Coronavirus disease (COVID-19) technical guidance by WHO Household transmission investigation protocol | Page 23

Case-fatality ratio
The number of deaths in
household caused by
COVID-19 in cases,
compared to the total
number of cases with
COVID-19 in the household.
(Proportion of COVID-19
cases who die).
Dead/alive status
and case
confirmation.
•
Form 2, 3, 4, 5
•
Genomic data,
including phylogenetic
analysis.
Laboratory data

Form 2, 3, 4, 5
•
Basic reproduction
number R 0
A measure of the number of
infections produced, on
average, by an infected
individual in the early stages
of the epidemic, when
virtually all contacts are
susceptible. Note that it can
be assumed that there will
be very little to no immunity
to COVID-19.
(Average number of
infections/diseases arising
from one infection).
Reminder: R 0 – everyone is
susceptible and there is no
control; the maximum value
that R can take is equal to
the transmission potential.
-23-
Laboratory data,
dates of contact,
symptoms in
contacts.
•
Form 1A: Q5
Form 2, 3, 4, 5
Symptom diary
•
A large number of cases
will probably be needed
before a significant
number of deaths are
seen, in order to allow
reliable estimates
through household
investigations (also
follow-up may end
before deaths due to
secondary infections
can be observed).
More likely to be an
overestimate in this
investigation, owing to
reporting/selection bias
of the initial cases.
An alternate means to
estimate the basic
reproduction number,
from comparing the
relatedness of strains
between cases and their
close contacts and
confirming transmission
between individuals.
The data may supplement
other transmission data to
inform transmission
parameter estimates,
although these data are
likely to be delayed
beyond the initial public
health response phase.
Can be calculated using
different approaches;
identifying clusters and
cluster size (using epi
methods and potentially
genetic information to
identify how many
secondary cases are
occurring) and using the
epidemic curve and how
steep it is.
R 0 can be calculated using
multiple sources of
information: incident case
notifications, incident
hospitalizations by age (as
a potentially more stable
alternative), or genomic
data, all of which will be
taken together as an
estimate of
transmissibility.