Dan Jarosch asks about modeling the SARS epidemic.
Roy Anderson in the UK, probably the leading expert on dynamic
modeling of infectious diseases, is currently developing a SARS
model. I dont know what has been published. He previously
developed models for the BSE (mad cow) epidemic and has estimated how
many human cases of BSE-induced vCJD can be expected.
For analogous work on infectious diseases, see the terrific papers on
smallpox and anthrax bioterror attacks by Ed Kaplan, Larry Wein, and
David Craft. There are two:
Kaplan, E., Craft, D. and Wein, L. (2002) Emergency response to a
smallpox attack: The case for mass vaccination PNAS (Proc. Nat. Acad.
of Sciences), August 6, 2002, 99 (16), 10935-10940
Wein, L., Craft, D., and Kaplan, E., (2003) Emergency response to an
anthrax attack. PNAS April 1, 2003100(7) 4346-4351.
These papers provide models of infectious disease spread after
bioterror attack, and include structure for quarantine and different
vaccination strategies. The smallpox model, in particular, could
readily be adapted for SARS.
John Sterman
From: John Sterman <jsterman@MIT.EDU>
Work on the Spread of SARS
-
John Sterman
- Senior Member
- Posts: 117
- Joined: Fri Mar 29, 2002 3:39 am
-
Daniel Jarosch
- Newbie
- Posts: 1
- Joined: Fri Mar 29, 2002 3:39 am
Work on the Spread of SARS
Knowing that there have been plenty SD research streams on the diffusion of
epidemics, I was wondering what the experts on this list say to the current
newspaper filling epidemic? If anyone is having a SD look at SARS I would
appreciate her/his opinion? Is it media hysteria or justified caution?
Best regards from a sunny winter autumn day in Sydney
daniel
------------------------------------------------------
Daniel Jarosch
IBM Business Consulting Services
Darling Park Tower 2
201, Sussex Street
Sydney, NSW, 2001, Australia
phone: + 61 2 8266 2619
email: daniel.jarosch@au1.ibm.com
epidemics, I was wondering what the experts on this list say to the current
newspaper filling epidemic? If anyone is having a SD look at SARS I would
appreciate her/his opinion? Is it media hysteria or justified caution?
Best regards from a sunny winter autumn day in Sydney
daniel
------------------------------------------------------
Daniel Jarosch
IBM Business Consulting Services
Darling Park Tower 2
201, Sussex Street
Sydney, NSW, 2001, Australia
phone: + 61 2 8266 2619
email: daniel.jarosch@au1.ibm.com
Work on the Spread of SARS
A short answer to Daniel Janoschs question is: The SARS epidemic is
worthy of caution but not yet panic. Media hysteria has not been abated
by the way public health authorities have handled the ever-changing
information about the disease and its spread.
From my vantage point here in Quebec City, I would say that Torontos
handling of information about the progress of the disease is a
case-study in dynamic ineptitude. The authorities presented information
on the total number of "probable and suspected" cases encountered since
the beginning of the outbreak there on March 5. This number, of course,
only increases over time. Only in the last two or three days have some
reporters mentioned that of these cases (some 260), most never became
real SARS-infected people, in fact only a little over a 100 (so far)
were infected and required treatment. This number is also fated to
increase. What is now being reported is the number currently infected
(i.e. subtracting deaths and recoveries) and the number of new "probable
and suspected" cases, both of which are decreasing in the expected
stochastic fashion.
Of course what would be most helpful would be a day-by-day barchart of
new "probable and suspected" cases, of new released-as-uninfecteds, of
new recovereds and of new deaths with an accompanying piece-wise
linear curve of currently infected. If the outflow bars were stacked
to be directly comparable to the inflow bar, it would be much clearer
how the campaign is going.
The Toronto case, when compared to the Vancouver experience, shows that
chance plays a big part in actual outcomes, but when compared with the
Vietnamese and Sinagaporean cases, shows that there are fairly
straightforward policies to control the epidemic. Vancouver quarantined
its first case because he fit a profile (not specifally for SARS) that
the main hospital had been sensitized to; Toronto did not because the
first case did not fit the SARS profile enough and the suburban hospital
did not quarantine him, much less his contacts, rapidly enough. Pure
chance, failure of leadership, take your pick.
The success that Vietnam (and to some extent Singapore) have had has
been due to a very vigorous programme of restricting arrivals from hot
areas regardless of the means of transport and severe penalties for
breaking quarantine. Toronto has been free of community-sourced cases
for 21 days now and has had no hospital-sourced cases for about two
weeks. However, early quarantine efforts were not strict enough and
there were several cases of quarantined people returning to work early,
thereby disrupting work and adding to the pressure on the health system.
There is some confusion about whether the modest efforts to control
departures of possibly-infected people from the airport has not
prevented the export of SARS from Toronto to other countries. Most of
the exported cases that figured in the WHO travel advisory are
disputed vigorously by Health Canada.
One other feature of the Toronto experience is more worrying. The
percentage of fatalities is high; I have heard of numbers like 13% but I
think this is based on the number of infected not the total "probable
and suspected". The worldwide average is around 5%, although at the
beginning of the outbreak, before the Chinese data came in, it was
expected to be about 1% and this compared with the so-called
Spanish-flu pandemic after the First World War.
What are the lessons for system dynamicists? Learn how to present your
data clearly. As with any novel product, abysmal data quality will make
accurate modelling difficult (especially in the face of data-hiding). A
simple model could help to clarify the flow of events. A more complete
model could help to avoid some unintended consequences such as
medical-staff burn-out currently being experienced in Toronto. It would
be interesting and perhaps useful to compare the Toronto-SARS experience
with the foot-and-mouth disease outbreak in the UK a couple of years
ago. There might be interesting capacity-limited dynamic parallels to
explore in order to inform public policy better.
R. Joel Rahn
Ste-Foy, Quebec
jrahn@sympatico.ca
worthy of caution but not yet panic. Media hysteria has not been abated
by the way public health authorities have handled the ever-changing
information about the disease and its spread.
From my vantage point here in Quebec City, I would say that Torontos
handling of information about the progress of the disease is a
case-study in dynamic ineptitude. The authorities presented information
on the total number of "probable and suspected" cases encountered since
the beginning of the outbreak there on March 5. This number, of course,
only increases over time. Only in the last two or three days have some
reporters mentioned that of these cases (some 260), most never became
real SARS-infected people, in fact only a little over a 100 (so far)
were infected and required treatment. This number is also fated to
increase. What is now being reported is the number currently infected
(i.e. subtracting deaths and recoveries) and the number of new "probable
and suspected" cases, both of which are decreasing in the expected
stochastic fashion.
Of course what would be most helpful would be a day-by-day barchart of
new "probable and suspected" cases, of new released-as-uninfecteds, of
new recovereds and of new deaths with an accompanying piece-wise
linear curve of currently infected. If the outflow bars were stacked
to be directly comparable to the inflow bar, it would be much clearer
how the campaign is going.
The Toronto case, when compared to the Vancouver experience, shows that
chance plays a big part in actual outcomes, but when compared with the
Vietnamese and Sinagaporean cases, shows that there are fairly
straightforward policies to control the epidemic. Vancouver quarantined
its first case because he fit a profile (not specifally for SARS) that
the main hospital had been sensitized to; Toronto did not because the
first case did not fit the SARS profile enough and the suburban hospital
did not quarantine him, much less his contacts, rapidly enough. Pure
chance, failure of leadership, take your pick.
The success that Vietnam (and to some extent Singapore) have had has
been due to a very vigorous programme of restricting arrivals from hot
areas regardless of the means of transport and severe penalties for
breaking quarantine. Toronto has been free of community-sourced cases
for 21 days now and has had no hospital-sourced cases for about two
weeks. However, early quarantine efforts were not strict enough and
there were several cases of quarantined people returning to work early,
thereby disrupting work and adding to the pressure on the health system.
There is some confusion about whether the modest efforts to control
departures of possibly-infected people from the airport has not
prevented the export of SARS from Toronto to other countries. Most of
the exported cases that figured in the WHO travel advisory are
disputed vigorously by Health Canada.
One other feature of the Toronto experience is more worrying. The
percentage of fatalities is high; I have heard of numbers like 13% but I
think this is based on the number of infected not the total "probable
and suspected". The worldwide average is around 5%, although at the
beginning of the outbreak, before the Chinese data came in, it was
expected to be about 1% and this compared with the so-called
Spanish-flu pandemic after the First World War.
What are the lessons for system dynamicists? Learn how to present your
data clearly. As with any novel product, abysmal data quality will make
accurate modelling difficult (especially in the face of data-hiding). A
simple model could help to clarify the flow of events. A more complete
model could help to avoid some unintended consequences such as
medical-staff burn-out currently being experienced in Toronto. It would
be interesting and perhaps useful to compare the Toronto-SARS experience
with the foot-and-mouth disease outbreak in the UK a couple of years
ago. There might be interesting capacity-limited dynamic parallels to
explore in order to inform public policy better.
R. Joel Rahn
Ste-Foy, Quebec
jrahn@sympatico.ca