform as of 20 September 2021, and are shown in colours according to reported clinical
Imperial College COVID‐19 Response Team
Page 7 of 45
06 October 2021 Imperial College COVID‐19 Response Team
400 300
60
300
200
40
200
100 20
100
0 0 0
Goiânia João Pessoa Macapá
150 60
200
100 40
observed and expected COVID−19 attributable deaths
100
50 20
0 0 0
Manaus Natal Porto Alegre
1000
400
150
750
300
100
500 200
50
250 100
0 0 0
Porto Velho Rio de Janeiro Salvador
160
300
120 1200
200
80 800
100
40 400
0 0 0
2020− 3−01
2020− 4− 1
2020− 5− 1
20 0− 6− 1
2020− 7− 1
2020− 8− 1
20 0− 9− 1
2020− 0− 1
20 0− 1− 1
2021− 2− 1
2021−01−01
20 1−02−01
2021− 3−01
2021− 4− 1
2021− 5− 1
21 06 01
7− 1
01
São Luís São Paulo
0 0
0 0
0 0
0 0
0 0
0 0
1 0
1 0
1 0
0 0
0 0
−0 −0
2020−02−
0
0
2020−
300
2
2
20
1500
200
1000
100
500
0 0
2020− 3−01
2020− 4− 1
2020− 5− 1
20 0− 6− 1
2020− 7− 1
20 0− 8− 1
2020− 9− 1
20 − 0− 1
2020− 1− 1
2021− 2− 1
20 1−01−01
2021−02−01
2021− 3−01
2021− 4− 1
2021− 5− 1
21 06 01
7− 1
01
2020− 3−01
2020− 4− 1
2020− 5− 1
20 − 6− 1
2020− 7− 1
2020− 8− 1
20 0− 9− 1
2020− 0− 1
20 0− 1− 1
2021− 2− 1
2021−01−01
20 1−02−01
2021− 3−01
2021− 4− 1
2021− 5− 1
21 06 01
7− 1
01
0 0
0 0
0 0
0 0
0 0
0 0
1 0
1 0
1 0
0 0
0 0
−0 −0
0 0
0 0
0 0
0 0
0 0
0 0
1 0
1 0
1 0
0 0
0 0
−0 −0
2020−02−
2020−02−
0
0
2020−
2020−
0
0
2
2
20
20
in public hospitals in private or public hospitals expected deaths among COVID−19 hospitalised patients with as of yet unknown outcome
Figure S3: Underreporting‐adjusted COVID‐19 attributable deaths. Reported COVID‐19 attributable deaths in the
SIVEP‐Gripe platform were adjusted for in‐hospital underreporting, by counting a proportion of hospitalised pa‐
tients with as of yet unknown outcome as fatal, and for likely out‐of‐hospital under‐reporting, by comparison
against population excess deaths derived from all‐cause mortality data of the Brazilian Civil Registry (Supplemen‐
tary Text, page 23). The date of Gamma’s first detection in each city is shown as a vertical dotted line.
0%
20%
40%
0%
20%
40%
20 2
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
20 0−
−0 20 0
7−
20 6− 2
20 01 0− 01
−0 20 10−
20 9−
20 01 21 0
−1 −0 1
20 20 1
21 2−0 21 −0
1
0−15
−
70−74
0−15
−0 1
70−74
20 3
06 October 2021
20 04−
21 −0
−0 1
21 0
−0 1
6− 7−
01 20 01
20 20
20
−0 −
20 6 20 07−
20 −0 20 01
− 1 −
20 09 20 10−
20 −0 21 0
− 1 −0 1
20 12 20 1
21 −0 21 −0
− 1 − 1
75−79
16−29
75−79
16−29
age
20 03
age
20 04−
21 −0
−0 1
21 0
−0 1
6− 7−
01 20 01
20 20
20 −
−
0−15
0−15
70−74
70−74
20 06 20 07−
20 −0 20 01
− 1 −
20 09 20 10−
20 −0 21 0
− 1 −0 1
20 12
21 −0
20 1
1 21 −0
−
75−79
16−29
80−84
30−39
75−79
16−29
80−84
30−39
20 03 −0 1
21 −0 20 4
−0 1 21 −0
6− −0 1
01 7−
20 20 01
DOI: https://doi.org/10.25561/91875
20 20
− −
80−84
30−39
80−84
30−39
20 06 20 07−
20 −0 20 01
− 1 −
20 09 20 10−
20 −0
1
21 0
− −0 1
20 12 20 1
21 −0
85−89
40−49
85−89
40−49
− 1 21 −0
85−89
40−49
85−89
40−49
20 03 −0 1
21 −0 20 4
−0 1 21 −0
6− −0 1
01 7−
90+
01
90+
20 20
50−59
20 20
50−59
− −
20 06 20 07−
20 −0 20 01
− 1 −
20 09 20 10−
20 −0
− 1 21 0
20 12 −0 1
1
60−69
20
60−69
21 −0
90+
− 1
90+
21 −0
50−59
50−59
20 03 −0 1
21 −0 20 4
−0 1 21 −0
6− −0 1
01 7−
a vertical dotted black line. (A) For Belo Horizonte. (B) For Curitiba.
20 01 20
20 20
− −
20 06 20 07−
20 −0 20 01
− 1 −
20 09 20 10−
20 −0 0
− 1 21
−0 1
20 12
21 −0 20 1
− 1 21 −0
60−69
60−69
20 03 −0 1
21 −0 20 4
−0 1 21 −0
6− −0 1
01 7−
01
solid line along with 95% confidence intervals as grey ribbon. The date of Gamma’s first detection is indicated as
hospital fatality rates are shown as dots, and non‐parametric loess estimates of time trends are shown as block
Figure S4: Time trends in age‐specific COVID‐19 in‐hospital fatality rates. Weekly, age‐specific COVID‐19 in‐
Imperial College COVID‐19 Response Team
Page 9 of 45
E
C
D
smoothed weekly in−hospital fatality rate smoothed weekly in−hospital fatality rate smoothed weekly in−hospital fatality rate
20
2
0%
40%
80%
0%
40%
80%
20 20
0%
50%
100%
0%
50%
100%
0%
50%
100%
150%
0%
50%
100%
150%
20 0− 20−
−0 20 06 0
20 6− 20 −0 20 6−
20 01 − 1 20− 01
−0 20 09
20 9− 20 −0 20 09
20 01 − 1 20 −0
−1 20 12 −1 1
20 2− 21 −0 20 2
21 − 1 21 −0
0−15
01 1
−
70−74
0−15
0−15
−0 20 03
70−74
70−74
20 3− 20 03
06 October 2021
21 21 −0 21 −0
01 −0 1
−0 6− −0 1
6− 01 6−
01 20 01
20 20 20
20 − 20
−0 20 06 −
20 6− 20 −0 20 06
20 01 − 1 20 −0
−0 20 09 − 1
20 9− 20 −0 20 09
20 01 − 1 20 −0
−1 20 12 − 1
20 2− 21 −0 20 12
21 01 − 1 21 −0
1
−
75−79
16−29
−0 20 03
75−79
16−29
75−79
16−29
age
20 20 03
age
age
3− 21 −0
21 01 −0 1 21 −0
−0 6− −0 1
6− 01 6−
01 20 01
20 20 20
−
0−15
−0
0−15
0−15
20 06
70−74
20 20 06
70−74
70−74
6− 20 −0
20 01 − 1 20 −0
1
−0 20 09 −
20 9− 20 −0 20 09
20 01 − 1 20 −0
1
−1 20 12 −
20 2− 21 −0 20 12
21 01 − 1 21 −0
1
−
75−79
16−29
80−84
30−39
−0
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
20 20 03 20 03
3− 21 −0
21 01 −0 1 21 −0
−0 6− −0 1
6− 01 6−
01 20 01
20 20
DOI: https://doi.org/10.25561/91875
20
20 − 20
−
80−84
30−39
−0
80−84
30−39
80−84
30−39
20 20 06 20 06
6− 20 −0
20 01 − 1 20 −0
1
−0 20 09 −
20 9− 20 −0 20 09
20
−1
01 − 1 20 −0
− 1
20 20 12 20 12
2− 21 −0
21 1 21 −0
85−89
40−49
01 − 1
85−89
40−49
85−89
40−49
−
85−89
40−49
−0
85−89
40−49
85−89
40−49
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 01
90+
20
90+
90+
20 20 20
50−59
20 − 20
50−59
50−59
−0 20 06 −
20 6− 20 06
20 −0
20
−0
01 − 1 20 −0
− 1
20 9− 20 09 20 09
20 −0
20
−1
01 − 1 20 −0
− 1
20 2− 20 12 20 12
60−69
60−69
60−69
21 21 −0 21 −0
01 − 1 1
90+
−
90+
90+
−0
50−59
50−59
50−59
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 01
20 20 20
20 20 20
−0 − −0
20 6− 20 06
20 −0
20 6
−0
20 1 20
−0
01 − −0 1
20 9− 20 09 20 9−
20 01 20 −0
− 1 20 0
−1 20 12 −1 1
20 2− 20 2−
21 21 −0 21
−0
01 − 1 −
01
60−69
60−69
60−69
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 01
Figure S4: (continued) Time trends in age‐specific COVID‐19 in‐hospital fatality rates. (C) For Florianópolis. (D)
Imperial College COVID‐19 Response Team
Page 10 of 45
F
H
G
smoothed weekly in−hospital fatality rate smoothed weekly in−hospital fatality rate smoothed weekly in−hospital fatality rate
20 20 20
2 2
0%
40%
80%
120%
0%
40%
80%
120%
0%
40%
80%
120%
0%
40%
80%
120%
−50%
0%
50%
100%
150%
−50%
0%
50%
100%
150%
20 0− 0−
−0 20 05 20 06
20 6− 20 −0
1 20 −0
20
−0
01 − − 1
20 20 08 20 09
9− 20 −0
20 01 − 1 20 −0
− 1
−1 20 11 20 12
20 2− 21 −0
21 01 1 21 −0
1
20 −02 −
0−15
0−15
−0
0−15
70−74
70−74
70−74
20 20 03
3− 21 −0 −
06 October 2021
21 01 −0 1 21 0
−0 5− −0 1
6− 01 6−
01 20 20 01
20 20
20 20
−0 − −
20 20 05 20 06
6− 20 −0 20 −0
20 − 1 1
75−79
16−29
75−79
16−29
75−79
16−29
20 21 −0 20 03
age
age
age
3−
21 01 −0 1 21 −0
−0 5− −0 1
6− 01 6−
01 20 20 01
20 20
20 20
−0 − −
0−15
0−15
0−15
20 20 05 20 06
70−74
70−74
70−74
6− 20 −0 20 −0
20
−0
01 − 1 − 1
20 20 08 20 09
9− 20 −0
20 01 − 1 20 −0
− 1
−1 20 11 20 12
20 2− 21 −0
21 01 1 21 −0
1
−0 20 −02 −
75−79
16−29
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
80−84
30−39
20 3− 21 −0 20 03
21 01 −0 1 21 −0
−0 5− −0 1
6− 01 6−
01 20 20 01
20
DOI: https://doi.org/10.25561/91875
20 20 20
−0 − −
80−84
30−39
80−84
30−39
80−84
30−39
20 20 05 20 06
6− 20 −0 20 −0
20
−0
01 − 1 − 1
20 20 08 20 09
9− 20 −0
20 01 − 1 20 −0
− 1
−1 20 11 20 12
20 2− 21 −0
21 01 1 21 −0
1
−
85−89
40−49
85−89
40−49
85−89
40−49
−0 20 −02
85−89
40−49
85−89
40−49
85−89
40−49
20 3− 21 −0 20 03
21 01 −0 1 21 −0
−0 5− −0 1
6− 01 6−
01 20 20 0 1
20
90+
90+
90+
20 20 20
− −
50−59
50−59
50−59
−0 20 05 20 06
20 6− 20 − 20 −0
20 01 − 01
−0 20 08 −0 1
20 9− 20 −0 20 9
20 01 − 1 20 −0
− 1
−1 20 11 20 12
20 2− 21 −0
1
60−69
60−69
60−69
21 21 −0
01 − 1
90+
90+
90+
−0 20 −02
50−59
50−59
50−59
20 3− 21 −0 20 03
21 01 −0 1 21 −0
−0 5− −0 1
6− 01 6−
01 20 01 20
20 20 20
20 − −0
−0
20 6−
20 05
20 −0
20 6−0
20
20
−0
01 − 1 −0 1
20 20 08 20 9−
9− 20 −0
20 01 − 1 20 0
−1 20 11 −1 1
20 2− 21 −0 20 2
21 01 1 21 −0
1
−0 20 −02 −
60−69
60−69
60−69
20 3− 21 −0 20 03
21 01 −0 1 21 −0
−0 5− −0 1
6− 01 6−
01 01
Figure S4: (continued) Time trends in age‐specific COVID‐19 in‐hospital fatality rates. (F) For Macapá. (G) For
Imperial College COVID‐19 Response Team
Page 11 of 45
I
K
smoothed weekly in−hospital fatality rate smoothed weekly in−hospital fatality rate smoothed weekly in−hospital fatality rate
20 20
2 20
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
40%
80%
120%
0%
40%
80%
120%
−50%
0%
50%
100%
150%
−50%
0%
50%
100%
150%
20 0− 20−
−0
20 6− 20 06 20 0
6−
20 −0 2
20
−0
01 − 1 0− 01
20 9− 20 09 20 09
20 20 −0
−1
01 − 1 20 −0
20 20 12 −1 1
2− 20
21 21 −0 21 2−0
01 − 1 − 1
0−15
0−15
−0
0−15
70−74
70−74
70−74
20 20 03 20 03
3− −
06 October 2021
21 01 21 0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 20 01 01
20 20
20 20
− 20
20
−0 20 06 −
6− 20 −0 20 06
20 01 − 1 20 −0
−0 20 09 − 1
20 9− 20 09
20 20 −0
01 − 1 20 −0
1
20
−1 20 12 −
2− 20 12
21 21 −0 21 −0
−0
01 − 1 − 1
75−79
16−29
75−79
16−29
75−79
16−29
20 20 03 20 03
age
age
age
3−
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 20 01 01
20 20
20 20
− 20
−0 −
0−15
0−15
0−15
20 20 06
70−74
70−74
70−74
6− 20 −0 20 06
20 01 − 1 20 −0
− 1
75−79
16−29
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
80−84
30−39
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 20 01 01
20 20
DOI: https://doi.org/10.25561/91875
20 20
− 20
−0 −
80−84
30−39
80−84
30−39
80−84
30−39
20 6− 20 06 20 06
20 −0
20 01 − 1 20 −0
−0 20 09 − 1
20 9− 20 09
20 20 −0
01 − 1 20 −0
1
20
−1 20 12 −
2− 20 12
21 21 −0
01 − 1 21 −0
85−89
40−49
85−89
40−49
85−89
40−49
−0 − 1
85−89
40−49
85−89
40−49
85−89
40−49
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 20 0 1 01
20
90+
90+
90+
20 20
20
− 20
50−59
50−59
50−59
20
−0 20 06 −
6− 20 −0 20 06
20 01 20 −0
−0 −0 1 − 1
20 9− 20 9 20 09
20 20 −0
01 − 1 20 −0
1
20
−1 20 12 −
2− 20 12
60−69
60−69
60−69
21 21 −0
01 − 1 21 −0
− 1
90+
90+
90+
−0
50−59
50−59
50−59
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 20 01
20 20 20
20 − 20
−0 20 06 −0
20 6− 20 6
20 −0
20
−0
01 − 1 20 −0
20 20 09 −0 1
9− 20 −0 20 9−
20 01 1 20 0
−1 − −1 1
20 2− 20 12 20
21 21 −0
−0
01 − 1 21 2−0
− 1
60−69
60−69
60−69
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 01
Figure S4: (continued) Time trends in age‐specific COVID‐19 in‐hospital fatality rates. (I) For Porto Alegre. (J) For
Imperial College COVID‐19 Response Team
Page 12 of 45
L
N
M
smoothed weekly in−hospital fatality rate smoothed weekly in−hospital fatality rate smoothed weekly in−hospital fatality rate
20 20 20
2 2
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
50%
100%
0%
50%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
20 0− 0−
20 −04
20 −0
20 05 20 0
6−
20 −0 2
20 −07 1 − 1 0− 01
20 08 20 09
20 −0 20 −0
20 −10 1 − 1 20 −0
20 11 −1 1
21 −0 21 −0 20 2
20 −01 1 1 21 −0
1
20 −02 −
0−15
0−15
0−15
70−74
70−74
70−74
21 −0 20 03
21 −0
06 October 2021
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
20 7− 20 01
20 01 20 20
− 20
20 −04 20 05 −
20 −0 20 −0 20 06
− − 1 20 −0
1
20 07 1 20 08 −
20 −0 20 −0 20 09
− 1 20 −0
1
20 −10 1
21 −0 20 11 −
21 −0 20 12
1 21 −0
75−79
16−29
75−79
16−29
75−79
16−29
21 −0 20 03
21 −0
age
age
age
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
20 7− 20 01
20 01 20 20
− 20
20 −04 −
0−15
0−15
0−15
20 05 20 06
70−74
70−74
70−74
20 −0 20 −0
− − 1 20 −0
1
20 07 1 20 08 −
20 −0 20 −0 20 09
− 1 20 −0
1
20 −10 1
21 −0 20 11 −
21 −0 20 12
20 −01 1 1 21 −0
1
20 −02 −
75−79
16−29
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
80−84
30−39
21 −0 20 03
21 −0
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
20 7− 20 01
01 20
DOI: https://doi.org/10.25561/91875
20 20 20
20 −04 − −
80−84
30−39
80−84
30−39
80−84
30−39
20 05 20 06
20 −0 20 −0
− − 1 20 −0
1
20 07 1 20 08 −
20 −0 20 −0 20 09
− 1 20 −0
1
20 −10 1
21 −0 20 11 −
21 −0 20 12
20 −01 1 1 21 −0
1
85−89
40−49
85−89
40−49
85−89
40−49
20 −02 −
85−89
40−49
85−89
40−49
85−89
40−49
21 −0 20 03
21 −0
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
20 7− 20 01
20
90+
90+
90+
20 01 20
− 20
50−59
50−59
50−59
20 −04 20 05 −
20 −0 20 − 20 06
− − 01 20 −0
1
20 07 1 20 08 −
20 −0 20 −0 20 09
− 1 20 −0
1
20 −10 1
21 −0 20 11 −
21 −0 20 12
1
60−69
60−69
60−69
21 −0
20 −01 1
− 1
90+
90+
90+
20 −02
50−59
50−59
50−59
21 −0 20 03
21 −0
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
7− 20 20 01
01 20 20 20
− 20
20 −04 20 05 −0
20 −0 20 −0 20 6
−0
20 −07 1 − 1 20
−0 1
20 08 20
20 −0 20 −0 9−
20 −10 1 − 1 20 0
21 −0 20 11 −1 1
21 −0 20 2−
20 −01 1 1 21 01
20 −02 −
60−69
60−69
60−69
21 −0 20 03
21 −0
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
7− 01
01
Figure S4: (continued) Time trends in age‐specific COVID‐19 in‐hospital fatality rates. (L) For Salvador. (M) For São
Imperial College COVID‐19 Response Team
Page 13 of 45
06 October 2021 Imperial College COVID‐19 Response Team
600
Physicians
400
200
Intensive Care
6
specialists
4
80
adult ICU
60
beds
40
20
75
Critical care
beds
resource per 100,000
50
25
160
Ventilators
120
80
40
300
Nurses
250
200
150
100
800
Nurse assistants
600
400
200
120
Physiotherapists
90
60
30
20 −03 1
20 −05 1
20 −07 1
20 −09 1
20 −11 1
20 −0 1
20 −03 1
20 −05 1
20 1−0 01
20 −0 01
20 −03 1
20 −05 1
20 −07 1
20 −0 1
20 −11 1
20 −0 1
20 −0 1
20 −05 1
20 −0 01
20 −0 01
20 −03 1
20 −05 1
20 −07 1
20 −09 1
20 −1 1
20 −0 1
20 −03 1
20 −05 1
20 −0 01
20 −0 01
20 −0 1
20 −05 1
20 −07 1
20 −09 1
20 −11 1
20 −0 1
20 −03 1
20 −05 1
−0 01
01
0
0
0
20 −0
0
0
0
0
0
0
0
20 −0
0
0
0
0
0
0
0
20 −0
0
0
0
0
0
0
0
20 −0
0
0
0
0
20 1−
20 −
20 −
20 −
21 −
21 −
21 −
−
20 7−
20 1−
20 −
20 −
20 9−
21 −
21 1−
21 3−
21 −
20 7−
20 1−
20 −
20 −
20 −
21 1−
21 1−
21 −
21 −
20 7−
20 1−
20 3−
20 −
20 −
21 −
21 1−
21 −
21 −
7−
1
20 −0
20
2
20
Natal Porto Alegre Porto Velho Rio de Janeiro Salvador São Luís São Paulo
Figure S5: Healthcare resources in hospital settings, per 100,000. Monthly data on health care resources in each
location were obtained from the National Register of Health Facilities (Cadastro Nacional de Estabelecimentos de
Saúde ‐ CNES), and were aggregated from facility‐level microdata to city‐level as described in Supplementary Text,
page 13.
0.1
0.2
0.3
0.4
0.1
0.2
0.3
0.4
0.1
0.2
0.3
0.4
0.5
1.0
1.5
2.0
0.5
1.0
1.5
2.0
0.5
1.0
1.5
2.0
20 −0 20 −0
20 3− 20 3−
0 0
20 −0 1 20 −0 1
20 5− 20 5−
0 0
r=0.5
r=0.44
r=0.35
r=0.59
r=0.54
r=0.58
20 −07 1 20 −07 1
20 −0 20 −0
20 −0 1 20 −0 1
20 9− 20 9−
− 0 − 0
20 1 1 20 1 1
21 1− 21 1−
0 0
20 −0 1 20 −0 1
Macapá
Macapá
21 1− 21 1−
0 0
06 October 2021
20 −0 1 20 −0 1
Rio de Janeiro
Rio de Janeiro
Belo Horizonte
21 3−
0
21 3−
0 Belo Horizonte
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
20 7− 20 7−
20 01 20 01
20 −0 20 −0
20 3− 20 3−
0 0
20 −0 1 20 −0 1
20 − 5 20 − 5
0 0
r=0.3
r=0.8
r=0.7
r=0.3
r=0.81
r=0.54
20 −07 1 20 −07 1
20 −0 20 −0
20 −0 1 20 −0 1
20 9− 20 9−
− 0 − 0
20 1 1 20 1 1
21 1− 21 1−
0 0
20 −0 1 20 −0 1
Curitiba
Curitiba
Manaus
Manaus
21 1− 21 1−
Salvador
Salvador
0 0
20 −0 1 20 −0 1
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
20 7− 20 7−
20 01 20 01
20 −0 20 −0
20 3− 20 3−
0 0
20 −0 1 20 −0 1
20 5− 20 5−
0 0
r=0.3
r=0.47
r=0.75
r=0.49
r=0.69
r=0.26
20 −07 1 20 −07 1
20 −0 20 −0
20 −0 1 20 −0 1
20 9− 20 9−
− 0 − 0
20 1 1 20 1 1
DOI: https://doi.org/10.25561/91875
21 1− 21 1−
0 0
20 −0 1 20 −0 1
Natal
Natal
21 1− 21 1−
São Luís
São Luís
0 0
Florianópolis
Florianópolis
20 −0 1 20 −0 1
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
20 7− 20 7−
20 01 20 01
20 −0 20 −0
20 3− 20 3−
0 0
20 −0 1 20 −0 1
20 5− 20 5−
0 0
r=0.73
r=0.72
r=0.67
r=0.67
20 −07 1 20 −07 1
r=−0.26
r=−0.27
20 −0 20 −0
20 −0 1 20 −0 1
Goiânia
Goiânia
21 1− 21 1−
0 0
São Paulo
São Paulo
− −
Porto Alegre
Porto Alegre
20 0 1 20 0 1
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
7− 20 7− 20
01 20 01 20
20 −0 20 −0
20 3− 20 3−
0 0
0%
0%
20%
40%
60%
20%
40%
60%
20 −0 1 20 −0 1
20 5− 20 5−
0 0
r=0.65
r=0.51
r=0.72
r=0.63
20 −07 1 20 −07 1
20 −0 20 −0
20 −0 1 20 −0 1
20 9− 20 9−
− 0 − 0
20 1 1 20 1 1
21 1− 21 1−
0 0
20 −0 1 20 −0 1
21 1− 21 1−
− 0 − 0
Porto Velho
Porto Velho
João Pessoa
João Pessoa
20 0 1 20 0 1
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
7− 7−
01 01
0%
0%
0%
0%
20%
40%
60%
20%
40%
60%
20%
40%
60%
20%
40%
60%
Figure S6: Time evolution of pandemic healthcare pressure indices. (A) ICU admissions in this and the following
standardised in‐hospital fatality rates are shown in black, with y‐axis on the right. Pearson correlation coefficients
(r) are shown in the upper left corner, and dates of Gamma’s first detection as vertical dotted lines. (B) ICU
two weeks per ICU bed with ventilator are shown in colour, with y‐axis on the left. The biweekly, empirical age‐
Imperial College COVID‐19 Response Team
Page 15 of 45
C
D
20 3−wk ICU admissions per ventilator 20 3−wk ICU admissions per physician
20 20
0.2
0.4
0.6
0.2
0.4
0.6
0.2
0.4
0.6
0.1
0.2
0.3
0.1
0.2
0.3
0.1
0.2
0.3
20 −0 20 −0
20 3− 20 3−
0 0
20 −0 1 20 −0 1
20 5− 20 5−
0 0
r=0.46
r=0.52
r=0.62
r=0.44
r=0.32
r=0.59
20 −07 1 20 −07 1
20 −0 20 −0
20 −0 1 20 −0 1
20 9− 20 9−
− 0 − 0
20 1 1 20 1 1
21 1− 21 1−
0 0
20 −0 1 20 −0 1
Macapá
Macapá
21 1− 21 1−
0 0
06 October 2021
20 −0 1 20 −0 1
Rio de Janeiro
Rio de Janeiro
Belo Horizonte
Belo Horizonte
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
20 7− 20 7−
20 01 20 01
20 −0 20 −0
20 3− 20 3−
0 0
20 −0 1 20 −0 1
20 − 5 20 − 5
0 0
r=0.3
r=0.81
r=0.66
r=0.28
r=0.79
r=0.74
20 −07 1 20 −07 1
20 −0 20 −0
20 −0 1 20 −0 1
20 9− 20 9−
− 0 − 0
20 1 1 20 1 1
21 1− 21 1−
0 0
20 −0 1 20 −0 1
Curitiba
Curitiba
Manaus
Manaus
21 1− 21 1−
Salvador
Salvador
0 0
20 −0 1 20 −0 1
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
20 7− 20 7−
20 01 20 01
20 −0 20 −0
20 3− 20 3−
0 0
20 −0 1 20 −0 1
20 5− 20 5−
0 0
r=0.3
r=0.47
r=0.71
r=0.29
r=0.47
20 −07 1 20 −07 1 r=0.76
20 −0 20 −0
20 −0 1 20 −0 1
20 9− 20 9−
− 0 − 0
20 1 1 20 1 1
DOI: https://doi.org/10.25561/91875
21 1− 21 1−
0 0
20 −0 1 20 −0 1
Natal
Natal
21 1− 21 1−
São Luís
São Luís
0 0
Florianópolis
Florianópolis
20 −0 1 20 −0 1
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
r=0.7
r=0.7
r=0.75
r=0.72
20 −07 1 20 −07 1
r=−0.26
20 −0 20 −0 r=−0.26
20 −0 1 20 −0 1
20 9− 20 9−
− 0 − 0
20 1 1 20 1 1
21 1− 21 1−
0 0
20 −0 1 20 −0 1
Goiânia
Goiânia
21 1− 21 1−
0 0
São Paulo
São Paulo
− −
Porto Alegre
Porto Alegre
20 0 1 20 0 1
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
7− 20 7− 20
01 20 01 20
20 −0 20 −0
20 3− 20 3−
0 0
0%
0%
20%
40%
60%
20%
40%
60%
20 −0 1 20 −0 1
20 5− 20 5−
0 0
r=0.68
r=0.56
r=0.63
r=0.51
20 −07 1 20 −07 1
20 −0 20 −0
20 −0 1 20 −0 1
20 9− 20 9−
− 0 − 0
20 1 1 20 1 1
21 1− 21 1−
0 0
20 −0 1 20 −0 1
21 1− 21 1−
− 0 − 0
Porto Velho
Porto Velho
João Pessoa
João Pessoa
20 0 1 20 0 1
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
7− 7−
01 01
0%
0%
0%
0%
20%
40%
60%
20%
40%
60%
20%
40%
60%
20%
40%
60%
Figure S6: (continued) (C) ICU admissions in this and the following two weeks per physician. (D) ICU admissions
Imperial College COVID‐19 Response Team
Page 16 of 45
F
E
20 3−wk ICU admissions per physiotherapist 3−wk ICU admissions per nurse assistant
20 20
0.4
0.8
1.2
1.6
0.4
0.8
1.2
1.6
0.4
0.8
1.2
1.6
20
0.05
0.10
0.15
0.20
0.05
0.10
0.15
0.20
0.05
0.10
0.15
0.20
20 −0
20 3− 20 −0
0 20 3−
20 −0 1 0
20 5− 20 −0 1
0 20 5−
r=0.6
0
r=0.44
r=0.32
r=0.3
r=0.6
20 −07 1
r=0.41
20 −0 20 −07 1
20 −
20 −0 1 − 0
20 9− 20 0 1
− 0 20 9−
20 1 1 0
21 1− 20 −1 1
0 21 1−
20 −0 1 −
20 0 01
Macapá
21 1−
0
06 October 2021
Macapá
21 1−
20 −0 1
Rio de Janeiro
Belo Horizonte
20 −0 01
Rio de Janeiro
21 3−
Belo Horizonte
0 21 3−
0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
20 7− 7−
20 01 20
20 01
20 −0 20 −0
20 3−
0 20 3−
0
20 −0 1
20 − 5 20 −0 1
0 20 5−
0
r=0.8
r=0.29
r=0.64
r=0.3
20 −07 1
r=0.79
r=0.66
20 −0 20 −07 1
20 −
20 −0 1 0
20 9− 20 −0 1
− 0 20 9−
0
20 1 1 20 −1 1
21 1−
0 21 1−
20 −0 1 −
20 0 01
Curitiba
Manaus
21 1−
Salvador
0
Curitiba
Manaus
21 1−
Salvador
20 −0 1 20 −0 01
21 3−
0 21 3−
0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
20 7− 7−
20 01 20
20 01
20 −0 20 −0
20 3−
0 20 3−
0
20 −0 1 20 −0 1
20 5−
0 20 5−
0
r=0.46
r=0.74
r=0.33
20 −07 1
r=0.47
r=0.75
r=0.31
20 −0 20 −07 1
20 −
20 −0 1 0
20 −0 1
20 9−
− 0 20 9−
0
20 1 1
DOI: https://doi.org/10.25561/91875
20 −1 1
21 1−
0 21 1−
20 −0 1
Natal
20 −0 01
Natal
21 1−
São Luís
0 21 1−
São Luís
Florianópolis
20 −0 1
Florianópolis
20 −0 01
21 3−
0 21 3−
0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
20 7− 20 7−
20 01 20 01
20 −0 20 −0
20 3−
0 20 3−
0
20 −0 1 20 −0 1
20 5− 5
0 20 −
0
r=0.72
r=0.72
r=0.66
r=0.72
20 −07 1
r=−0.26
20 −07 1
20 −0 20 −
0 r=−0.26
20 −0 1 20 −0 1
20 9−
− 0 20 9−
− 0
20 1 1 20 1 1
21 1−
0 21 1−
20 −0 1 20 −0 01
Goiânia
Goiânia
21 1− 21 1−
0
São Paulo
−
São Paulo
20 0 1 20 −0 01
21 3−
0 21 3−
0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
7− 20 7− 20
01 20 01 20
20 −0 20 −0
20 3− 20 3−
0 0
0%
0%
20%
40%
60%
20%
40%
60%
20 −0 1 20 −0 1
20 5− 20 5−
0 0
r=0.66
r=0.54
r=0.64
r=0.51
20 −07 1 20 −07 1
20 −0 20 −
0
20 −0 1 20 −0 1
20 9− 20 9−
− 0 − 0
20 1 1 20 1 1
21 1− 21 1−
0
20 −0 1 20 −0 01
21 1− 21 1−
− 0 −
Porto Velho
Porto Velho
João Pessoa
João Pessoa
20 0 1 20 0 01
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
7− 7−
01 01
0%
0%
0%
0%
20%
40%
60%
20%
40%
60%
20%
40%
60%
20%
40%
60%
Figure S6: (continued) (E) ICU admissions in this and the following two weeks per nurse assistant. (F) ICU admis‐
Imperial College COVID‐19 Response Team
Page 17 of 45
H
G
3−wk SARI admissions per nurse 20 3−wk SARI admissions per physician
20 20
0.3
0.6
0.9
0.3
0.6
0.9
0.3
0.6
0.9
20
0.25
0.50
0.75
1.00
1.25
0.25
0.50
0.75
1.00
1.25
0.25
0.50
0.75
1.00
1.25
20 −0
20 −0 20 3−
20 3− 0
0 20 −0 1
20 −0 1 20 5−
20 5− 0
0
r=0.47
r=0.14
r=0.47
20 −07 1
r=0.47
r=0.18
r=0.48
20 −07 1 20 −0
20 −
− 0 20 −0 1
20 0 1 20 9−
20 9− − 0
0 20 1 1
20 −1 1 21 1−
21 1− 0
−
20 0 01 20 −0 1
Macapá
21 1−
0
06 October 2021
Macapá
21 1−
20 −0 1
Rio de Janeiro
Belo Horizonte
20 −0 01
Rio de Janeiro
21 3−
Belo Horizonte
21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
7− 20 7−
20 20 01
20 01
20 −0 20 −0
20 3−
20 3−
0 0
20 −0 1
20 −0 1 20 − 5
20 5−
0 0
r=0.3
r=0.81
r=0.73
20 −07 1
r=0.32
r=0.82
r=0.69
20 −07 1 20 −0
20 −
0 20 −0 1
20 −0 1 20 9−
20 9−
0 − 0
20 −1 1 20 1 1
21 1−
21 1− 0
−
20 0 01 20 −0 1
Curitiba
Manaus
21 1−
Salvador
0
Curitiba
Manaus
21 1−
Salvador
20 −0 01 20 −0 1
21 3−
21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
7− 20 7−
20 20 01
20 01
20 −0 20 −0
20 3−
20 3−
0 0
20 −0 1 20 −0 1
20 5−
20 5−
0 0
r=0.3
r=0.3
r=0.7
r=0.71
r=0.31
20 −07 1
r=0.31
20 −07 1 20 −0
20 −
0 20 −0 1
20 −0 1
20 9−
20 9−
0 − 0
20 1 1
DOI: https://doi.org/10.25561/91875
20 −1 1
21 1−
21 1− 0
20 −0 1
Natal
20 −0 01
Natal
21 1−
São Luís
Florianópolis
20 −0 1
Florianópolis
20 −0 01
21 3−
21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
20 7− 20 7−
20 01 20 01
20 −0 20 −0
20 3−
20 3−
0 0
20 −0 1 20 −0 1
5 20 5−
20 −
0 0
r=0.6
r=0.63
r=0.72
r=0.73
20 −07 1
r=−0.28
20 −07 1
r=−0.28
20 − 20 −0
0 20 −0 1
20 −0 1
20 9−
20 9−
− 0 − 0
20 1 1 20 1 1
21 1−
21 1− 0
20 −0 01 20 −0 1
Goiânia
Goiânia
21 1− 21 1−
0
São Paulo
São Paulo
Porto Alegre
Porto Alegre
20 −0 01 20 0 1
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
7− 20 7− 20
01 20 01 20
20 −0 20 −0
20 3− 20 3−
0 0
0%
0%
20%
40%
60%
20%
40%
60%
20 −0 1 20 −0 1
20 5− 20 5−
0 0
r=0.56
r=0.46
r=0.59
r=0.46
20 −07 1 20 −07 1
20 − 20 −0
0
20 −0 1 20 −0 1
20 9− 20 9−
− 0 − 0
20 1 1 20 1 1
21 1− 21 1−
0
20 −0 01 20 −0 1
21 1− 21 1−
− − 0
Porto Velho
Porto Velho
João Pessoa
João Pessoa
20 0 01 20 0 1
21 3− 21 3−
0 0
20 −0 1 20 −0 1
21 5− 21 5−
−0 01 −0 01
7− 7−
01 01
0%
0%
0%
0%
20%
40%
60%
20%
40%
60%
20%
40%
60%
20%
40%
60%
Figure S6: (continued) (G) SARI admissions in this and the following two weeks per physician. (H) SARI admissions
Imperial College COVID‐19 Response Team
Page 18 of 45
I
J
20 20
20 3−wk SARI admissions per critical bed with ventilator 20 3−wk SARI admissions per ventilator
2
4
6
8
2
4
6
8
2
4
6
8
1
2
3
1
2
3
1
2
3
20 −03 20 −03
20 − 20 −
0 0
20 −05 1 20 −05 1
20 − 20 −
0 0
r=0.5
r=0.5
r=0.34
r=0.48
r=0.49
r=0.36
20 −07 1 20 −07 1
20 −0 20 −0
20 −09 1 20 −09 1
20 − 20 −
0 0
20 −11 1 20 −11 1
21 − 21 −
0 0
20 −0 1 20 −0 1
Macapá
Macapá
21 1− 21 1−
0 0
06 October 2021
20 −03 1 20 −03 1
Rio de Janeiro
Rio de Janeiro
Belo Horizonte
Belo Horizonte
21 − 21 −
0 0
20 −05 1 20 −05 1
21 − 21 −
−0 01 −0 01
20 7− 20 7−
20 01 20 01
20 −03 20 −03
20 − 20 −
0 0
20 −05 1 20 −05 1
20 − 20 −
0 0
r=0.32
r=0.83
r=0.62
r=0.32
r=0.83
r=0.67
20 −07 1 20 −07 1
20 −0 20 −0
−
20 09 1 −
20 09 1
20 − 20 −
0 0
20 −11 1 20 −11 1
21 − 21 −
0 0
20 −0 1 20 −0 1
Curitiba
Curitiba
Manaus
Manaus
21 1− 21 1−
Salvador
Salvador
0 0
20 −03 1 20 −03 1
21 − 21 −
0 0
20 −05 1 20 −05 1
21 − 21 −
−0 01 −0 01
20 7− 20 7−
20 01 20 01
20 −03 20 −03
20 − 20 −
0 0
20 −05 1 20 −05 1
20 − 20 −
0 0
r=0.3
r=0.27
r=0.63
r=0.29
r=0.28
20 −07 1
20 −0
20 −07 1
20 −0
r=0.64
−
20 09 1 −
20 09 1
20 − 20 −
0 0
20 −11 1 20 −11 1
DOI: https://doi.org/10.25561/91875
21 − 21 −
0 0
20 −0 1 20 −0 1
Natal
Natal
21 1− 21 1−
São Luís
São Luís
0 0
Florianópolis
Florianópolis
20 −03 1 20 −03 1
21 − 21 −
0 0
20 −05 1 20 −05 1
21 − 21 −
−0 01 −0 01
20 7− 20 7−
20 01 20 01
20 −03 20 −03
20 − 20 −
0 0
20 −05 1 20 −05 1
20 − 20 −
0 0
r=0.52
r=0.68
r=0.55
r=0.71
20 −07 1 20 −07 1
r=−0.28
r=−0.28
20 −0 20 −0
20 −09 1 20 −09 1
20 − 20 −
0 0
20 −11 1 20 −11 1
21 − 21 −
0 0
20 −0 1 20 −0 1
Goiânia
Goiânia
21 1− 21 1−
0 0
São Paulo
São Paulo
− −
Porto Alegre
Porto Alegre
20 03 1 20 03 1
21 − 21 −
0 0
20 −05 1 20 −05 1
21 − 21 −
−0 01 −0 01
7− 20 7− 20
01 20 01 2
0
20 −03 20 −03
20 − 20 −
0 0
0%
0%
20%
40%
60%
20%
40%
60%
20 −05 1 20 −05 1
20 − 20 −
0 0
r=0.64
r=0.63
r=0.61
r=0.52
20 −07 1 20 −07 1
20 −0 20 −0
20 −09 1 20 −09 1
20 − 20 −
− 0 − 0
20 11 1 20 11 1
21 − 21 −
0 0
20 −0 1 20 −0 1
21 1− 21 1−
− 0 − 0
Porto Velho
Porto Velho
João Pessoa
João Pessoa
20 03 1 20 03 1
21 − 21 −
0 0
20 −05 1 20 −05 1
21 − 21 −
−0 01 −0 01
7− 7−
01 01
in this and the following two weeks per ICU and intermediate beds with ventilator.
0%
0%
0%
0%
20%
40%
60%
20%
40%
60%
20%
40%
60%
20%
40%
60%
Figure S6: (continued) (I) SARI admissions in this and the following two weeks per ventilator. (J) SARI admissions
Imperial College COVID‐19 Response Team
Page 19 of 45
H
20
20 3−wk SARI admissions per critical bed
2
4
6
2
4
6
2
4
6
20 −03
20 −
0
20 −05 1
20 −
0
r=0.51
r=0.33
r=0.48
20 −07 1
20 −0
20 −09 1
20 −
0
20 −11 1
21 −
0
20 −0 1
Macapá
21 1−
0
20 −03 1
Rio de Janeiro
Belo Horizonte
06 October 2021
21 −
− 0
20 05 1
21 −
−0 01
20 7−
20 01
20 −03
20 −
0
20 −05 1
20 −
0
r=0.29
r=0.83
r=0.61
20 −07 1
20 −0
20 −09 1
20 −
0
20 −11 1
21 −
0
20 −0 1
Curitiba
Manaus
21 1−
Salvador
0
20 −03 1
21 −
− 0
20 05 1
21 −
−0 01
20 7−
20 01
20 −03
20 −
0
20 −05 1
20 −
0
r=0.26
r=0.66
r=0.27
20 −07 1
20 −0
20 −09 1
20 −
0
20 −11 1
21 −
DOI: https://doi.org/10.25561/91875
0
20 −0 1
Natal
21 1−
São Luís
0
Florianópolis
20 −03 1
21 −
0
20 −05 1
21 −
−0 01
20 7−
20 01
20 −03
20 −
0
20 −05 1
20 −
0
r=0.52
r=0.67
20 −07 1
r=−0.28
20 −0
−
20 09 1
20 −
0
20 −11 1
21 −
0
20 −0 1
Goiânia
21 1−
0
São Paulo
Porto Alegre
20 −03 1
21 −
0
20 −05 1
21 −
−0 01
7− 20
01 20
20 −03
20 −
0
0%
20%
40%
60%
20 −05 1
20 −
0 r=0.63
r=0.63
20 −07 1
20 −0
−
20 09 1
20 −
0
20 −11 1
21 −
0
20 −0 1
21 1−
0
Porto Velho
João Pessoa
20 −03 1
21 −
0
20 −05 1
21 −
−0 01
7−
01
0%
0%
20%
40%
60%
20%
40%
60%
in this and the following two weeks per ICU and intermediate beds with or without ventilator.
Figure S6: (continued) (I) SARI admissions in this and the following two weeks per ventilator. (J) SARI admissions
Imperial College COVID‐19 Response Team
Page 20 of 45
06 October 2021 Imperial College COVID‐19 Response Team
A 100%
Belo Horizonte Curitiba Florianópolis Goiânia João Pessoa
75%
0−15
50%
25%
0%
100%
75%
16−29
50%
25%
0%
100%
75%
30−39
50%
25%
0%
100%
75%
40−49
50%
25%
0%
having received one or two vaccine doses
100%
75%
50−59
50%
proportions of population
25%
0%
100%
75%
60−69
50%
25%
0%
100%
75%
70−74
50%
25%
0%
100%
75%
75−79
50%
25%
0%
100%
75%
80−84
50%
25%
0%
100%
75%
85−89
50%
25%
0%
100%
75%
90+
50%
25%
0%
20 03 1
1
−0 1
01
20 03 1
1
−0 1
01
20 03 1
1
−0 1
01
20 03 1
1
−0 1
01
20 03 1
1
1
01
−0 1
01
− 0
21 −0
21 −0
21 −0
21 −0
− 0
21 −0
21 −0
21 −0
21 −0
− 0
21 −0
21 −0
21 −0
21 −0
− 0
21 −0
21 −0
21 −0
21 −0
− 0
21 −0
21 −0
21 −0
21 −
7−
21 −
7−
21 −
7−
21 −
7−
21 −
7−
20 −02
20 −04
20 05
20 −06
20 −02
20 −04
20 05
20 −06
20 −02
20 −04
20 05
20 −06
20 −02
20 −04
20 05
20 −06
20 −02
20 −04
20 05
20 −06
−
−
21
21
21
21
21
21
20
20
20
20
20
vaccine Covishield Sinovac Pfizer Janssen doses administred Only one Two
Figure S7: COVID‐19 vaccine coverage. Individual‐level data on administered vaccine doses from the Brazilian
Ministry of Health database [2] were retrieved on 05 August 21, and preprocessed as described in the Supple‐
mentary Text, page 18 and 31. Estimated vaccine coverage is shown by vaccine (colour) and number of doses
administered (colour intensity). The date of Gamma’s detection is added as a grey vertical line. (A) Data for Belo
Horizonte, Curitiba, Florianópolis, Goiânia, João Pessoa.
B 100%
Macapá Manaus Natal Porto Alegre
75%
0−15
50%
25%
0%
100%
75%
16−29
50%
25%
0%
100%
75%
30−39
50%
25%
0%
100%
75%
40−49
50%
25%
0%
having received one or two vaccine doses
100%
75%
50−59
50%
proportions of population
25%
0%
100%
75%
60−69
50%
25%
0%
100%
75%
70−74
50%
25%
0%
100%
75% 75−79
50%
25%
0%
100%
75%
80−84
50%
25%
0%
100%
75%
85−89
50%
25%
0%
100%
75%
90+
50%
25%
0%
−0 1
01
01
01
01
01
−0 1
01
01
01
01
01
−0 1
01
01
01
01
01
−0 1
01
01
01
01
01
0
0
20 02−
3−
4−
5−
6−
7−
20 02−
3−
4−
5−
6−
7−
20 02−
3−
4−
5−
6−
7−
20 02−
3−
4−
5−
6−
7−
−0
−0
−0
−0
−0
−0
−0
−0
−0
−0
−0
−0
−0
−0
−0
−0
−
−
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
vaccine Covishield Sinovac Pfizer Janssen doses administred Only one Two
Figure S7: (continued) (B) Estimated COVID‐19 vaccine coverage in Macapá, Manaus, Natal, Porto Alegre.
C 100%
Porto Velho Rio de Janeiro Salvador São Luís São Paulo
75%
0−15
50%
25%
0%
100%
75%
16−29
50%
25%
0%
100%
75%
30−39
50%
25%
0%
100%
75%
40−49
50%
25%
0%
having received one or two vaccine doses
100%
75%
50−59
50%
proportions of population
25%
0%
100%
75%
60−69
50%
25%
0%
100%
75%
70−74
50%
25%
0%
100%
75%
50% 75−79
25%
0%
100%
75%
80−84
50%
25%
0%
100%
75%
85−89
50%
25%
0%
100%
75%
90+
50%
25%
0%
20 03 1
1
−0 1
01
20 03 1
1
−0 1
01
20 03 1
1
−0 1
01
20 03 1
1
−0 1
01
20 03 1
1
1
01
−0 1
01
− 0
21 −0
21 −0
21 −0
21 −0
− 0
21 −0
21 −0
21 −0
21 −0
− 0
21 −0
21 −0
21 −0
21 −0
− 0
21 −0
21 −0
21 −0
21 −0
− 0
21 −0
21 −0
21 −0
21 −
7−
21 −
7−
21 −
7−
21 −
7−
21 −
7−
20 −02
20 −04
20 05
20 −06
20 −02
20 −04
20 05
20 −06
20 −02
20 −04
20 05
20 −06
20 −02
20 −04
20 05
20 −06
20 −02
20 −04
20 05
20 −06
−
−
21
21
21
21
21
21
20
20
20
20
20
vaccine Covishield Sinovac Pfizer Janssen doses administred Only one Two
Figure S7: (continued) (C) Estimated COVID‐19 vaccine coverage in Porto Velho, Rio De Janeiro, Salvador, São Luís,
São Paulo city.
300 200
40
75
150 100
30
200
50
100 20
50
100
25
50 10
60 60
20 120
200
40 40
80
10 100
20 20
40
1000
500
150
400 100
750
300 100
500
200 50
50
250
100
100 200 300 400 500 50 100 150 200 0 50 100 0 300 600 900 1200
actual COVID−19 hospital admissions
Figure S8: Model fit of expected weekly COVID‐19 attributable hospital admissions. Posterior median estimates
of the expected hospital admissions in residents in location l and age band a in week w obtained with the Bayesian
multi‐strain fatality model, Supplementary Text Equation (S37), are shown on the y‐axis along with 95% credible
intervals against the observed, COVID‐19 attributable hospital admissions among residents in location l and age
band a in week w defined in Supplementary Text, page 12. Locations are shown across facets, and age bands are
shown using colours and plot symbols.
60 30
12
40 40
40 8 20
20 20
20 4 10
0 30 60 90 0 20 40 60 80 0 5 10 15 0 20 40 60 0 10 20 30 40
9
150 30
30 75
6 20
100 20 50
3 50 10 25 10
0 5 10 0 50 100 150 0 10 20 30 40 0 30 60 90 0 10 20 30 40
60
60 300
200
40
40 200
100
20
20 100
Figure S9: Model fit of expected weekly COVID‐19 attributable deaths. Posterior median estimates of the ex‐
pected deaths in location l and age band a following hospital admission of residents in week w obtained with the
Bayesian multi‐strain fatality model, Supplementary Text Equation (S40), are shown on the y‐axis along with 95%
credible intervals against the underreporting‐adjusted COVID‐19 attributable deaths following hospital admission,
Supplementary Text Equation (S11), on the x‐axis. Locations are shown across facets, and age bands are shown
using different colours and plot symbols.
20 20 20 20
0
200
400
600
0
200
400
600
0
100
200
300
400
0
500
1000
20 20 20
20 −0
20 4− 20 −05 20 −05 20 −06
0 20 −0 20 −0 20 −0
20 −0 1
20 6− 20 −07 1 20 −07 1 1
0 20 −0 20 −0 20 −08
20 −0 1 20 −0
20 8− 20 −09 1 1
0 20 −09 1
20 −0 20 −0 20 −10
20 −1 1 20 −0
20 0−
0 20 −11 1 20 −11 1 − 1
20 −12 1 21 −0 21 −0 20 12
21 −0 20 −0 1 20 −0 1 21 −0
21 1− 21 1− 20 −02 1
20 −0 1 0 0
São Luís
21 2− 20 −03 1 20 −03 1 21 −0
0
Porto Alegre
João Pessoa
21 −0 20 −04 1
Belo Horizonte
20 −0 1 21 −0
21 4− 20 −05 1 20 −05 1 21 −0
−0 01 21 −0 21 −0 −0 1
6− −0 1 −0 1 6−
01 7− 7− 01
01 01
20 20 20
non−Gamma
20 20 20 20
0
1000
2000
3000
4000
5000
0
100
200
300
0
25
50
75
100
0
200
400
600
800
20 −0 20 −05 20 −05 20
20 3− 20 −0 20 −0 20 −05
20 −0 01 20 −0
20 5− 20 −07 1 20 −07 1
20 −0 20 −0 20 −07 1
DOI: https://doi.org/10.25561/91875
20 −0 01 20 −0
20 7− 20 −09 1 20 −09 1 20 −09 1
20 −0 20 −0
Gamma
20 −0 01 20 −0
20 9− 20 −11 1 20 −11 1 20 −11 1
20 −1 01 21 −0 21 −0
age
21 1−
Curitiba
Macapá
20 −03 1 20 −03 1
São Paulo
20 −03 1
Porto Velho
20 −0 01 21 −0 21 −0
21 3− 21 −0
20 −0 01 20 −05 1 20 −05 1 20 −05 1
21 5− 21 −0 21 −0 21 −0
−0 01 −0 1 −0 1 −0 1
7− 7− 7− 7−
0−15
01 01 01
70−74
01
20 20
20 20
20
0
500
1000
1500
2000
2500
0
500
1000
0
50
100
150
200
20
20 −05 20 −0
20 4− 20 −05
20 −0 0 20 −0
20 −0 1
75−79
16−29
20 −07 1 20 6− 20 −07 1
20 −0 0 20 −0
20 −0 1
20 −09 1 20 − 8 20 −09 1
0
80−84
30−39
20 −01 1 21 −0 20 −0 1
21 − 21 1−
0 20 −0 1 0
Manaus
20 −03 1 21 2− 20 −03 1
0
Florianópolis
21 −0
Rio de Janeiro
20 −0 1 21 −0
20 −05 1 21 4− 20 −05 1
21 −0 −0 01 21 −0
−0 1 6− −0 1
85−89
40−49
7− 01 7−
01 01
20 20 20
20 20 20
0
250
500
750
0
100
200
300
0
200
400
600
90+
20 −05 20 −05 20 −05
50−59
20 −0 20 −0 20 −0
20 −07 1 20 −07 1 20 −07 1
20 −0 20 −0 20 −0
20 −09 1 20 −09 1 20 −09 1
20 −0 20 −0 20 −0
20 −11 1 20 −11 1 20 −11 1
60−69
21 −0 21 −0 21 −0
20 −01 1 20 −01 1 20 −01 1
Natal
21 − 21 − 21 −
0 0 0
Goiânia
Salvador
Figure S10: Estimated COVID‐19 attributable hospital admissions by SARS‐Cov‐2 variant. Posterior median esti‐
shown for each age band (color) in lighter shades, while those for the Gamma variant are shown in darker shades.
mates of hospital admissions among residents in each location, that are attributed to non‐Gamma variants are
shown across facets. The date of Gamma’s detection is shown as a vertical dotted line. Observed weekly hospital
Estimates are derived using the Bayesian multi‐strain fatality model, Supplementary Text page 41. Locations are
Imperial College COVID‐19 Response Team
Page 26 of 45
estimated in−hospital COVID19 deaths by variant
20 (posterior median)
20 20 20 20
06 October 2021
0
50
100
150
200
250
0
100
200
300
0
40
80
120
160
0
100
200
300
20 20 20
20 −04
20 − 20 −05 20 −05
0 20 −0 20 −0 20 −06
20 −0
20 −06 1
20 − 20 −07 1 20 −07 1 1
0 20 −0 20 −0 20 −08
20 −08 1 20 −0
20 − 20 −09 1 1
0 20 −09 1
20 −0 20 −0 20 −10
20 −10 1 20 −0
20 −
0 20 −11 1 20 −11 1 − 1
20 −12 1 21 −0 21 −0 20 12
21 −0
21 −0 20 −0 1 20 −01 1 1
21 1− 21 − 20 −02
20 −0 1
2 0 0
São Luís
21 − 20 −03 1 21 −0
0 20 −03 1
Porto Alegre
João Pessoa
21 −0
Belo Horizonte
20 −04 1 21 −0 20 −04 1
21 − 20 −05 1 20 −05 1 21 −0
−0 01 21 −0 21 −0 −0 1
6− −0 1 −0 1 6−
01 7− 7− 01
01 01
20 20 20
non−Gamma
20 20 20 20
0
400
800
1200
1600
0
40
80
120
160
0
10
20
30
40
0
100
200
20 −0 20 −05 20 −05 20
20 3− 20 −0 20 −0 20 −05
20 −0 01 1 1 20 −0
20 5− 20 −07 20 −07
20 −0 20 −0 20 −07 1
20 −0 01 1 1 20 −0
DOI: https://doi.org/10.25561/91875
20 7− 20 −09 20 −09 20 −09 1
20 −0 20 −0
Gamma
20 −0 01 20 −0
20 9− 20 −11 1 20 −11 1 20 −11 1
20 −1 01 21 −0 21 −0 21 −0
21 1− 20 −01 1 20 −01 1
20 −0 01 20 −0 1
21 −0 21 −0 21 1−
0
age
21 1−
Curitiba
Macapá
São Paulo
20 −03 1 20 −03 1 20 −03 1
Porto Velho
20 −0 01 21 −0 21 −0
21 3− 21 −0
20 −0 01 20 −05 1 20 −05 1 20 −05 1
21 5− 21 −0 21 −0 21 −0
−0 01 −0 1 −0 1 −0 1
7− 7− 7− 7−
0−15
01 01 01
70−74
01
20 20
20 20
20
0
250
500
750
1000
0
200
400
600
800
0
20
40
60
20
20 −05 20 −04
20 − 20 −05
20 −0 0 20 −0
75−79
16−29
20 −06 1
20 −07 1 20 − 20 −07 1
20 −0 0 20 −0
20 −08 1
20 −09 1 20 − 20 −09 1
0
80−84
30−39
20 −01 1 21 −0 20 −01 1
21 −0 21 −
20 −0 1
2 0
Manaus
20 −03 1 21 − 20 −03 1
0
Florianópolis
21 −0
Rio de Janeiro
20 −04 1 21 −0
20 −05 1 21 − 20 −05 1
21 −0 −0 01 21 −0
−0 1 6− −0 1
85−89
40−49
7− 01 7−
01 01
20 20 20
weekly deaths following hospital admissions in residents are shown as black dots.
20 20 20
0
100
200
300
0
50
100
150
0
100
200
90+
20 −05 20 −05 20 −05
50−59
20 −0 20 −0 20 −0
20 −07 1 20 −07 1 20 −07 1
20 −0 20 −0 20 −0
20 −09 1 20 −09 1 20 −09 1
20 −0 20 −0 20 −0
20 −11 1 20 −11 1 20 −11 1
60−69
21 −0 21 −0 21 −0
20 −01 1 20 −01 1 20 −01 1
Natal
21 −0 21 −0 21 −0
Goiânia
Salvador
mates of deaths following hospital admissions of residents in each location, that are attributed to non‐Gamma
Figure S11: Estimated COVID‐19 attributable deaths in hospitals by SARS‐Cov‐2 variant. Posterior median esti‐
Locations are shown across facets. The date of Gamma’s detection is shown as a vertical dotted line. Observed
variants are shown for each age band (color) in lighter shades, while those for the Gamma variant are shown in
darker shades. Estimates are derived using the Bayesian multi‐strain fatality model, Supplementary Text page 41.
Imperial College COVID‐19 Response Team
Page 27 of 45
06 October 2021 Imperial College COVID‐19 Response Team
50.0%
40.0%
30.0%
weekly in−hospital fatality rate
20.0%
10.0%
0.0%
−1 1
−0 1
−0 1
−0 1
01
70−74 75−79 80−84 85−89 90+
−0
21 −0
21 −0
0−
7−
7
4
−0
20
20
21
50.0%
20
20
20
20
20
40.0%
30.0%
20.0%
10.0%
0.0%
0− 01
01
−0 1
−0 1
01
01
01
−0 1
−0 1
01
−1 1
−0 1
−1 1
−0 1
01
01
−0 1
−0 1
01
21 −0
21 −0
21 −0
21 −0
21 −0
21 −0
21 −0
21 −0
21 −0
21 −0
21 −0
21 −0
7−
20 10−
7−
20 07−
20 10−
7−
20 07−
7−
20 07−
7−
20 07−
20 10−
7−
20 01
20 01
20 01
20 04
20 01
20 04
20 04
0
0−
−
21
20
20
21
20
20
20
20
20
20
21
2
2
20
20
20
20
20
20
20
20
20
20
20
B 100%
0−15 16−29 30−39 40−49 50−59 60−69
75%
50%
weekly in−hospital fatality rate
25%
0%
1
−0 1
01
20 −0
21 −0
21 −0
6−
100%
20 06
20 09
20 12
20 03
−
−
20
20
75%
50%
25%
0%
1
− 01
−0 1
01
−0 1
01
−0 1
01
−0 1
01
−0 1
01
20 −0
20 −0
20 −0
20 −0
21 −0
21 −0
20 −0
20 −0
21 −0
21 −0
20 −0
20 −0
21 −0
21 −0
20 −0
20 −0
21 −0
21 −0
−
−
6−
6−
6−
6−
6−
20 06
20 09
20 12
20 03
20 06
20 09
20 12
20 03
20 06
20 09
20 12
20 03
20 06
20 09
20 12
20 03
20 06
20 09
20 12
20 03
−
−
20
21
21
20
20
20
20
20
20
20
20
20
Figure S12: Model fits to age‐specific COVID‐19 in‐hospital fatality rates. Weekly, age‐specific in‐hospital fatality
rates are shown as dots. Posterior median estimates of the expected in‐hospital fatality rates across variants from
the Bayesian multi‐strain fatality model, Supplementary Text Equation (S34), are shown on the y‐axis (black line)
along with 95% credible intervals (grey ribbon). The expected in‐hospital fatality rates of non‐Gamma variants,
Supplementary Text Equation (S33a), are shown as dotted line. The date of Gamma’s first detection is indicated
as a vertical dotted black line. (A) For Belo Horizonte. (B) For Curitiba.
D
weekly in−hospital fatality rate weekly in−hospital fatality rate weekly in−hospital fatality rate
20 20 20
2
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
20 0− 20−
20
−0 20 0
6− 20 0
6− 2 6−
20 01 0− 01 20− 01
−0 20 09
20 9− 20 09
20 01 20 −0 20 −0
−1 −1 1 −1 1
20 2− 20 2 20 2
21 21 −0 21 −0
01 − 1 − 1
0−15
0−15
0−15
−0
70−74
70−74
70−74
20 3− 20 03 20 03
06 October 2021
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 01
20 20 20
20 20 20
−0 − −
20 6− 20 06 20 06
20 20 −0 20 −0
−0
01 − 1 − 1
20 9− 20 09 20 09
20 20 −0 20 −0
−1
01 − 1 − 1
20 2− 20 12 20 12
21 21 −0 21 −0
−0
01 − 1 − 1
75−79
16−29
75−79
16−29
75−79
16−29
20 20 03 20 03
age
age
age
3−
0−15
0−15
0−15
20 20 06 20 06
70−74
70−74
70−74
6− 20 −0
20 01 1 20 −0
−0 − − 1
20 9− 20 09 20 09
20 20 −0 20 −0
−1
01 − 1 − 1
20 2− 20 12 20 12
21 21 −0 21 −0
−0
01 − 1 − 1
75−79
16−29
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
80−84
30−39
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 01
20 20 20
DOI: https://doi.org/10.25561/91875
20 20 20
−0 − −
80−84
30−39
80−84
30−39
80−84
30−39
20 6− 20 06 20 06
20 20 −0 20 −0
−0
01 − 1 − 1
20 9− 20 09 20 09
20 20 −0 20 −0
−1
01 − 1 − 1
20 2− 20 12 20 12
21 21 −0 21 −0
01 1 1
85−89
40−49
85−89
40−49
85−89
40−49
−0 − −
85−89
40−49
85−89
40−49
85−89
40−49
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 01
20
90+
90+
90+
20 20
20 20 20
50−59
50−59
50−59
−0 − −
20 6− 20 06 20 06
20 20 −0 20 −0
−0
01 − 1 − 1
20 9− 20 09 20 09
20 20 −0 20 −0
−1
01 − 1 − 1
20 2− 20 12 20 12
60−69
60−69
60−69
21 21 −0 21 −0
01 − 1 − 1
90+
90+
90+
−0
50−59
50−59
50−59
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 01
20 20 20
20 20 20
−0 − −0
20 6− 20 06
20 −0
20 6
−0
20 01 1 20
−0 − −0 1
20 9− 20 09 20 9−
20 01 20 −0
− 1
20 0
−1 20 12 −1 1
20 2− 20 2−
21 21 −0 21
−0
01 − 1 −
01
60−69
60−69
60−69
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 01
Figure S12: (continued) Model fits to age‐specific COVID‐19 in‐hospital fatality rates. (C) For Florianópolis. (D) For
Imperial College COVID‐19 Response Team
Page 29 of 45
F
H
G
weekly in−hospital fatality rate weekly in−hospital fatality rate weekly in−hospital fatality rate
20 20 20
2 2
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
20 0− 0−
−0 20 05 20 06
20 6− 20 −0 20 −0
20
−0
01 − 1 − 1
20 20 08 20 09
9− 20 −0
20 01 − 1 20 −0
− 1
−1 20 11 20 12
20 2− 21 −0
21 01 1 21 −0
1
20 −02 −
0−15
0−15
0−15
−0
70−74
70−74
70−74
20 20 03
3− 21 −0 −
06 October 2021
21 01 −0 1 21 0
−0 5− −0 1
6− 01 6−
01 20 20 01
20 20 20
20 − −
−0 20 05 20 06
20 6− 20 −0 20 −0
20 − 1 1
75−79
16−29
75−79
16−29
75−79
16−29
20 21 −0 20 03
age
age
age
3−
21 01 −0 1 21 −0
−0 5− −0 1
6− 01 6−
01 20 20 01
20 20 20
20 − −
−0
0−15
0−15
0−15
20 20 05 20 06
70−74
70−74
70−74
6− 20 −0 20 −0
20
−0
01 − 1 − 1
20 20 08 20 09
9− 20 −0
20 01 − 1 20 −0
− 1
−1 20 11 20 12
20 2− 21 −0
21 01 1 21 −0
1
−0 20 −02 −
75−79
16−29
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
80−84
30−39
20 3− 21 −0 20 03
21 01 −0 1 21 −0
−0 5− −0 1
6− 01 6−
01 20 20 01
20
DOI: https://doi.org/10.25561/91875
20 20 20
−0 − −
80−84
30−39
80−84
30−39
80−84
30−39
20 20 05 20 06
6− 20 −0 20 −0
20
−0
01 − 1 − 1
20 20 08 20 09
9− 20 −0
20 01 − 1 20 −0
− 1
−1 20 11 20 12
20 2− 21 −0
21 01 1 21 −0
1
−
85−89
40−49
85−89
40−49
85−89
40−49
−0 20 −02
85−89
40−49
85−89
40−49
85−89
40−49
20 3− 21 −0 20 03
21 01 −0 1 21 −0
−0 5− −0 1
6− 01 6−
01 20 20 0 1
20
90+
90+
90+
20 20 20
− −
50−59
50−59
50−59
−0 20 05 20 06
20 6− 20 − 20 −0
20 01 − 01
−0 20 08 −0 1
20 9− 20 −0 20 9
20 01 − 1 20 −0
− 1
−1 20 11 20 12
20 2− 21 −0
1
60−69
60−69
60−69
21 21 −0
01 − 1
90+
90+
90+
−0 20 −02
50−59
50−59
50−59
20 3− 21 −0 20 03
21 01 −0 1 21 −0
−0 5− −0 1
6− 01 6−
01 20 01 20
20 20 20
20 − −0
−0
20 6−
20 05 20 6−0
20 −0 20
20
−0
01 − 1 −0 1
20 20 08 20 9−
9− 20 −0
20 01 − 1 20 0
−1 20 11 −1 1
20 2− 21 −0 20 2
21 01 1 21 −0
1
−0 20 −02 −
60−69
60−69
60−69
20 3− 21 −0 20 03
21 01 −0 1 21 −0
−0 5− −0 1
6− 01 6−
01 01
Figure S12: (continued) Model fits to age‐specific COVID‐19 in‐hospital fatality rates. (F) For Macapá. (G) For
Imperial College COVID‐19 Response Team
Page 30 of 45
I
K
weekly in−hospital fatality rate weekly in−hospital fatality rate weekly in−hospital fatality rate
20 20
2 20
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
20 0− 20−
−0
20 6− 20 06
20 −0 20 0
6−
20 01 − 1 20− 01
−0 20 09
20 9− 20 09
20 20 −0
−1
01 − 1 20 −0
20 20 12 −1 1
2− 20
21 21 −0
01 − 1 21 2−0
− 1
0−15
0−15
0−15
−0
70−74
70−74
70−74
20 20 03 20 03
3− −
06 October 2021
21 01 21 0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 20 01 01
20 20 20
20 − 20
20
−0 20 06 −
6− 20 −0 20 06
20 01 − 1 20 −0
−0 20 09 − 1
20 9− 20 09
20 20 −0
01 − 1 20 −0
1
20
−1 20 12 −
2− 20 12
21 21 −0
−0
01 − 1 21 −0
− 1
75−79
16−29
75−79
16−29
75−79
16−29
20 20 03 20 03
age
age
age
3−
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 20 01 01
20 20 20
20 − 20
−0 −
0−15
0−15
0−15
20 20 06
70−74
70−74
70−74
6− 20 −0 20 06
20 01 − 1 20 −0
− 1
75−79
16−29
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
80−84
30−39
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 20 01 01
20 20
DOI: https://doi.org/10.25561/91875
20 20
− 20
−0 −
80−84
30−39
80−84
30−39
80−84
30−39
20 6− 20 06 20 06
20 −0
20 01 − 1 20 −0
−0 20 09 − 1
20 9− 20 09
20 −0
20 01 − 1 20 −0
1
20
−1 20 12 −
2− 20 12
21 21 −0
01 − 1 21 −0
85−89
40−49
85−89
40−49
85−89
40−49
−0 − 1
85−89
40−49
85−89
40−49
85−89
40−49
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 20 0 1 01
20
90+
90+
90+
20 20
20 − 20
50−59
50−59
50−59
20
−0 20 06 −
6− 20 −0 20 06
20 01 20 −0
−0 20
−0 1
9 − 1
20 9− 20 09
20 20 −0
01 − 1 20 −0
1
20
−1 20 12 −
2− 20 12
60−69
60−69
60−69
21 21 −0
01 − 1 21 −0
− 1
90+
90+
90+
−0
50−59
50−59
50−59
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 20 01
20 20 20
20 − 20
−0 20 06 −0
20 6− 20 6
20 −0
20
−0
01 − 1 20 −0
20 20 09 −0 1
9− 20 −0 20 9−
20 01 1 20 0
−1 − −1 1
20 2− 20 12 20
21 21 −0
−0
01 − 1 21 2−0
− 1
60−69
60−69
60−69
20 3− 20 03 20 03
21 01 21 −0 21 −0
−0 −0 1 −0 1
6− 6− 6−
01 01 01
Figure S12: (continued) Model fits to age‐specific COVID‐19 in‐hospital fatality rates. (I) For Porto Alegre. (J) For
Imperial College COVID‐19 Response Team
Page 31 of 45
L
N
M
weekly in−hospital fatality rate weekly in−hospital fatality rate weekly in−hospital fatality rate
20 20 20
2 2
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
20 0− 0−
20 −04
20 −0
20 05 20 0
6−
20 −0 2
20 −07 1 − 1 0− 01
20 08 20 09
20 −0 20 −0
20 −10 1 − 1 20 −0
20 11 −1 1
21 −0 21 −0 20 2
20 −01 1 1 21 −0
1
20 −02 −
0−15
0−15
0−15
70−74
70−74
70−74
21 −0 20 03
21 −0
06 October 2021
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
20 7− 20 01
20 01 20 20
− 20
20 −04 20 05 −
20 −0 20 −0 20 06
− − 1 20 −0
1
20 07 1 20 08 −
20 −0 20 −0 20 09
− 1 20 −0
1
20 −10 1
21 −0 20 11 −
21 −0 20 12
1 21 −0
75−79
16−29
75−79
16−29
75−79
16−29
21 −0 20 03
21 −0
age
age
age
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
20 7− 20 01
20 01 20 20
− 20
20 −04 −
0−15
0−15
0−15
20 05 20 06
70−74
70−74
70−74
20 −0 20 −0
− − 1 20 −0
1
20 07 1 20 08 −
20 −0 20 −0 20 09
− 1 20 −0
1
20 −10 1
21 −0 20 11 −
21 −0 20 12
20 −01 1 1 21 −0
1
20 −02 −
75−79
16−29
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
80−84
30−39
21 −0 20 03
21 −0
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
20 7− 20 01
01 20
DOI: https://doi.org/10.25561/91875
20 20 20
20 −04 − −
80−84
30−39
80−84
30−39
80−84
30−39
20 05 20 06
20 −0 20 −0
− − 1 20 −0
1
20 07 1 20 08 −
20 −0 20 −0 20 09
− 1 20 −0
1
20 −10 1
21 −0 20 11 −
21 −0 20 12
20 −01 1 1 21 −0
1
85−89
40−49
85−89
40−49
85−89
40−49
20 −02 −
85−89
40−49
85−89
40−49
85−89
40−49
21 −0 20 03
21 −0
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
20 7− 20 01
20
90+
90+
90+
20 01 20
− 20
50−59
50−59
50−59
20 −04 20 05 −
20 −0 20 − 20 06
− − 01 20 −0
1
20 07 1 20 08 −
20 −0 20 −0 20 09
− 1 20 −0
1
20 −10 1
21 −0 20 11 −
21 −0 20 12
1
60−69
60−69
60−69
21 −0
20 −01 1
− 1
90+
90+
90+
20 −02
50−59
50−59
50−59
21 −0 20 03
21 −0
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
7− 20 20 01
01 20 20 20
− 20
20 −04 20 05 −0
20 −0 20 −0 20 6
−0
20 −07 1 − 1 20
−0 1
20 08 20
20 −0 20 −0 9−
20 −10 1 − 1 20 0
21 −0 20 11 −1 1
21 −0 20 2−
20 −01 1 1 21 01
20 −02 −
60−69
60−69
60−69
21 −0 20 03
21 −0
20 −04 1 −0 1 21 −0
21 −0 5− −0 1
−0 1 01 6−
7− 01
01
Figure S12: (continued) Model fits to age‐specific COVID‐19 in‐hospital fatality rates. (L) For Salvador. (M) For São
Imperial College COVID‐19 Response Team
Page 32 of 45
B
A
share of age groups share of age groups
among COVID−19 attributable hospital admissions among COVID−19 attributable hospital admissions
20
20 2
10%
20%
10%
20%
2 20 0−0
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
20 0− 2 6
2 0 20 0−0 −0
20 0− 5−0 2 8 1
2 0 1 20 0−1 −0
20 0− 7−0 2 0 1
2 0 1 20 0−1 −0
20 0− 9−0 2 2 1
2 1 1 20 1− −0
20 1− 1−0
2 0 1 2 02 1
0−15
06 October 2021
70−74
20 1− 1−0 20 1−0 −0
0−15
70−74
2 0 1 21 4− 1
20 1− 3−0 −0 01
21 05 1 6−
−0 −0 20 01
20 7− 1 2
2 01 20 0−0
20 0− 2 6
20 05 20 0−0 −0
20 − −0 2 8 1
2 0 1 20 0−1 −0
20 0− 7−0 2 0 1
2 0 1 20 0−1 −0
20 0− 9−0
2 1 1 2 2 1
20 1− 1−0 20 1− −0
2 0 1 2 02 1
75−79
16−29
20 1− 1−0
age
20 1−0 −0
75−79
16−29
2 0 1
age
21 4− 1
20 1− 3−0 −0 01
21 05 1 6−
−0 −0 20 01
20 7− 1 2
2 01 20 0−0
20 0− 2 6
0−15
2 0
0−15
70−74
70−74
20 0− 5−0 20 0−0 −0
20 07 1 2 8 1
20 − −0 20 0−1 −0
2 0 1 2 0 1
20 0− 9−0 20 0−1 −0
2 1 1 2 2 1
20 1− 1−0 20 1− −0
2 0 1 2 02 1
75−79
16−29
80−84
30−39
20 1− 1−0
75−79
16−29
20 1−0 −0
80−84
30−39
2 0 1 21 4− 1
20 1− 3−0 −0 0
21 05 1 6− 1
−0 −0 01
20 7− 1 20
DOI: https://doi.org/10.25561/91875
2 01 20
20 0− 20 −0
2 0
80−84
30−39
20 6−
80−84
30−39
20 0− 5−0 20 −0 0
2 0 1 2 8 1
20 0− 7−0 20 0−1 −0
2 0 1 2 0 1
20 0− 9−0 20 0−1 −0
2 1 1 2 2 1
20 1− 1−0 20 1− −0
2 0 1 2 02 1
85−89
40−49
85−89
40−49
20 1− 1−0
85−89
40−49
20 1−0 −0
85−89
40−49
2 0 1 21 4− 1
20 1− 3−0
21 05 1 −0 01
−0 −0 6−
20 7− 1 20 01
2
90+
2 01
90+
20 0− 20 0−0
50−59
50−59
20 05 20 6−
20 − −0 20 −0 0
2 0 1 20 8− 1
20 0− 7−0 20 −1 0
2 0 1 2 0 1
20 0− 9−0 20 0−1 −0
2 1 1 2 2 1
20 1− 1−0 20 1− −0
60−69
21 01 1 2 02 1
60−69
90+
90+
20 − −0
50−59
50−59
2 0 1 20 1−0 −0
21 4− 1
20 1− 3−0
21 05 1 −0 01
−0 −0 6−
7− 1 20 01 20
01 2 2
20 0− 20 0−0
2 0 2 6
20 0− 5−0 20 0−0 −0
2 0 1 2 8 1
20 0− 7−0 20 0−1 −0
2 0 1 2 0 1
20 0− 9−0
2 1 1 20 0−1 −0
2 2 1
20 1− 1−0
2 0 1 20 1− −0
2 02 1
20 1− 1−0
60−69
60−69
2 0 1 20 1−0 −0
20 1− 3−0 21 4− 1
21 05 1 −0 0
−0 −0 6− 1
7− 1 01
01
(x‐axis). The empirical proportions are shown as dots. (A) For Belo Horizonte. (B) For Curitiba.
y‐axis as a black line along with 95% credible intervals. Time trends are shown by week of hospital admission
obtained with the Bayesian multi‐strain fatality model, Supplementary Text Equation (S31b), are shown on the
Figure S13: Model fits of the expected age composition of COVID‐19 attributable hospital admissions. Posterior
median estimates of the expected age composition of hospital admissions in each location, age band, and week,
Imperial College COVID‐19 Response Team
Page 33 of 45
E
C
D
share of age groups share of age groups share of age groups
among COVID−19 attributable hospital admissions among COVID−19 attributable hospital admissions among COVID−19 attributable hospital admissions
20 20
20 2 2
0%
10%
20%
30%
0%
10%
20%
30%
0%
20%
40%
60%
0%
20%
40%
60%
2 20 0− 20 0−
2 05 2 0
0.0%
10.0%
20.0%
30.0%
40.0%
0.0%
10.0%
20.0%
30.0%
40.0%
20 0−
2 0 20 0− −0 20 0− 5−0
20 0− 5−0 2 07 1 2 0 1
2 0 1 20 0− −0 20 0− 7−0
20 0− 7−0 2 09 1 2 0 1
2 0 1 20 0− −0 20 0− 9−0
20 0− 9−0 2 1 1 2 1 1
2 1 1 20 1− 1−0 20 1− 1−0
20 1− 1−0 2 0 1 2 0 1
2 0 1 20 1− 1−0 20 1− 1−0
0−15
0−15
06 October 2021
2 0 1
70−74
70−74
20 1− 1−0 2 03 1
0−15
2 0 1
70−74
20 1− −0 20 1− 3−0
20 1− 3−0 21 05 1 21 05 1
21 05 1 − −0 − −0
− −0 20 07− 1 20 07− 1
20 07− 1 2 01 2 01
2 01 20 0− 20 0−
20 0− 20 05 2 0
20 05 20 − −0 20 0− 5−0
20 − −0 2 07 1 20 07 1
2 0 1 20 0− −0 20 − 0 −
20 0− 7−0 2 09 1 2 0 1
2 0 1 20 0− −0 20 0− 9−0
20 0− 9−0 2 1 1 2 1 1
2 1 1 20 1− 1−0 20 1− 1−0
20 1− 1−0 2 0 1 2 0 1
2 0 1 20 1− 1−0 20 1− 1−0
2 0 1
75−79
16−29
75−79
16−29
20 1− 1−0 2 03 1
age
age
2 0 1
75−79
16−29
age
20 1− −0 20 1− 3−0
20 1− 3−0 21 05 1 21 05 1
21 05 1 − −0 − −0
− −0 20 07− 1 20 07− 1
20 07− 1 2 01 2 01
2 01 20 0− 20 0−
20 0− 20 05 20 05
0−15
0−15
2 0
0−15
20 − −0
70−74
70−74
20 − −0
70−74
20 0− 5−0 2 07 1 2 0 1
20 07 1 20 0− −0 20 0− 7−0
20 − −0
2 0 1
2 09 1 2 0 1
20 0− −0 20 0− 9−0
20 0− 9−0 2 1 1 2 1 1
2 1 1 20 1− 1−0 20 1− 1−0
20 1− 1−0 2 0 1 2 0 1
2 0 1 20 1− 1−0 20 1− 1−0
75−79
16−29
75−79
16−29
2 0 1
80−84
30−39
80−84
30−39
20 1− 1−0 2 03 1
75−79
16−29
2 0 1
80−84
30−39
20 1− 3−0 20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
− −0 − −0 − −0
20 07− 1 20 07− 1
20 07− 1 2 2
DOI: https://doi.org/10.25561/91875
2 01 01 01
20 0− 20 0−
20 0− 20 05 20 05
80−84
30−39
80−84
30−39
20 05 20 − −0
80−84
30−39
20 − −0 20 − −0
2 0 1 2 07 1 2 0 1
20 0− −0 20 0− 7−0
20 0− 7−0 0
2 9 1 2 0 1
2 0 1 20 0− −0 20 0− 9−0
20 0− 9−0
2 1 1 2 1 1 2 1 1
20 1− 1−0 20 1− 1−0 20 1− 1−0
2 0 1 2 0 1 2 0 1
85−89
40−49
85−89
40−49
20 1− 1−0 20 1− 1−0
85−89
40−49
20 1− 1−0 2 0 1
85−89
40−49
85−89
40−49
2 0 1 2 03 1
85−89
40−49
20 1− 3−0 20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
− −0 − −0 − −0
20 07− 1 20 07− 1
20 07− 1
2 2 01 2 01
90+
90+
01
90+
20 0− 20 0− 20 0−
2 05 2 0
50−59
50−59
50−59
20 05 20 0− −0
(C) For Florianópolis. (D) For Goiânia. (E) For João Pessoa.
20 − −0 20 0− 5−0
2 0 1 20 07 1
20 − −0
20 07 1
20 − −0
20 0− 7−0
2 0 1 2 09 1 2 0 1
20 0− 9−0 20 0− −0 20 0− 9−0
2 1 1 2 1 1 2 1 1
20 1− 1−0 20 1− 1−0 20 1− 1−0
2 0 1 2 0 1 2 0 1
60−69
60−69
60−69
90+
90+
20 1− 1−0 20 1− 1−0
90+
20 1− 1−0
2 0 1
50−59
50−59
2 0 1 2 03 1
50−59
20 1− 3−0 20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
−0 −0 −0 −0 −0 −0
7− 1 20 7− 1 20 7− 1 20
01 2 01 2 01 2
20 0− 20 0− 20 0−
2 0 2 05 2 0
20 0− 5−0 20 0− −0 20 0− 5−0
2 0 1 2 07 1 2 0 1
20 0− 7−0 20 0− −0 20 0− 7−0
2 0 1 2 09 1 2 0 1
20 0− 9−0 20 0− −0 20 0− 9−0
2 1 1 2 1 1 2 1 1
20 1− 1−0 20 1− 1−0 20 1− 1−0
2 0 1 2 0 1 2 0 1
20 1− 1−0 20 1− 1−0 20 1− 1−0
2 0 1
60−69
60−69
2 0 1
60−69
2 03 1
20 1− 3−0 20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
−0 −0 −0 −0 −0 −0
7− 1 7− 1 7− 1
01 01 01
Figure S13: (continued) Model fits of the expected age composition of COVID‐19 attributable hospital admissions.
Imperial College COVID‐19 Response Team
Page 34 of 45
F
H
G
share of age groups share of age groups share of age groups
among COVID−19 attributable hospital admissions 20 among COVID−19 attributable hospital admissions among COVID−19 attributable hospital admissions
20 2 20
2 20 0− 2
0%
20%
40%
60%
80%
0%
20%
40%
60%
80%
2 0 20 0−
0.0%
10.0%
20.0%
30.0%
40.0%
0.0%
10.0%
20.0%
30.0%
40.0%
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
20 0−
2 0
20 0− 4−0 2 05
20 0− 5−0 2 0 1 20 0−0 −0
20 0− 6−0 2 7 1
2 0 1 2 0 1 20 0− −0
20 0− 7−0 20 0− 8−0
2 0 1 2 1 1 2 09 1
20 0− 9−0 20 0− 0−0 20 0− −0
2 1 1 2 1 1 2 11 1
20 1− 1−0 20 1− 2−0 20 1− −0
2 0 1 2 0 1 2 01 1
0−15
06 October 2021
70−74
20 1− 1−0 20 1− 2−0 20 1− −0
0−15
0−15
2 0 1 2 03 1
70−74
70−74
21 04 1
20 1− 3−0 −0 −0 20 1− −0
21 05 1 21 05 1
− −0 20 6− 1 −
20 07 1 2 01 20 −07 01
2 −0 20 0− 2 −
20 0− 1 20 04 20 0− 01
20 05 20 − −0 2 05
20 − −0 20 06 1 20 0−0 −0
20 − −0 20 7− 1
20 07 1
20 − −0 2 0 1 20 − 0
2 0 1 20 0− 8−0 2 09 1
20 0− 9−0 2 1 1 20 0− −0
2 1 1 20 0− 0−0 2 11 1
20 1− 1−0 2 1 1 20 1− −0
20 1− 2−0
2 0 1 2 0 1 2 01 1
75−79
16−29
20 1− 1−0 20 1− −0
age
20 1− 2−0
2 0 1 2 03 1
75−79
16−29
75−79
16−29
age
age
20 1− 3−0 21 04 1 20 1− −0
21 05 1 −0 −0 21 05 1
− −0 20 6− 1 −
20 07 1 2 01 20 −07 01
2 −
2 −0 20 0−
20 0− 1 2 0 20 0− 01
2 0 2 05
0−15
20 0− 4−0
0−15
0−15
70−74
20 06 1 20 0−0 −0
70−74
70−74
20 0− 5−0 20 − −0
20 07 1 20 7− 1
20 − −0 2 0 1 20 − 0
20 09 1 20 0− 8−0 2 09 1
20 − −0 2 1 1 20 0− −0
2 1 1 20 0− 0−0 2 11 1
20 1− 1−0 2 1 1 20 1− −0
2 0 1 20 1− 2−0
2 0 1 2 01 1
75−79
16−29
20 1− −0
80−84
30−39
20 1− 1−0
75−79
16−29
75−79
16−29
2 0 1 20 1− 2−0 2 03 1
80−84
30−39
80−84
30−39
20 1− 3−0 21 04 1 20 1− −0
21 05 1 −0 −0 21 05 1
− −0 20 6− 1 −
20 07 1 2 01 20 −07 01
2 2 −
DOI: https://doi.org/10.25561/91875
−0 20 0−
20 0− 1 2 0 20 0− 01
20 05
80−84
30−39
20 05 20 0− 4−0
80−84
30−39
80−84
30−39
20 − −0 20 06 1 20 −0 −0
2 7 1
85−89
40−49
2 0 1
85−89
40−49
85−89
40−49
20 1− 1−0 20 1− −0
85−89
40−49
2 0 1 2 03 1
85−89
40−49
85−89
40−49
20 1− 2−0
20 1− 3−0 21 04 1 20 1− −0
21 05 1 −0 −0 21 05 1
− −0 20 6− 1 −
20 −07 01
20 07 1 2 01
2 −
2 −0
90+
1 20 0−
90+
90+
20 0− 20 04 20 0− 01
2 05
50−59
50−59
50−59
20 05 20 − −0
20 − −0 2 0 1 20 0−0 −0
20 7− 1
20 07 1 20 0− 6−0
20 − −0
0 2 0 1 20 − 0
2 1 20 0− 8−0 2 09 1
20 0− 9−0 2 1 1 20 0− −0
2 1 1 20 0− 0−0 2 11 1
20 1− 1−0 2 1 1 20 1− −0
2 0 1 2 01 1
60−69
20 1− 2−0
60−69
60−69
2 0 1
90+
90+
90+
20 1− 1−0 20 1− −0
0
50−59
2 1 2 03 1
50−59
50−59
20 1− 2−0
20 1− 3−0 21 04 1 20 1− −0
21 05 1 −0 −0 21 05 1
−0 −0 6− 1 20 −0 −0 20
7− 1 20 01 2 7− 1 2
01 2 20 0− 01 20 0−
20 0−
2 0 2 0 2 05
20 0− 4−0 20 0−0 −0
20 0− 5−0
2 0 1 2 0 1 2 7 1
20 0− 6−0 20 0− −0
20 0− 7−0 2 0 1 2 09 1
2 0 1 20 0− 8−0
20 0− 9−0 2 1 1 20 0− −0
2 11 1
2 1 1 20 0− 0−0
20 1− 1−0 2 1 1 20 1− −0
2 0 1 20 1− 2−0 2 01 1
20 1− 1−0 2 0 1 20 1− −0
2 0 1
60−69
2 03 1
60−69
60−69
20 1− 2−0
20 1− 3−0 21 04 1 20 1− −0
21 05 1 −0 −0 21 05 1
−0 −0 6− 1 −0 −0
7− 1 01 7− 1
01 01
Figure S13: (continued) Model fits of the expected age composition of COVID‐19 attributable hospital admissions.
Imperial College COVID‐19 Response Team
Page 35 of 45
I
K
share of age groups share of age groups share of age groups
among COVID−19 attributable hospital admissions among COVID−19 attributable hospital admissions among COVID−19 attributable hospital admissions
20 20
2 20
2
2
10%
20%
10%
20%
0%
20%
40%
60%
0%
20%
40%
60%
20 0− 20 0−
2 05 20 0−
0.0%
10.0%
20.0%
30.0%
40.0%
0.0%
10.0%
20.0%
30.0%
40.0%
20 0− −0 2 05 2 0
20 0− 5−0
2 07 1 20 0−0 −0 2 0 1
20 0− −0 2 7 1 20 0− 7−0
2 09 1 20 0− −0 2 0 1
20 0− −0 2 09 1 20 0− 9−0
2 1 1 20 0− −0 2 1 1
20 1− 1−0 2 11 1 20 1− 1−0
2 0 1 20 1− −0 2 0 1
20 1− 1−0 2 01 1 20 1− 1−0
0−15
0−15
06 October 2021
70−74
70−74
2 03 1 20 1− −0 2 0 1
0−15
2 03 1
70−74
20 1− −0 20 1− 3−0
21 05 1 20 1− −0 21 05 1
− −0 21 05 1
− −0 −0
20 07− 1 20 −07 01 20 7− 1
2 01 2 − 2 01
20 0− 20 0− 01 20 0−
20 05
20 − −0 2 05 2 0
20 0− 5−0
2 07 1 20 0−0 −0
20 7− 1 20 07 1
20 0− −0 20 − 0 20 − −0
2 09 1 2 09 1 2 0 1
20 0− −0 20 0− 9−0
2 1 1 20 0− −0
2 11 1 2 1 1
20 1− 1−0 20 1− 1−0
2 0 1 20 1− −0
2 01 1 2 0 1
20 1− 1−0 20 1− 1−0
75−79
16−29
75−79
16−29
2 03 1 20 1− −0 2 0 1
age
age
2 03 1
75−79
16−29
age
20 1− −0 20 1− 3−0
21 05 1 20 1− −0 21 05 1
− −0 21 05 1
20 07− 1 − −0 −0
20 −07 01 20 7− 1
2 01 2 − 2 01
20 0− 20 0− 01 20 0−
20 05 2 05
0−15
0−15
20 05
0−15
20 − −0
70−74
70−74
20 0−0 −0 20 − −0
70−74
2 07 1 20 7− 1 2 0 1
20 0− −0 20 − 0 20 0− 7−0
2 09 1 2 09 1 2 0 1
20 0− −0 20 0− −0 20 0− 9−0
2 1 1 2 11 1 2 1 1
20 1− 1−0 20 1− −0 20 1− 1−0
2 0 1 2 01 1 2 0 1
20 1− 1−0 20 1− 1−0
75−79
16−29
75−79
16−29
20 1− −0
80−84
30−39
80−84
30−39
2 03 1 2 0 1
75−79
16−29
2 03 1
80−84
30−39
20 1− −0 20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
− −0 − −0 −0
20 07− 1 20 −07 01 20 7− 1
2 2 − 2
DOI: https://doi.org/10.25561/91875
01 01
20 0− 20 0− 01 20 0−
20 05 20 05 20 05
80−84
30−39
80−84
30−39
20 − −0
80−84
30−39
20 −0 −0 20 − −0
2 07 1 2 7 1 2 0 1
20 0− −0
0
2 9 1
20 0− −0
2 09 1
20 0− 7−0
2 0 1
20 0− −0 20 0− −0 20 0− 9−0
2 1 1 2 11 1 2 1 1
20 1− 1−0 20 1− −0 20 1− 1−0
2 0 1 2 01 1 2 0 1
85−89
40−49
85−89
40−49
20 1− 1−0 20 1− 1−0
85−89
40−49
20 1− −0
85−89
40−49
85−89
40−49
2 03 1 2 03 1 2 0 1
85−89
40−49
20 1− −0 20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
− −0 − −0 −0
20 07− 1 20 −07 01 20 7− 1
2 01 2 − 2 01
90+
90+
90+
20 0− 20 0− 01 20 0−
2 05 2 05
50−59
50−59
50−59
20 0− −0 20 0−0 −0 20 05
20 − −0
20 07 1 20 7− 1 20 07 1
20 − −0 20 − 0 20 − −0
2 09 1 2 09 1 2 0 1
20 0− −0 20 0− −0 20 0− 9−0
2 1 1 2 11 1 2 1 1
20 1− 1−0 20 1− −0 20 1− 1−0
2 0 1 2 01 1 2 0 1
(I) For Porto Alegre. (J) For Porto Velho. (K) For Rio De Janeiro.
60−69
60−69
60−69
90+
90+
20 1− 1−0
90+
20 1− −0 20 1− 1−0
50−59
50−59
2 03 1 2 03 1 2 0 1
50−59
20 1− −0 20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
−0 −0 −0 −0 −0 −0
7− 1 20 20 7− 1
01 2
7− 1
01 2 01
20
2
20 0− 20 0− 20 0−
2 05 2 05 2 0
20 0− −0 20 0−0 −0 20 0− 5−0
2 07 1 2 7 1 2 0 1
20 0− −0 20 0− −0 20 0− 7−0
2 09 1 2 09 1 2 0 1
20 0− −0 20 0− −0 20 0− 9−0
2 1 1 2 11 1 2 1 1
20 1− 1−0 20 1− −0 20 1− 1−0
2 0 1 2 01 1 2 0 1
20 1− 1−0 20 1− −0 20 1− 1−0
60−69
60−69
2 03 1 2 0 1
60−69
2 03 1
20 1− −0 20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
−0 −0 −0 −0 −0 −0
7− 1 7− 1 7− 1
01 01 01
Figure S13: (continued) Model fits of the expected age composition of COVID‐19 attributable hospital admissions.
Imperial College COVID‐19 Response Team
Page 36 of 45
L
N
M
share of age groups share of age groups share of age groups
among COVID−19 attributable hospital admissions 20 among COVID−19 attributable hospital admissions among COVID−19 attributable hospital admissions
20
20 2 2
20 0−
0%
20%
40%
60%
0%
20%
40%
60%
10%
20%
30%
10%
20%
30%
2020− 2 0 20 0−
2 05
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
0 20 0− 4−0 20 0− −0
2020− 3− 2 0 1 2 07 1
2 0 0 20 0− 6−0
20 0− 5− 1
0 0 2 0 1 20 0− −0
2 09 1
20 0− 8−0
2020− 7− 1
0 0 2 1 1 20 0− −0
2020− 9− 1 20 0− 0−0 2 1 1
1 0 2 1 1 20 1− 1−0
2 0 1
2021− 1− 1 20 1− 2−0
0 0 2 0 1 20 1− 1−0
0−15
0−15
06 October 2021
70−74
70−74
20 1− 2−0 2 03 1
0−15
2021− 1− 1
70−74
2 0 0 21 04 1 20 1− −0
20 1− 3− 1 −0 −0 21 05 1
21 05 01 6− 1 − −0
− − 20 20 07− 1
20 07 01 2 01
2
− 20 0− 20 0−
01
2020− 01 20 04
0 20 − −0 20 05
20 − −0
2020− 3− 20 06 1
2 0 0 20 − −0 2 07 1
20 0− 5− 1
0 0 2 0 1 20 0− −0
2 09 1
2020− 7− 1 20 0− 8−0
0 0 2 1 1 20 0− −0
2020− 9− 1 20 0− 0−0 2 1 1
1 0 2 1 1 20 1− 1−0
2 0 1
2021− 1− 1 20 1− 2−0
0 0 2 0 1 20 1− 1−0
75−79
16−29
75−79
16−29
2 03 1
age
age
2021− 1− 1 20 1− 2−0
75−79
16−29
0 0
age
21 04 1 20 1− −0
2021− 3− 1 −0 −0 21 05 1
21 05 01 − −0
− 20 6− 1
20 07− 1
20 −07 01 2 01
2
2 − 20 0− 20 0−
01
20 0− 01
0 2 0 20 05
0−15
0−15
20 0− 4−0
0−15
2020− 3− 20 − −0
70−74
70−74
2 0 0 20 06 1
70−74
20 0− 5− 1 20 − −0 2 07 1
0 0 2 0 1 20 0− −0
2 09 1
2020− 7− 1 20 0− 8−0
0 0 2 1 1 20 0− −0
2020− 9− 1 20 0− 0−0 2 1 1
1 0 2 1 1 20 1− 1−0
2021− 1− 1 20 1− 2−0 2 0 1
0 0 2 0 1 20 1− 1−0
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
2021− 1− 1 2 03 1
75−79
16−29
20 1− 2−0
80−84
30−39
0 0 21 04 1 20 1− −0
2021− 3− 1 −0 −0 21 05 1
21 05 01 − −0
− 20 6− 1
20 07− 1
20 −07 01
− 2 01
2
DOI: https://doi.org/10.25561/91875
20 0− 01
2020− 01 2 0 20 0−
0 20 05
80−84
30−39
80−84
30−39
2020− 3− 20 0− 4−0 20 − −0
80−84
30−39
20 05 01 20 06 1 2 07 1
20 − − 20 − −0
2 0 0 2 0 1 20 0− −0
0
2 9 1
20 0− 7− 1 20 0− 8−0
0 0 2 1 1 20 0− −0
2020− 9− 1 20 0− 0−0 2 1 1
1 0 2 1 1 20 1− 1−0
2021− 1− 1 20 1− 2−0 2 0 1
0 0
85−89
40−49
85−89
40−49
2 0 1 20 1− 1−0
85−89
40−49
85−89
40−49
85−89
40−49
2021− 1− 1 2 03 1
85−89
40−49
0 0 20 1− 2−0
21 04 1 20 1− −0
2021− 3− 1 21 05 1
21 05 01 −0 −0
− − 20 6− 1 − −0
20 07 01 2 01 20 07− 1
2 − 2 01
90+
90+
(L) For Salvador. (M) For São Luís. (N) For São Paulo city.
20 0− 01 20 0−
90+
20 0−
0 20 04 2 05
50−59
50−59
50−59
2020− 3− 20 − −0 20 0− −0
2 0 0 2 0 1 20 07 1
20 0− 5− 1 20 0− 6−0
0 0 2 0 1 20 − −0
2020− 7− 1 20 0− 8−0 2 09 1
0 0 2 1 1 20 0− −0
2020− 9− 1 20 0− 0−0 2 1 1
1 0 2 1 1 20 1− 1−0
2021− 1− 1 2 0 1
60−69
60−69
0 0 20 1− 2−0
60−69
2 0 1
90+
90+
20 1− 1−0
90+
2021− 1− 1
50−59
50−59
2 03 1
50−59
0 0 20 1− 2−0
21 04 1 20 1− −0
2021− 3− 1
21 05 01 −0 −0 21 05 1
−0 −0 6− 1 20 −0 −0
20 7− 1 20
7− 1 01 2 01 2
01 2020− 20 0− 20 0−
0 2 0 2 05
2020− 3− 20 0− 4−0
2 0 0 2 0 1 20 0− −0
2 07 1
20 0− 5− 1 20 0− 6−0
0 0 2 0 1 20 0− −0
2020− 7− 1 20 0− 8−0 2 09 1
0 0 2 1 1 20 0− −0
2020− 9− 1
1 0 20 0− 0−0 2 1 1
2021− 1− 1 2 1 1 20 1− 1−0
2 0 1
2 0 0 20 1− 2−0
2 0 1 20 1− 1−0
20 1− 1− 1
60−69
60−69
60−69
0 0 20 1− 2−0 2 03 1
2021− 3− 1 21 04 1 20 1− −0
21 05 01 −0 −0 21 05 1
−0 −0 6− 1 −0 −0
7− 1 01 7− 1
01 01
Figure S13: (continued) Model fits of the expected age composition of COVID‐19 attributable hospital admissions.
Imperial College COVID‐19 Response Team
Page 37 of 45
B
A
share of age groups among COVID−19 deaths share of age groups among COVID−19 deaths
20
20 2
0%
10%
20%
30%
0%
10%
20%
30%
2
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
20 0− 20 0−0
2 0 2 6
20 0− 5−0 20 0−0 −0
2 0 1 2 8 1
20 0− 7−0 20 0−1 −0
2 0 1 2 0 1
20 0− 9−0 20 0−1 −01
2 1 1 2 2
20 1− 1−0 20 1− −0
2 0 1 2 02 1
0−15
70−74
20 1− 1−0
0−15
20 1−0 −0
70−74
2 0 1
06 October 2021
20 1− 3−0 21 4− 1
21 05 1 −0 01
−0 −0 6−
7− 1 20 01
20
2 01 2
20 0− 20 0−0
20 05 2 6
20 − −0 20 0−0 −0
20 07 1 2 8 1
20 − −0 20 0−1 −0
2 0 1 2 0 1
20 0− 9−0 20 0−1 −01
2 1 1 2 2
20 1− 1−0 20 1− −0
2 0 1 2 02 1
75−79
16−29
20 1− 1−0 20 1−0 −0
75−79
16−29
2 0 1
age
age
20 1− 3−0 21 4− 1
21 05 1 −0 01
−0 −0 6−
7− 1 20 01
20 2
2 01
20 0−0
20 0−
2 6
0−15
20 05
0−15
70−74
70−74
20 − −0 20 0−0 −0
20 07 1 2 8 1
20 − −0 20 0−1 −0
2 0 1 2 0 1
20 0− 9−0 20 0−1 −01
2 1 1 2 2
20 1− 1−0 20 1− −0
2 0 1 2 02 1
75−79
16−29
20 1− 1−0
80−84
30−39
75−79
16−29
80−84
30−39
2 0 1 20 1−0 −0
21 4− 1
20 1− 3−0 −0 01
21 05 1
−0 −0 6−
20 7− 1 20 01
2 01 2
DOI: https://doi.org/10.25561/91875
20 0− 20 0−0
2 0 2 6
80−84
30−39
80−84
30−39
20 0− 5−0 20 0−0 −0
20 07 1 2 8 1
20 − −0 20 0−1 −0
2 0 1 2 0 1
20 0− 9−0 20 0−1 −01
2 1 1 2 2
20 1− 1−0 20 1− −0
2 0 1
85−89
40−49
2 02 1
85−89
40−49
20 1− 1−0
85−89
40−49
85−89
40−49
2 0 1 20 1−0 −0
21 4− 1
20 1− 3−0
21 05 1 −0 01
−0 −0 6−
20 7− 1 20 01
90+
2 2
90+
01
50−59
20 0− 20 0−0
50−59
2 0 2 6
20 0− 5−0 20 0−0 −0
20 07 1 2 8 1
20 − −0 20 0−1 −0
2 0 1 2 0 1
20 0− 9−0 20 0−1 −01
2 1 1 2 2
20 1− 1−0
60−69
2 0 1 20 1− −0
60−69
2 02 1
90+
90+
20 1− 1−0
50−59
50−59
2 0 1 20 1−0 −0
20 1− 3−0 21 4− 1
21 05 1 −0 01
−0 −0 6−
7− 1 20 01 20
01 2 2
20 0− 20 0−0
2 0 2 6
20 0− 5−0 20 0−0 −0
2 0 1 2 8 1
20 0− 7−0 20 0−1 −0
2 0 1 2 0 1
20 0− 9−0
2 1 1 20 0−1 −01
2 2
20 1− 1−0
2 0 1 20 1− −0
20 1− 1−0 2 02 1
60−69
60−69
2 0 1 20 1−0 −0
20 1− 3−0 21 4− 1
21 05 1 −0 01
−0 −0 6−
7− 1 01
01
(x‐axis). The empirical proportions are shown as dots. (A) For Belo Horizonte. (B) For Curitiba.
Figure S14: Model fits of the expected age composition of COVID‐19 attributable deaths following hospital ad‐
the y‐axis as a black line along with 95% credible intervals. Time trends are shown by week of hospital admission
week, obtained with the Bayesian multi‐strain fatality model, Supplementary Text Equation (S42), are shown on
mission. Posterior median estimates of the expected age composition of deaths in each location, age band, and
Imperial College COVID‐19 Response Team
Page 38 of 45
E
C
D
share of age groups among COVID−19 deaths share of age groups among COVID−19 deaths share of age groups among COVID−19 deaths
20
2 20 20
2
0%
20%
40%
60%
0%
20%
40%
60%
20 0− 2
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0.0%
10.0%
20.0%
30.0%
40.0%
0.0%
10.0%
20.0%
30.0%
40.0%
2 05 20 0− 20 0−
20 0− −0 2 05 2 05
2 07 1 20 0− −0 20 0− −0
20 0− −0 2 07 1 2 07 1
2 09 1 20 0− −0 20 0− −0
20 0− −0 2 09 1 2 0 1
2 1 1 20 0− −0 20 0− 9−0
20 1− 1−0 2 1 1 2 1 1
2 0 1 20 1− 1−0 20 1− 1−0
2 0 1
20 1− 1−0 2 0 1
0−15
70−74
2 03 1 20 1− 1−0
0−15
20 1− 1−0
0−15
70−74
2 0 1
70−74
20 1− −0 2 03 1
06 October 2021
21 05 1 20 1− −0 20 1− 3−0
− −0 21 05 1 21 05 1
20 07− 1 − −0 − −0
2 01 20 07− 1 20 07− 1
2
20 0− 2 01 01
2 05 20 0− 20 0−
20 0− −0 20 05 20 05
20 07 1 20 − −0 20 − −0
20 − 0 − 2 07 1 2 07 1
2 09 1 20 0− −0 20 0− −0
20 0− −0 2 09 1 2 0 1
2 1 1 20 0− −0 20 0− 9−0
2 1 1
20 1− 1−0 2 1 1
2 0 1 20 1− 1−0 20 1− 1−0
2 0 1
20 1− 1−0 2 0 1
75−79
16−29
2 03 1 20 1− 1−0 20 1− 1−0 75−79
16−29
2 0 1
age
75−79
16−29
2 03 1
age
age
20 1− −0 20 1− 3−0
21 05 1 20 1− −0
− −0 21 05 1 21 05 1
− −0 − −0
20 07− 1 20 07− 1
2 01 20 07− 1
2 01 2 01
20 0− 20 0− 20 0−
2 05 2 05
0−15
20 05
0−15
0−15
20 0− −0
70−74
20 0− −0
70−74
20 − −0
70−74
20 07 1
20 − 0 − 2 07 1 20 07 1
20 − −0
2 09 1 20 0− −0
0
2 9 1 2 0 1
20 0− −0 20 0− 9−0
2 1 1 20 0− −0
2 1 1 2 1 1
20 1− 1−0 20 1− 1−0
2 0 1 20 1− 1−0
2 0 1 2 0 1
20 1− 1−0
75−79
16−29
20 1− 1−0
80−84
30−39
2 03 1 20 1− 1−0
75−79
16−29
80−84
30−39
2 0 1
75−79
16−29
80−84
30−39
20 1− −0 2 03 1
20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
− −0 − −0 − −0
20 07− 1 20 07− 1 20 07− 1
2 01 2 01 2 01
DOI: https://doi.org/10.25561/91875
20 0− 20 0− 20 0−
20 05 2 05 2 05
80−84
30−39
80−84
30−39
80−84
30−39
20 − −0 20 0− −0 20 0− −0
20 07 1
20 − 0 − 20 07 1
20 − −0
20 07 1
20 − −0
2 09 1 2 09 1 2 0 1
20 0− −0 20 0− −0 20 0− 9−0
2 1 1 2 1 1 2 1 1
20 1− 1−0 20 1− 1−0 20 1− 1−0
2 0 1 2 0 1 2 0 1
85−89
40−49
85−89
20 1− 1−0 40−49
85−89
40−49
20 1− 1−0 20 1− 1−0
85−89
40−49
2 03 1
85−89
40−49
2 0 1
85−89
40−49
20 1− −0 2 03 1
20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
− −0 − −0 − −0
20 07− 1 20 07− 1 20 07− 1
2 2
90+
01 2
90+
01
90+
20 0− 01
20 0− 20 0−
50−59
50−59
50−59
20 05 2 05 2 05
20 − −0 20 0− −0 20 0− −0
2 07 1 20 07 1 20 07 1
20 0− −0 20 − 0 − 20 − 0 −
2 09 1 2 09 1 2 0 1
20 0− −0 20 0− −0 20 0− 9−0
2 1 1 2 1 1 2 1 1
20 1− 1−0 20 1− 1−0 20 1− 1−0
60−69
2 0 1 2 0 1 2 0 1
60−69
60−69
90+
90+
90+
20 1− 1−0 20 1− 1−0 20 1− 1−0
50−59
50−59
2 03 1 2 0 1
50−59
2 03 1
20 1− −0 20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
−0 −0 −0 −0 −0 −0
7− 1 20 7− 1 20 7− 1 20
01 2 01 2 01 2
20 0− 20 0− 20 0−
2 05 2 05 2 05
20 0− −0 20 0− −0 20 0− −0
2 07 1 2 07 1 2 07 1
20 0− −0 20 0− −0 20 0− −0
2 09 1 2 09 1 2 0 1
20 0− −0 20 0− −0 20 0− 9−0
2 1 1 2 1 1 2 1 1
20 1− 1−0 20 1− 1−0 20 1− 1−0
2 0 1 2 0 1 2 0 1
20 1− 1−0 20 1− 1−0 20 1− 1−0
60−69
60−69
60−69
2 03 1 2 03 1 2 0 1
20 1− −0 20 1− −0 20 1− 3−0
21 05 1 21 05 1 21 05 1
−0 −0 −0 −0 −0 −0
hospital admission. (C) For Florianópolis. (D) For Goiânia. (E) For João Pessoa.
7− 1 7− 1 7− 1
01 01 01
Figure S14: (continued) Model fits of the expected age composition of COVID‐19 attributable deaths following
Imperial College COVID‐19 Response Team
Page 39 of 45
F
H
G
share of age groups among COVID−19 deaths share of age groups among COVID−19 deaths share of age groups among COVID−19 deaths
20 20 20
2 2 2
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
20 0− 20 0− 20 0−0
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
2 05 2 0 2 5
20 0− −0 20 0− 4−0 20 0−0 −0
2 07 1 2 0 1 2 7 1
20 0− −0 20 0− 6−0 20 0−0 −0
2 09 1 2 0 1 2 9 1
20 0− −0 20 0− 8−0 20 0−1 −0
2 1 1 2 1 1 2 1 1
20 1− 1−0 20 0− 0−0 20 1− −0
2 0 1 2 1 1 2 01 1
20 1− 1−0 20 1− 2−0 20 1−0 −0
2 0 1
0−15
0−15
0−15
70−74
70−74
2 03 1 2 3 1
70−74
06 October 2021
20 1− −0 20 1− 2−0 20 1−0 −0
21 05 1 21 04 1 21 5− 1
− −0 −0 −0
20 6− 1 20 −07 01
20 07− 1
2 01 2 01 2 −
20 0− 20 0− 20 0−0 01
20 05 2 0 2 5
20 − −0 20 0− 4−0 20 0−0 −0
2 07 1 20 06 1
20 − −0
20 7− 1
20 −0 0
20 0− −0
0
2 9 1 2 0 1 2 9 1
20 0− −0 20 0− 8−0 20 0−1 −0
2 1 1 2 1 1 2 1 1
20 1− 1−0 20 0− 0−0 20 1− −0
2 0 1 2 1 1 2 01 1
20 1− 1−0 20 1− 2−0
2 0 1 20 1−0 −0
75−79
16−29
75−79
16−29
2 03 1 2 3 1
75−79
16−29
age
age
age
20 1− −0 20 1− 2−0 20 1−0 −0
21 05 1 21 04 1 21 5− 1
− −0 −0 −0
20 6− 1 20 −07 01
20 07− 1
2 01 2 01 2 −
20 0− 20 0− 20 0−0 01
2 05 2 0 2 5
0−15
0−15
0−15
20 0− −0 20 0− 4−0 20 0−0 −0
70−74
70−74
70−74
20 06 1 20 7− 1
20 07 1
20 − −0 20 − 0 − 20 −0 0
2 09 1 2 0 1 2 9 1
20 0− −0 20 0− 8−0 20 0−1 −0
2 1 1 2 1 1 2 1 1
20 1− 1−0 20 0− 0−0 20 1− −0
2 0 1 2 1 1 2 01 1
20 1− 2−0
20 1− 1−0 2 0 1 20 1−0 −0
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
2 3 1
75−79
16−29
2 03 1
80−84
30−39
20 1− 2−0
20 1− −0 21 04 1 20 1−0 −0
21 05 1 −0 −0 21 5− 1
− −0 6− 1 20 −07 01
20
20 07− 1
2 01 2 01 2 −
DOI: https://doi.org/10.25561/91875
20 0− 20 0− 20 0−0 01
2 05 20 04 2 5
80−84
30−39
80−84
30−39
80−84
30−39
20 0− −0 20 − −0 20 0−0 −0
20 07 1 2 0 1 2 7 1
20 − 0 − 20 0− 6−0 20 0−0 −0
2 09 1 2 0 1 2 9 1
20 0− −0 20 0− 8−0 20 0−1 −0
2 1 1 2 1 1 2 1 1
20 1− 1−0 20 0− 0−0 20 1− −0
2 0 1 2 1 1 2 01 1
20 1− 2−0
85−89
40−49
85−89
40−49
85−89
40−49
20 1− 1−0 2 0 1 20 1−0 −0
85−89
40−49
85−89
40−49
2 3 1
85−89
40−49
2 03 1 20 1− 2−0
20 1− −0 21 04 1 20 1−0 −0
21 05 1 −0 −0 21 5− 1
− −0 6− 1 20 −07 01
20
20 07− 1
2 2 01 2 −
90+
90+
01
90+
20 0− 20 0− 20 0−0 01
2 5
50−59
50−59
20 04
50−59
20 05 20 − −0 20 0−0 −0
20 − −0 2 0 1 2 7 1
20 07 1
20 − 0 − 20 0− 6−0 20 0−0 −0
2 09 1 2 0 1 2 9 1
20 0− 8−0 20 0−1 −0
20 0− −0
2 1 1 2 1 1 2 1 1
20 0− 0−0
20 1− 1−0 2 1 1 20 1− −0
2 0 1 2 01 1
60−69
60−69
60−69
20 1− 2−0
90+
90+
2 0 1
90+
20 1− 1−0 20 1−0 −0
50−59
50−59
2 3 1
50−59
2 03 1 20 1− 2−0
20 1− −0 21 04 1 20 1−0 −0
21 05 1 −0 −0 21 5− 1
−0 −0 6− 1 −0 0 20
7− 1 20 20 7− 1
2 2
hospital admission. (F) For Macapá. (G) For Manaus. (H) For Natal.
01
01 2 20 0− 01 20 0−0
20 0−
2 05 2 0 2 5
20 0− −0 20 0− 4−0 20 0−0 −0
2 07 1 2 0 1 2 7 1
20 0− 6−0 20 0−0 −0
20 0− −0
2 09 1 2 0 1 2 9 1
20 0− 8−0
20 0− −0
2 1 1 2 1 1 20 0−1 −0
2 1 1
20 0− 0−0 20 1− −0
20 1− 1−0
2 0 1 2 1 1 2 01 1
20 1− 2−0
20 1− 1−0 2 0 1 20 1−0 −0
60−69
60−69
60−69
2 03 1 20 1− 2−0 2 3 1
20 1− −0 21 04 1 20 1−0 −0
21 05 1 −0 −0 21 5− 1
−0 −0 6− 1 −0 0
7− 1 01 7− 1
01 01
Figure S14: (continued) Model fits of the expected age composition of COVID‐19 attributable deaths following
Imperial College COVID‐19 Response Team
Page 40 of 45
I
K
share of age groups among COVID−19 deaths share of age groups among COVID−19 deaths share of age groups among COVID−19 deaths
20 20
2 2 20
0%
10%
20%
30%
0%
10%
20%
30%
20 0− 2
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
20 0−0
2 05 2 5 20 0−
20 0−0 −0 20 0−0 −0 2 0
2 7 1 2 7 1 20 0− 5−0
20 0− −0 20 0−0 −0 2 0 1
2 09 1 2 9 1 20 0− 7−0
20 0− −0 20 0−1 −0 2 0 1
2 11 1 2 1 1
20 0− 9−0
20 1− −0 20 1− −0 2 1 1
2 01 1 2 01 1
20 1− 1−0
20 1− −0 2 0 1
0−15
70−74
2 03 1 20 1−0 −0 20 1− 1−0
0−15
0−15
70−74
70−74
20 1− −0 2 3 1 2 0 1
06 October 2021
21 05 1 20 1−0 −0 20 1− 3−0
− −0 21 5− 1 21 05 1
20 07− 1 20 −07 01 −0 −0
2 01 2 − 20 7− 1
20 0− 20 0−0 01 2 01
20 05 2 5 20 0−
20 −0 −0 20 0−0 −0 20 05
2 7 1 20 7− 1 20 − −0
20 0− −0 20 −0 0 20 07 1
2 09 1 2 9 1 20 − −0
20 0− −0 20 0−1 −0 2 0 1
2 11 1 2 1 1
20 0− 9−0
20 1− −0 20 1− −0 2 1 1
2 01 1 20 1− 1−0
20 1− −0 2 01 1 2 0 1
75−79
16−29
2 03 1 20 1−0 −0 20 1− 1−0
75−79
16−29
75−79
16−29
2 3 1 2 0 1
age
age
age
20 1− −0 20 1−0 −0 20 1− 3−0
21 05 1 21 5− 1 21 05 1
− −0 −0 −0
20 07− 1 20 −07 01
2 01 2 − 20 7− 1
20 0− 20 0−0 01 2 01
20 05 2 5 20 0−
0−15
20 05
0−15
0−15
20 −0 −0 20 0−0 −0
70−74
70−74
70−74
2 7 1 20 7− 1 20 − −0
20 0− −0 20 −0 0 20 07 1
2 09 1 2 9 1 20 − −0
20 0− −0 20 0−1 −0 2 0 1
2 11 1 20 0− 9−0
20 1− −0 2 1 1 2 1 1
20 1− −0 20 1− 1−0
2 01 1 2 01 1 2 0 1
20 1− −0
75−79
16−29
20 1−0 −0 20 1− 1−0
80−84
30−39
2 03 1
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
20 1− −0 2 3 1 2 0 1
21 05 1
20 1−0 −0 20 1− 3−0
− −0
21 5− 1 21 05 1
20 07− 1 20 −07 01 −0 −0
2 01 2 − 20 7− 1
2 01
DOI: https://doi.org/10.25561/91875
20 0− 20 0−0 01
20 05 2 5 20 0−
2 0
80−84
30−39
80−84
30−39
80−84
30−39
20 −0 −0 20 0−0 −0 20 0− 5−0
2 7 1 2 7 1 20 07 1
20 0− −0 20 0−0 −0 20 − −0
2 09 1 2 9 1 2 0 1
20 0− −0 20 0−1 −0 20 0− 9−0
2 11 1 2 1 1 2 1 1
20 1− −0 20 1− −0 20 1− 1−0
2 01 1 2 01 1 2 0 1
85−89
40−49
85−89
40−49
85−89
20 1− −0 40−49
20 1−0 −0 20 1− 1−0
85−89
40−49
2 03 1
85−89
40−49
85−89
40−49
2 3 1 2 0 1
20 1− −0 20 1−0 −0 20 1− 3−0
21 05 1 21 5− 1 21 05 1
− −0 −0 −0
20 07− 1 20 −07 01
2 2 − 20 7− 1
90+
01 2
90+
90+
20 0− 20 0−0 01 01
50−59
2 5 20 0−
50−59
50−59
20 05 2 0
20 −0 −0 20 0−0 −0 20 0− 5−0
2 7 1 2 7 1 20 07 1
20 0− −0 20 0−0 −0 20 − −0
2 09 1 2 9 1 2 0 1
20 0− −0 20 0−1 −0 20 0− 9−0
2 11 1 2 1 1 2 1 1
20 1− −0 20 1− −0 20 1− 1−0
60−69
2 01 1 2 01 1 2 0 1
60−69
60−69
90+
90+
90+
20 1− −0 20 1−0 −0 20 1− 1−0
50−59
50−59
50−59
2 03 1 2 3 1 2 0 1
20 1− −0 20 1−0 −0 20 1− 3−0
21 05 1 21 5− 1 21 05 1
−0 −0 −0 0 −0 −0
7− 1 20 7− 1 20
2 2 7− 1 20
01
20 0−
01 20 0−0 01 2
2 05 2 5 20 0−
20 0−0 −0 20 0−0 −0 2 0
2 7 1 20 0− 5−0
2 7 1 2 0 1
20 0− −0 20 0−0 −0 20 0− 7−0
2 09 1 2 9 1 2 0 1
20 0− −0 20 0−1 −0 20 0− 9−0
2 11 1 2 1 1 2 1 1
20 1− −0 20 1− −0 20 1− 1−0
2 01 1 2 01 1 2 0 1
20 1− −0 20 1−0 −0 20 1− 1−0
60−69
60−69
60−69
2 03 1 2 3 1 2 0 1
20 1− −0 20 1−0 −0 20 1− 3−0
21 05 1 21 5− 1 21 05 1
−0 −0 −0 0 −0 −0
7− 1 7− 1 7− 1
01 01 01
hospital admission. (I) For Porto Alegre. (J) For Porto Velho. (K) For Rio De Janeiro.
Figure S14: (continued) Model fits of the expected age composition of COVID‐19 attributable deaths following
Imperial College COVID‐19 Response Team
Page 41 of 45
L
N
M
share of age groups among COVID−19 deaths share of age groups among COVID−19 deaths share of age groups among COVID−19 deaths
20 20
2 2 20
20 0− 20 0− 2
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
0%
25%
50%
75%
100%
2 0
0.0%
10.0%
20.0%
30.0%
40.0%
0.0%
10.0%
20.0%
30.0%
40.0%
2 0 20 0−
20 0− 3− 20 0− 4−0 2 05
2 0 0 2 0 1 20 0− −0
20 0− 5− 1 20 0− 6−0 2 07 1
2 0 0 2 0 1 20 0− −0
20 0− 7− 1 20 0− 8−0 2 09 1
2 0 0 2 1 1 20 0− −0
20 0− 9− 1
1 0 20 0− 0−0 2 1 1
2021− 1− 1 2 1 1 20 1− 1−0
2 0 0 20 1− 2−0 2 0 1
2 0 1
0−15
0−15
20 1− 1− 1 20 1− 1−0
0−15
70−74
70−74
70−74
21 03 01 20 1− 2−0 2 03 1
06 October 2021
20 − − 21 04 1 20 1− −0
21 05 01 −0 −0 21 05 1
− 6− 1 − −0
20 −07 01 20
2 − 2 01 20 07− 1
2
20 0− 01 20 0− 01
2 0 2 0 20 0−
20 0− 3− 20 0− 4−0 20 05
2 0 0 2 0 1 20 − −0
20 0− 5− 1 20 0− 6−0 2 07 1
2 0 0 2 0 1 20 0− −0
20 0− 7− 1 20 0− 8−0 2 09 1
2 0 0 2 1 1 20 0− −0
20 0− 9− 1 2 1 1
1 0 20 0− 0−0
2021− 1− 1 2 1 1 20 1− 1−0
2 0 0 20 1− 2−0 2 0 1
20 1− 1− 1 2 0 1 20 1− 1−0
75−79
16−29
75−79
16−29
75−79
16−29
2 0 0 20 1− 2−0 2 03 1
age
age
age
20 1− 3− 1 21 04 1 20 1− −0
21 05 01 −0 −0 21 05 1
− 6− 1 − −0
20 −07 01 20
2 − 2 01 20 07− 1
2
20 0− 01 20 0− 01
2 0 20 04 20 0−
20 0− 3− 20 05
0−15
0−15
20 − −0
0−15
2 0 0 2 0 1
70−74
70−74
20 − −0
70−74
20 0− 5− 1 20 0− 6−0 2 07 1
2 0 0 2 0 1 20 0− −0
20 0− 7− 1 20 0− 8−0 0
2 9 1
2 0 0 2 1 1 20 0− −0
20 0− 9− 1 2 1 1
1 0 20 0− 0−0
2021− 1− 1 2 1 1 20 1− 1−0
2 0 0 20 1− 2−0 2 0 1
20 1− 1− 1 2 0 1 20 1− 1−0
75−79
16−29
75−79
16−29
80−84
30−39
80−84
30−39
75−79
16−29
80−84
30−39
2 0 0 20 1− 2−0 2 03 1
20 1− 3− 1 21 04 1 20 1− −0
21 05 01 −0 −0 21 05 1
− 6− 1 − −0
20 −07 01 20
2 − 2 01 20 07− 1
2 01
20 0− 01 20 0−
DOI: https://doi.org/10.25561/91875
2 0 20 04 20 0−
2 05
20 0− 3−
80−84
30−39
80−84
30−39
0 20 − −0
80−84
30−39
2 0 2 0 1 20 0− −0
20 0− 5− 1 20 0− 6−0 20 07 1
2 0 0 2 0 1 20 − −0
20 0− 7− 1 20 0− 8−0 2 09 1
20 09 01
20 − − 2 1 1 20 0− −0
1 0 20 0− 0−0 2 1 1
2021− 1− 1 2 1 1 20 1− 1−0
2 0 0 20 1− 2−0 2 0 1
85−89
40−49
85−89
40−49
2 0 1
85−89
40−49
20 1− 1− 1 20 1− 1−0
85−89
40−49
85−89
40−49
85−89
40−49
2 0 0 20 1− 2−0 2 03 1
20 1− 3− 1 21 04 1 20 1− −0
21 05 01 −0 −0 21 05 1
− 6− 1 − −0
20 −07 01 20
2 − 2 01 20 07− 1
2
90+
90+
90+
20 0− 01 20 0− 01
2 0 2 0 20 0−
50−59
50−59
50−59
20 0− 3− 20 0− 4−0 2 05
20 05 01 2 0 1 20 0− −0
20 − − 20 07 1
2 0 0
20 0− 6−0
2 0 1 20 − 0 −
20 0− 7− 1 20 0− 8−0 2 09 1
2 0 0 2 1 1 20 0− −0
20 0− 9− 1
1 0 20 0− 0−0 2 1 1
2021− 1− 1 2 1 1 20 1− 1−0
2 0 1
60−69
60−69
60−69
2 0 0 20 1− 2−0
2 0 1
90+
90+
90+
20 1− 1− 1 20 1− 1−0
50−59
50−59
50−59
2 0 0 20 1− 2−0 2 03 1
20 1− 3− 1 21 04 1 20 1− −0
21 05 01 −0 −0 21 05 1
−0 −0 20 6− 1 20 −0 −0
7− 1 2 7− 1 20
01 2 01
01 2
20 0− 20 0−
2 0 2 0 20 0−
2 05
20 0− 3− 20 0− 4−0
2 0 0 2 0 1 20 0− −0
20 0− 5− 1 20 0− 6−0 2 07 1
2 0 0 2 0 1 20 0− −0
20 0− 7− 1 20 0− 8−0 2 09 1
2 0 0 2 1 1 20 0− −0
20 0− 9− 1
1 0 20 0− 0−0 2 1 1
2021− 1− 1 2 1 1 20 1− 1−0
2 0 0 20 1− 2−0 2 0 1
20 1− 1− 1 2 0 1 20 1− 1−0
60−69
60−69
60−69
2 0 0 20 1− 2−0 2 03 1
hospital admission. (L) For Salvador. (M) For São Luís. (N) For São Paulo city.
20 1− 3− 1 21 04 1 20 1− −0
21 05 01 −0 −0 21 05 1
−0 −0 6− 1 −0 −0
7− 1 01 7− 1
01 01
Figure S14: (continued) Model fits of the expected age composition of COVID‐19 attributable deaths following
Imperial College COVID‐19 Response Team
Page 42 of 45
06 October 2021 Imperial College COVID‐19 Response Team
2.0
Belo Horizonte
Curitiba
ratio of Gamma vs non−Gamma
Florianópolis
Goiânia
1.5
João Pessoa
Macapá
Manaus
Natal
1.0
Porto Alegre
Porto Velho
Rio de Janeiro
0.5 Salvador
São Luís
São Paulo
15
+
−2
−3
−4
−5
−6
−7
−7
−8
−8
90
0−
16
30
40
50
60
70
75
80
85
Figure S15: Estimated ratio in the share of age groups in Gamma versus non‐Gamma residents’ hospital admis‐
sions. Posterior median estimates are shown as black horizontal bar, 50% interquantile ranges as box and 95%
credible intervals as whiskers. Estimates were derived independently for each location (colour) and control for
changes in the population at risk of a fatal outcome over time, see Supplementary Text, page 50. Estimates for
which credible interval width was larger than 3 were removed from the figure.
60%
50%
50%
50%
0%
25% 40%
25%
−50%
0%
20%
−100%
0%
−25%
−150%
0%
projected avoidable COVID−19 attributable deaths in hospitals
60%
50.0% 50.0%
40%
40.0% 40.0%
40%
30.0% 30.0%
20%
20.0%
20.0%
20%
10.0%
10.0%
0%
or u te
Jo G ópo a
ão o lis
M ssoa
M capa
an á
rto N us
R Por Al tal
de V re
Sa ne o
S lva ro
Sã ão dor
Pa ís
o
n ib
Peiâni
Ja elh
ul
o Lu
Fl C izon
io to eg
i
a
a
80−84 85−89 90+ ia rit
a
or
H
Po
lo
Be
50.0%
40.0%
40.0%
40.0%
30.0%
30.0%
30.0%
20.0%
20.0% 20.0%
10.0% 10.0%
10.0%
0.0%
0.0%
0.0%
or u te
Jo G ópo a
ão o lis
M ssoa
M capa
an á
rto N us
R Por Al tal
de V re
Sa ne o
S lva ro
Sã ão dor
Pa ís
o
or u te
Jo G ópo a
ão o lis
M ssoa
M capa
an á
rto N us
R Por Al tal
de V re
Sa ne o
S lva ro
Sã ão dor
Pa ís
o
or u te
Jo G ópo a
ão o lis
M ssoa
M capa
an á
rto N us
R or Al tal
de V re
Sa e o
S lva ro
Sã ão dor
Pa ís
o
n ib
Peiâni
Ja elh
ul
n ib
Peiâni
Ja lh
ul
n ib
Peiâni
Ja lh
ul
o Lu
o Lu
o Lu
Fl C izon
Fl C izon
Fl C izon
io to eg
io to eg
io to eg
i
a
a
a
a
a
a
ia rit
ia rit
ia rit
e
n
a
a
or
or
or
H
H
Po
Po
Po
P
lo
lo
lo
Be
Be
Be
Figure S16: Projected proportion of avoidable COVID‐19 attributable deaths in hospitals without pandemic re‐
source limitations. The projections are based on counterfactual simulations assuming the lowest COVID‐19 in‐
hospital fatality rates that were achieved prior to Gamma’s detection in the corresponding location. Posterior
median estimates of the proportion of avoidable COVID‐19 attributable in‐hospital deaths are shown as black
horizontal bar, 50% interquantile ranges as box, and 95% credible intervals as whiskers.
60%
50%
60% 60.0%
40%
0% 40.0%
20%
in the absence of location inequities and pandemic healthcare pressure
50.0%
50.0%
40.0% 40.0%
40.0% 40.0%
30.0% 30.0%
30.0% 30.0%
20.0% 20.0%
20.0%
or u te
Jo G ópo a
ão o lis
M ssoa
M capa
an á
rto N us
R Por Al atal
de V re
Saane o
S lva ro
Sã ão dor
Pa ís
o
n ib
Peiâni
J elh
ul
o Lu
Fl C izon
io to eg
i
a
80−84 85−89 90+ ia rit
a
or
70.0%
H
Po
lo
Be
60%
60.0% 60%
50.0%
40%
40.0% 40%
30.0%
20%
20%
20.0%
or u te
Jo G ópo a
ão o lis
M ssoa
M capa
an á
rto N us
R or Al atal
de V re
Saane o
S lva ro
Sã ão dor
Pa ís
o
or u te
Jo G ópo a
ão o lis
M ssoa
M capa
an á
rto N us
R Por Al atal
de V re
Saane o
S lva ro
Sã ão dor
Pa ís
o
or u te
Jo G ópo a
ão o lis
M ssoa
M capa
an á
rto N us
R Por Al atal
de V re
Saane o
S lva ro
Sã ão dor
Pa ís
o
n ib
Peiâni
J elh
ul
n ib
Peiâni
J elh
ul
n ib
Peiâni
J elh
ul
o Lu
o Lu
o Lu
Fl C izon
Fl C izon
Fl C izon
io to eg
io to eg
io to eg
i
a
a
ia rit
ia rit
ia rit
a
a
or
or
or
H
H
Po
Po
Po
P
lo
lo
lo
Be
Be
Be
Figure S17: Projected proportion of avoidable COVID‐19 attributable deaths in hospitals without location in‐
equities and without pandemic resource limitations. The projections are based on counterfactual simulations
assuming the lowest COVID‐19 in‐hospital fatality rates that were achieved prior to Gamma’s detection across all
locations. Posterior median estimates of the proportion of avoidable COVID‐19 attributable in‐hospital deaths are
shown as black horizontal bar, 50% interquantile ranges as box, and 95% credible intervals as whiskers.