PreprintPDF Available

A COVID-19 Activity Risk Calculator as a Gamified Public Health Intervention

Authors:
Preprints and early-stage research may not have been peer reviewed yet.

Abstract

The public has consistently requested information about current COVID-19 risk levels associated with day-today activities, such as dining out and social gatherings. Risk has been highly variable during the pandemic due to its dependency on rapidly evolving factors such as community transmission levels and variants. There has also been confusion surrounding certain personal protective strategies such as risk reduction by mask wearing and vaccination. Here we describe COVrisk, our COVID-19 Activity Risk Calculator, which provides real time COVID-19 infection, death, and hospitalization risk ranges associated with activities for more than 250 countries at a regional level. While maintaining a streamlined user friendly interface, we factor in a sophisticated set of inputs including prevalence of variants, vaccine variant coverage, indoor vs. outdoor risk, under-reporting corrected case numbers, and mask types. The calculator serves as a gamified public health intervention as users can "play with" how their risk would change depending on factors such as hosting events outdoors vs. indoors, wearing a high quality mask, and getting vaccinated. Empowering the public to make informed data-driven decisions about safely engaging in activities may help to reduce COVID-19 levels in the community. We also demonstrate quantitatively the increased impact of interventions such as mask wearing when cases are higher, which could inform and support policy decisions around mask mandate case thresholds and other non-pharmaceutical interventions.
A COVID-19 Activity Risk Calculator as a Gamified1
Public Health Intervention2
Shreyasvi Natraj1,3, *, Nathan Yap3, Agrima Seth4, Malhar Bhide1,2, Leila Orszag1,2, Pawan3
Nandakishore1,2, Nina Rescic1,2, Shanice Hudson5, Carlos Baquero6, Davide Frey6,4
Antonio Fernandez Anta6, and Christin Glorioso2,3,6,*
5
1Psychiatry Department, Faculty of Medicine, University of Geneva, 1211, Geneva, Switzerland6
2University of California, San Francisco, Department of Anatomy, San Francisco, CA, USA7
3Academics for the Future of Science Inc., Cambridge, MA, USA8
4School of Information, University of Michigan, Ann Arbor, MI, USA9
5Hood Medicine Initiative, Inc, P.O. Box 55458, Indianapolis, IN, USA10
6CoronaSurveys Team, coronasurveys.org11
*shreyasvi.natraj@unige.ch12
*christin.glorioso@ucsf.edu13
ABSTRACT
14
The public has consistently requested information about current COVID-19 risk levels associated with day-to-day activities,
such as dining out and social gatherings. Risk has been highly variable during the pandemic due to its dependency on
rapidly evolving factors such as community transmission levels and variants. There has also been confusion surrounding
certain personal protective strategies such as risk reduction by mask wearing and vaccination. Here we describe COVrisk,
our COVID-19 Activity Risk Calculator, which provides real time COVID-19 infection, death, and hospitalization risk ranges
associated with activities for more than 250 countries at a regional level. While maintaining a streamlined user friendly interface,
we factor in a sophisticated set of inputs including prevalence of variants, vaccine variant coverage, indoor vs. outdoor risk,
under-reporting corrected case numbers, and mask types. The calculator serves as a gamified public health intervention as
users can "play with" how their risk would change depending on factors such as hosting events outdoors vs. indoors, wearing
a high quality mask, and getting vaccinated. Empowering the public to make informed data-driven decisions about safely
engaging in activities may help to reduce COVID-19 levels in the community. We also demonstrate quantitatively the increased
impact of interventions such as mask wearing when cases are higher, which could inform and support policy decisions around
mask mandate case thresholds and other non-pharmaceutical interventions.
15
Introduction16
Risk levels for gatherings and activities have fluctuated significantly since the pandemic began. The risk of becoming
17
infected with SARS-CoV-2 in the worst set ups (indoor and crowded) can range from greater than 60% at peak community
18
case levels to less than 0.01% when cases are at their lowest. Determinants of infection, hospitalization, and death risk include
19
community transmission levels, size of gatherings, social distancing, vaccination status, type of vaccine, dose of vaccine, face
20
mask usage, air filtration, circulating SARS-CoV-2 variants, health conditions, age, gender, previous infection, time since
21
previous infection or vaccination (due to waning immunity). Because risk levels are so rapidly evolving and have so many
22
determinants, it has been confusing for the public to maintain a clear picture of what their risk is for various activities over
23
time. The need has been so great to understand risk, that some non-scientist citizens have created tools
1
to try to assess risk
24
for their households and general audience publications such as the New York Times have attempted to create risk assessment
25
algorithms (https://www.nytimes.com/2020/12/04/briefing/california-covid-restrictions-warner-bros-stimulus.html). These
26
manual attempts can be complicated to use and at the same time lacking in enough determinants to accurately predict risk.
27
Confusion over cost-benefit surrounding harm reduction strategies such as use of face masks, restrictions on gathering size, and
28
vaccination, has further created an "infodemic" and undoubtedly cost many people their lives or livelihoods.29
To address aspects of these issues, research teams have created various tools for individuals to evaluate their risk, each
30
taking a somewhat different approach. A simple approach taken by many Dashboards, including the US Center for Disease
31
Control (CDC) is to report risk levels on a county-wide basis simply by community transmission levels
2
. While this is a
32
simple to understand and a useful metric, it doesn’t address differences in risk by either individual risk factors such as age and
33
health condition, differences in activities such as number of attendees, or differences in precautions such as mask wearing or
34
RISK CALCULATOR IS CONSIDERS CALCULATES RISK OF
Fast Vaccination Health
Conditions
Many
Locations
Mask
Type
Number of
People
Indoor or
Outdoor Variants Specific
Activities Infection Hospitalization Death
COVrisk (Our Model) X X X X X X X X X X X
QCovid8X X X
ASIMI Model9X X X X
19 an Me
(Princeton Model)4X X X X
MyCOVIDRisk6X X X X X X
COVID-19
Assessment Tool
(GA Tech Model)3
X X X
Covidtracker.fr10 X X
Max Planck Institute
COVID Risk Calculator11 X X X X
COVID-19 Indoor Safety
Guide (MIT Model)12 X X X X X
COVID-19 Risk Calculator
(Northwestern University
Model)13
X X X
microCOVID Project1X X X X X X
Table 1. Comparison between different Existing Risk calculators and our risk calculator based on different parameters.
vaccination. All of these factors are important and can change risk drastically.35
The approach factoring in the smallest number of determinants is the COVID-19 Event Risk Assessment Planning Tool, a
36
web-based tool
3
, developed by scientists at the Georgia Institute of Technology. The tool estimates the probability that a person
37
will encounter someone infected with COVID-19 at a particular gathering based on the size of the group and the location of the
38
event (factoring in under reporting adjusted community transmission levels). The tool works for the US only, at the county level,
39
and does not take into account the individual’s health risk factors, vaccination status, variants, mask usage, or indoor v. outdoor
40
activities. It simply predicts the chances that someone at a gathering of a user-set size will be actively infected with COVID-19.
41
The 19 and Me calculator
4
, developed by Mathematica, a policy-research company in Princeton, New Jersey, draws on
42
demographic and health information and user behaviors such as hand washing and masks used to determine the relative risk of
43
exposure, infection, and severe illness for the individual’s behavior on a weekly basis. It does not account for variants and
44
works for the US and Belgium only.45
In December 2020, a team led by biostatistician Nilanjan Chatterjee at Johns Hopkins University in Baltimore, Maryland,
46
released the COVID-19 Mortality Risk Calculator
5
, which estimates an individual’s relative risk of death from COVID-19
47
during an activity based on their location, pre-existing conditions, and general health status. It also reports estimated risk of
48
death in a user’s area in the next two weeks. Activities, personal precautions, vaccination, and variants are not factored in.49
Another approach for risk calculator called MyCOVIDRisk
6
takes a more situational approach, estimating the risks
50
associated with specific errands or recreational activities. The estimate is based on the location and duration and the number of
51
masked or unmasked people attending. This can help users avoid activities that are likely to be high risk in a specific pandemic
52
hotspot, such as spending an hour or more at an indoor gym, favoring safer alternatives — a masked meet-up in the park, for
53
instance
7
. It does not report risk numerically, instead opting for a low-very high scale, which does not take into account an
54
individuals own risk tolerance threshold. It also does not account for variants.55
We build upon the comprehensiveness of these tools both in terms of factors taken into account and geographical locations.
56
To our knowledge, this method of estimation is the only one that takes into account variants and vaccine coverage of them
57
and works in almost every country of the world. The app is lightweight and can be used on low bandwidth internet, which is
58
an important factor in many countries. We aim to aid people around the world to make informed decisions about how to do
59
activities more safely during a global Pandemic.60
Results61
For illustrative purposes here we show a few scenarios using our risk calculation system over time. We show that level of
62
risk increases with presence of high number of active cases, older age, less vaccination, and lower quality or no face mask usage
63
(Figure 1A-D). Activity risk also increases with indoor activities, more contacts at the event, vaccine type, health conditions,
64
male sex, and prevalence of variants of concern (not shown). Values used to toggle risk can be found in Table 4 and the flow
65
diagram that outlines the risk calculator’s backend code can be found in Figure 3.66
2/9
Change in Risk of Infection, Hospitalization and Death with Community Transmission Levels and Age67
In (Figure 1A) We first use take the instance of a 30 year old male with no chronic health conditions in Massachusetts,
68
USA. We observe that in the scenario of no vaccination or mask and 5 people passed indoors and 10 people passed outdoors,
69
there is a close correlation between the change in number of active cases, the risk of infection, risk of hospitalization, and risk
70
of death. Using the same inputs and changing the user’s age to 65 years, we see the same risk of infection, but a much elevated
71
risk of hospitalization and death (Figure 1B).72
Change in Risk of Infection with Vaccination and Dosage73
To illustrate the effect of vaccination, we take again the example of the 30 year old male with no chronic health conditions
74
in Massachusetts, USA, and add different doses of the Moderna Vaccine. In the backend of this calculation, prevalence of
75
variants of concern in a given region from the GISAID
14
dataset are taken into account (Figure 3). As expected we observe
76
reduced risk of infection upon vaccination with the first dose of the Moderna vaccine and further risk reduction with dose 2
77
(Figure 1C).78
Change in Risk of Infection with Mask Type79
To demonstrate risk stratification by face mask type, we use the same example of a 30 year old male with no past chronic
80
illness, in Massachusetts, USA, unvaccinated, and estimate infection risk when wearing no mask, a surgical mask, or an N95
81
respirator. We observe that there is significant risk reduction upon wearing mask which was in line with several past studies
82
conducted
1516
. Depending upon the fitted filtration efficacy of the mask, there was a large reduction in risk of COVID-19
83
infection for both outdoor and indoor activities (Figure 1D).84
Figure 1. For a 30 year old male with no chronic illness and 10 people passed outdoors and 5 people passed indoors, we
calculate for Massachusetts, USA the (A) Change in different risk factors with change in active cases for Unvaccinated 30 year
old male with no past chronic illness and not wearing any mask, (B) Change in risk factor in the same scenario in a 65 yr old
(C) with different doses of vaccine, and (D) with two types of face masks used.
Ease of Usage and Streamlined Estimation of COVID-19 Risk85
By identifying the significant inputs required for calculating several risk factors pertaining to COVID-19 and minimizing
86
the user-inputs, we were able to substantially streamline the process of risk calculation so that it could be easily implemented87
as a day-to-day usage tool. The user is able to get acquainted with the risk of infection, hospitalization and death in under 2
88
minutes on an average across 5 trial runs.89
Discussion90
We demonstrate that the COVrisk gamified activity risk calculator is an accurate and useful tool for the public and policy
91
makers. The risk estimation system is both more comprehensive and much simpler compared to existing alternatives. Due to its
92
streamlined interface, the user is enabled to get acquainted with the risk of infection, hospitalization and death in just a few
93
minutes. This is complimented by high accuracy of risk prediction with comprehensive inputs including age, gender, health
94
3/9
conditions, active cases, vaccination (type and dose), use of and type of face masks, and number of people in close contact with
95
outdoors and indoors.96
There are some factors that we don’t yet take into account in our calculations that could influence the accuracy of predictions
97
or be useful to users. The calculator does not take into account the behavior of others with regards to vaccination or mask usage.
98
Instead, we err on the side of caution and assume others are unmasked and unvaccinated, which could somewhat overestimate
99
risk in situations of high mask or vaccination compliance. In particular it would be useful to include the behavior of others
100
in future calculations for purposes of planning private gatherings or when going to establishments with proof of vaccination
101
requirements. The calculator also does not take into account natural immunity. Further, we do not yet account for waning
102
immunity from vaccines or natural infection or how variants of concern impact rate of decline. We do plan to add these inputs
103
for the next release. We also have not included risk of myocarditis with vaccination or infection, which is an outcome that many
104
people are concerned about and have asked us to include as a feature. Including this as an output may be useful to demonstrate
105
that risk of myocarditis is greater with COVID-19 infection than with vaccination. Lastly, we do not yet account for the newest
106
Variant of Concern, Omicron. As new variants emerge we plan to update COVrisk when reliable information is fully available.
107
How Omicron will effect transmission rates, hospitalization rates, death rates, or immune escape from vaccines has not yet
108
been confirmed.109
We hope that this calculator will empower the public to live their lives safely and that a beneficial side effect of use of the110
calculator will be education about risk reduction, which ultimately could result in reduction of COVID-19 infections. Future
111
directions will include quantifying the impact of COVrisk on COVID-19 community transmission levels. Gamifying risk
112
reduction could prove to be a useful public health strategy for this and future Pandemics.113
Methods114
Figure 2. Working web based prototype for the risk calculator17
Dataset usage115
As a part of initial pre-processing of the confirmed cases obtained through Johns Hopkins dataset
18
, we identify the number
116
of active cases and computes a 14 day aggregate. Furthermore, we determine a range of 14 day aggregate active cases by taking
117
the upper limit as the product of the 14 day aggregate active cases and the ratio between the confirmed cases obtained from
118
Facebook’s COVID-19 Trends and Impact Survey
19
and Johns hopkins dataset
18
. As a part of developing the system, we used
119
datasets from several different sources namely the Johns Hopkins University dataset
18
, used for calculating the number of
120
active cases by taking the number of confirmed cases for a given day and subtracting it with the number of confirmed cases of
121
the preceding day. Moreover, taking a 14 day aggregate for them. We use the GSAID dataset
14
for calculating the number of
122
different variant cases present in a particular region. In order to extract the number of active cases we subtracted the number of
123
confirmed cases from the previous day and took a 14 day aggregate of it. The dataset is regularly updated on our database server
124
4/9
from where the data for the number of active cases is fetched. We perform Gaussian smoothing over the variants dataset and
125
upload them to our amazon s3 bucket where it is updated daily. We further use the country and region-wise population dataset
126
to estimate number of active cases per unit population to identify the COVID density in the given region. To further accurate the
127
confirmed cases dataset, we calculated the ratio between the number of confirmed cases by Facebook surveys
19
and officially
128
reported confirmed cases
18
to calculated the variance in number of confirmed cases. We use the ratio as a multiplication factor
129
to calculate the upper limit for the active cases and calculate the maximum range for risk of infection.130
In addition to these datasets that were pre-existing, we used information provided by several different research studies and
131
articles in order to create custom datasets for the protection mask’s fitted filtration efficacy (FFE), efficacy of vaccine against
132
different variants of virus, indoor and outdoor risk, risk of hospitalization and death with their dependency on age, gender,
133
presence of variants and past chronic illness.134
We use all of these datasets and take into account the number of people the user passes outdoors while travelling to the
135
location and the number of people they are with at the destination (which can be indoors or outdoors) to estimate several risks
136
associated with COVID.137
Mask Type
Vaccine Type &
Dose
People passed
(Indoors)
Date
People passed
(Outdoors)
Johns Hopkins &
Oxford University
confirmed cases
dataset
GISAID variants
dataset
COVID19
Vaccine
Efficacy dataset
Mask Fitted
Filtration
Efficacy dataset
Region-wise
Population
dataset
Inputs
Active Cases for
a specific region
Average active
Variants Cases
14 day aggregated active cases
(ncas=
        Active cases per
unit population*
Region
If region matches
Vaccine Risk
Reduction
(rv)*
Risk factor
Current Active
Cases (rac)*
1-Alpha variant
efficacy
Mask Risk Reduction
(rm)
(1 mask FFE)
Average variant cases= 
   
If date, matches,
1-Beta variant
efficacy
1-Gamma variant
efficacy
     
 
Risk Factor (ro)
(Outdoors)*
Risk Factor (ri)
(Indoors)*
Risk Factor*
(Cumulative)
    
    
no
ni
 
1-Delta variant
efficacy
Country
Risk factor (No mask) = 1
No vaccine, efficacy reduction = 1
1- Normal
efficacy
Age, Gender,
Variant, Chronic
Illness data
Risk of
Hospitalization* Risk of Death*
Ratio
(Facebook
Surveys
Cases/Report
ed cases)
*: Consists of a range (lower limits & upper limit)
Risk of
Hospitalization
& death Dataset
Inputs
 

 

Figure 3. A.Workflow for the Risk Calculator
Pipeline and Workflow138
We then identify the new cases from the confirmed cases by subtracting the confirmed cases for a given day with confirmed
139
cases for the previous day. We then identify the number of new cases by taking a 14 day sum for number of active cases(
nac
).
140
Upon providing the input for region/state within the country and present day date, we identify the 14 day sum of active case
141
for a particular previous day and region/state. We then take into account the region and date to identify the variants and then
142
identify the average number of variants cases over a period of 31 days. We use the ratio between the confirmed cases reported
143
through Facebook surveys
19
and the officially reported confirmed cases and multiply them with the confirmed cases reported by
144
Johns Hopkins University dataset18 in order to estimate the upper limit for the number of reported cases.145
We calculated the reduced risk of infection for variants of COVID by subtracting one from the efficacy of vaccines as well
146
as fitted filtration efficacy of masks using the same method. For vaccines, we use the lower limit and upper limit of the efficacy
147
in order to estimate and multiply it with the higher limit and lower limit of confirmed cases respectively and estimate the range
148
of risk of infection. We then calculate the risk of variants by taking into account the regional data to identify the number of
149
variants present and multiply them with reduced risk after vaccination to identify the added risk of variants (rv). Furthermore,150
we multiply the number of people passed by indoors (ni), outdoors (no), risk due to active cases per unit population (rac), risk151
reduction due to vaccination (
rv
) and reduced risk due to mask (
rm
) to calculate indoor (
ri
) and outdoor risk (
ro
) range of
152
infection. The sum of which helps us to estimate the range of cumulative risk of infection.153
We further estimate the risk of hospitalization (
rh
) and risk of death (
rd
) by using factors related to age (
fa
and
fa1
),
154
gender(
fg
and
fg1
), past chronic illness(
fc
and
fc1
) and type of variants (
fv
and
fv1
) and multiply the range of risk factors with
155
5/9
the upper and lower limits of the cumulative risk of infection. Finally we use it to compute the range of risk of hospitalization
156
and risk of death.157
Testing158
We tested the risk calculator by creating a webpage using shinyapps and R, where all this estimation system could be
159
accessed by anyone in order to identify their personal risk of infection hospitalization and death when carrying out day to day
160
activities. In order to make it more accessible the webpage would be later on converted in the form of a smartphone application
161
which could further increase the accessibility of the same to the general public and enable smooth implementation of the system
162
in the daily life of individuals. The representation of the webpage can be seen Fig. 2.163
References164
1. The microcovid project. https://www.microcovid.org/paper/all. Accessed: 2021-10-29.165
2.
for Disease Control, C. & Prevention. Covid data tracker. https://covid.cdc.gov/covid-data- tracker/#county-view&list_
166
select_map_data_parent=Risk&map-metrics-cv-comm-transmission=community_transmission_level&list_select_map_
167
data_metro=all (2019).168
3. Covid-19 event risk assessment planning tool. https://covid19risk.biosci.gatech.edu/. Accessed: 2021-10-29.169
4. 19 and me: Covid-19 risk score calculator. https://19andme.covid19.mathematica.org/. Accessed: 2021-10-29.170
5.
Jin, J. et al. Assessment of individual-and community-level risks for covid-19 mortality in the us and implications for
171
vaccine distribution. medRxiv (2020).172
6. Mycovidrisk. https://mycovidrisk.app/. Accessed: 2021-10-29.173
7.
Rowe, B., Canosa, A., Drouffe, J. & Mitchell, J. Simple quantitative assessment of the outdoor versus indoor airborne
174
transmission of viruses and covid-19. medRxiv DOI: 10.1101/2020.12.30.20249058 (2021). https://www.medrxiv.org/
175
content/early/2021/01/04/2020.12.30.20249058.1.full.pdf.176
8.
Qcovid risk calculator. https://qcovid.org/Home/AcademicLicence?licencedUrl=%2FCalculation. Accessed: 2021-10-29.
177
9. Covid-19 risk calculator. https://covid-19.forhealth.org/covid-19-transmission-calculator/. Accessed: 2021-10-29.178
10. Covid19 exposure risk calculator. https://covidtracker.fr/covid19-risk- calculator/. Accessed: 2021-10-29.179
11.
Lelieveld, J. et al. Model calculations of aerosol transmission and infection risk of covid-19 in indoor environments. Int. J.
180
Environ. Res. Public Heal. 17, DOI: 10.3390/ijerph17218114 (2020).181
12.
Bazant, M. Z. & Bush, J. W. M. A guideline to limit indoor airborne transmission of covid-19. Proc. Natl. Acad. Sci.
118
,
182
DOI: 10.1073/pnas.2018995118 (2021). https://www.pnas.org/content/118/17/e2018995118.full.pdf.183
13. Covid-19 dashboard. https://www.northwestern.edu/coronavirus-covid-19-updates/university-status/dashboard/ (2021).184
14. Gisaid tracking of variants. https://www.gisaid.org/hcov19-variants/. Accessed: 2021-11-20.185
15.
for Disease Control, C. & Prevention. Science brief: Community use of masks to control the spread of sars-cov-2.
186
https://www.cdc.gov/coronavirus/2019-ncov/science/science-briefs/masking-science-sars-cov2.html (2021).187
16.
Agency, U. E. P. Epa researchers test effectiveness of face masks, disinfection methods against covid-19. https://www.epa.
188
gov/sciencematters/epa-researchers-test-effectiveness-face-masks-disinfection-methods-against-covid-19 (2021).189
17. Covid19 activity risk calculator webpage. https://realsciencecommunity.shinyapps.io/riskcalculator/ (2021).190
18. Gardner, L. Covid-19 data repository. https://github.com/CSSEGISandData/COVID-19 (2019).191
19.
Covid-19 trends and impact survey. https://dataforgood.facebook.com/dfg/tools/covid-19-trends-and-impact-survey
192
(2021).193
Author contributions statement194
Shrey Natraj created the code for the backend of the calculator, made the paper figures, and wrote a large part of the paper
195
draft. Nathan Yap interpreted Shrey’s code and created the user interface in R Shiny. Agrima Seth helped to draft the paper
196
and the text for the UI for the web app. Malhar Bhide researched the affects of interventions and created many of the paper
197
tables. Leila Orzag contributed to survey and data collection, and development of manuscript. Pawan Nandakishore created
198
the structure for pulling code daily into S3 buckets and formatting data frames. Nina Rescic created the variants data frame
199
and performed smoothing. Shanice Hudson contributed to survey and data collection, and development of manuscript. Carlos
200
6/9
Baquero, Davide Frey, and Antonio Fernandez Anta developed methods used to estimate real active case numbers using surveys
201
and provided valuable feedback on the manuscript. Christin Glorioso thought of the idea for the app, led the project, and helped
202
to draft and edit the manuscript and web app UI.203
Supplementary Materials204
Efficacy of vaccines205
Our major focus for this project was to use accurate data in order to estimate the risk reduction upon vaccination from differ-
206
ent versions of COVID vaccine. For the given purpose we used the reference from (https://yourlocalepidemiologist.substack.com/p/vaccine-
207
table-update-may-5-2021) in order to generate a table consisting of different vaccine efficacy from different manufacturing
208
companies and upon different dosage. This table (see Table 3) was used in order to identify the risk reduction for different
209
variants of COVID upon vaccination to estimate the cumulative risk of infection.210
Figure 4.
For an unvaccinated unmasked 30 year old male with no chronic illness and 10 people passed outdoors and 5 people
passed indoors, we calculate for Delhi, India the A. Change in different risk factors with change in active cases, D. We calculate
different risks for a 65 year old unvaccinated unmasked male with 10 outdoor and 5 indoor people passed C. Change in risk
factor with different doses of vaccine, D. Change in risk factor with whether mask is worn or not.
Mask Fitted Filtration Efficacy211
As can be observed in Fig. 1, masks play a significant role in risk reduction of COVID. A mask with higher fitted filtration effi-
212
cacy (FFE) reduces the risk of COVID infection significantly on a day-to-day basis. We used the (https://www.epa.gov/sciencematters/epa-
213
researchers-test-effectiveness-face-masks-disinfection-methods-against-covid-19) as reference to create a dataset represented in
214
Fig. 2. This was used as one of the factor taken into account when calculating the cumulative risk of infection.215
Risk of Hospitalization and Death216
In order to estimate the risk of hospitalization and risk of death, we took into account the age, gender, past chronic illness
217
and presence of a specific variant in the given region which increased or decreased the risk of hospitalization and death. The
218
estimated risk of hospitalization and death are calculated using the cumulative risk of infection and multiplying it with factor
219
influenced with age, gender, past chronic illness and specific COVID variant presence (see Table 4).220
Risk of infection in the case of Delhi, India221
As a part of the supplementary materials we have also included another test similar to the one indicated in Fig. 1. In the
222
second example, we calculated the change in risk of infection, hospitalization and death with change in aggregated 14 day
223
active cases for a 30 year old and 65 year old male living in Delhi, India, with no chronic illness, no vaccination, no mask,
224
5 people passed indoors and 10 people passed outdoors (see Fig. 4 A. B.). We also conducted the same when a vaccination
225
of Astrazeneca Dose 1 and Dose 2 are taken in order to check the reduced risk of infection for the 30 year old male with no
226
chronic illness (see Fig. 4 C.) and reduction in risk of infection when wearing procedure mask and N95 respirator mask (see
227
Fig. 4 D.) in order to further validate the results observed for the Massachusetts, USA example.228
7/9
Mask Type FFE
2-layer woven nylon mask without nose bridge 0.447
2-layer woven nylon mask with nose bridge 0.563
2-layer woven nylon with nose bridge and filter insert 0.744
2-layer woven nylon with nose bridge washed 0.79
Cotton Bandana folded surgeon general style 0.49
Cotton Bandana folded bandit style 0.49
Single-layer woven polyester gaiter 0.378
Single-layer woven polyester mask with ties 0.393
Non-woven polypropylene mask with fixed ear loops 0.286
3-layer knitted cotton mask with ear loops 0.265
N95 respirator 0.984
Surgical mask with ties 0.715
Procedure mask with ear loops 0.385
Procedure mask with loops tied, corners tucked 0.603
Procedure mask with loops tied, corners tucked and ear guard 0.617
Procedure mask with Clawed hair clip 0.648
Procedure mask with fix-the-mask technique (rubber bands) 0.782
Procedure mask with Nylon hosiery sleeve 0.802
No Mask 0
Table 2. Mask Fitted Filtration Efficacy
Vaccine Normal Alpha Beta Gamma Delta
Pfizer (Dose 1) 0.800-0.910 0.490 0.360-0.375 0.360-0.370 0.330
Pfizer (Dose 2) 0.950 0.870-0.950 0.720-0.850 0.750-0.770 0.790-0.920
Moderna (Dose 1) 0.800-0.900 0.490 0.490-0.720 0.490-0.720 0.330-0.490
Moderna (Dose 2) 0.900-0.960 0.910-0.960 0.900-0.960 0.900-0.960 0.855-0.960
JJ (Dose 1) 0.690-0.770 0.770 0.520-0.570 0.510-0.680 0.490-0.780
Astrazeneca (Dose 1) 0.550-0.670 0.330-0.370 0.100-0.110 0.110-0.243 0.329
Astrazeneca (Dose 2) 0.820-0.850 0.660-0.740 0.220-0.700 0.220-0.486 0.598
Novavax (Dose 1) 0.904-0.910 0.000-0.863 0.000-0.486 0.000 0.000
No Vaccine 0.000 0.000 0.000 0.000 0.000
Table 3. Vaccine Efficacy based on Dosage and Manufacturing Company
8/9
Hospitalization Rate (%) Death Rate (%)
By Age Group a [1], [2]
0-17 years 0.8% 0.0015%
18-49 years 2.5% 0.07%
50-64 years 7.9% 0.7%
65+ years 23% 6%
All ages 5% 0.75%
By SARS-CoV-2 Variant- Fold Higher Risk Compared to Original Variants
Alpha (B.1.1.7 B.1.1.7 with E484K) 1.5 (1.5–1.6) c [3] 1.6 (1.4–1.7) [3]
Beta (B.1.351) Under Investigation Possibly increased [4, 7]
Gamma (P.1) Possibly Increased [4] 1.5 (1.2–1.9) e, f [8, 9]
Delta (B.1.617.2) 2.3 (1.9–3.0) d, e [5, 6] 2.4 (1.5–3.3) [6]
By Gender - Fold Higher Risk g [10] [12]
Male - 1.5-2.3
Female - 1
By Any Chronic Health Condition - Fold Higher Risk [11][12]
Diabetes, Heart Disease, Cancer, Lung disease, High Blood Pressure,
Immunocompromised, Asthma, Kidney Disease, Obesity,
Sickle Cell Anemia, HIV, Liver Disease.
2.5 1.2-6.9
Table 4. Hospitalization and Death Risk by Age, Variant, Gender, Health Condition
9/9
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
The role of aerosolized SARS-CoV-2 viruses in airborne transmission of COVID-19 has been debated. The aerosols are transmitted through breathing and vocalization by infectious subjects. Some authors state that this represents the dominant route of spreading, while others dismiss the option. Here we present an adjustable algorithm to estimate the infection risk for different indoor environments, constrained by published data of human aerosol emissions, SARS-CoV-2 viral loads, infective dose and other parameters. We evaluate typical indoor settings such as an office, a classroom, choir practice, and a reception/party. Our results suggest that aerosols from highly infective subjects can effectively transmit COVID-19 in indoor environments. This “highly infective” category represents approximately 20% of the patients who tested positive for SARS-CoV-2. We find that “super infective” subjects, representing the top 5–10% of subjects with a positive test, plus an unknown fraction of less—but still highly infective, high aerosol-emitting subjects—may cause COVID-19 clusters (>10 infections). In general, active room ventilation and the ubiquitous wearing of face masks (i.e., by all subjects) may reduce the individual infection risk by a factor of five to ten, similar to high-volume, high-efficiency particulate air (HEPA) filtering. A particularly effective mitigation measure is the use of high-quality masks, which can drastically reduce the indoor infection risk through aerosols.
Assessment of individual-and community-level risks for covid-19 mortality in the us and implications for 171 vaccine distribution
  • J Jin
Jin, J. et al. Assessment of individual-and community-level risks for covid-19 mortality in the us and implications for 171 vaccine distribution. medRxiv (2020).
Simple quantitative assessment of the outdoor versus indoor airborne 174 transmission of viruses and covid-19
  • B Rowe
  • A Canosa
  • J Drouffe
  • J Mitchell
Rowe, B., Canosa, A., Drouffe, J. & Mitchell, J. Simple quantitative assessment of the outdoor versus indoor airborne 174 transmission of viruses and covid-19. medRxiv DOI: 10.1101/2020.12.30.20249058 (2021). https://www.medrxiv.org/ 175 content/early/2021/01/04/2020.12.30.20249058.1.full.pdf.
A guideline to limit indoor airborne transmission of covid-19
  • M Z Bazant
  • J W M Bush
Bazant, M. Z. & Bush, J. W. M. A guideline to limit indoor airborne transmission of covid-19. Proc. Natl. Acad. Sci. 118, 182 DOI: 10.1073/pnas.2018995118 (2021). https://www.pnas.org/content/118/17/e2018995118.full.pdf.
Epa researchers test effectiveness of face masks, disinfection methods against covid-19
  • U E Agency
16. Agency, U. E. P. Epa researchers test effectiveness of face masks, disinfection methods against covid-19. https://www.epa.
Covid-19 data repository
  • L Gardner
Gardner, L. Covid-19 data repository. https://github.com/CSSEGISandData/COVID-19 (2019).