Public health in the 21st century - the open-source outbreak | Jennifer Gardy | TEDxTerryTalks

Dr. Jennifer Gardy is a postdoctoral fellow in the Robert Hancocks laboratory at UBCs Centre for Microbial Diseases and Immunity Research, where she works on methods for the systems-level analysis of the mammalian innate immune system. Jennifer is also involved in several science media activities, including hosting CBC Televisions new half-hour science show, Project X. Jennifer completed her PhD at Simon Fraser University, finishing her degree in four years without ever having to put in late nights at the lab, and while simultaneously pursuing her interest in media activities! She was also awarded the Governor Generals Gold Medal as the top doctoral student.

Filmed by Craig Ross at TEDx Terry talks 2009 (October 3rd, 2009). Video edited by David Ng.

This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at http://ted.com/tedx

Thank you, Dave.

Can everybody hear me?

I can hear myself, excellent.

As the alumni speaker,

I thought I should post
a picture of myself

in my UBC undergraduate days
for your reference.

So there's me in 1999
as the radical beer faction candidate

for the AMS Vice President
External position.

I'm pictured there with my right-hand man,

the dictatorial sock puppet,
Generalissimo,

explaining our plan to secede UBC
from the rest of Vancouver.

It just goes to show

that you can drink a lot of beer
and be completely ridiculous

as an undergraduate,

and still become a doctor

and be gainfully employed
in the field of public health,

as I am now,

at the British Columbia
Centre for Disease Control.

Our executive director, Bob Brennaman,
is here in the audience today.

And I thought today

I'd talk about public health
in the 21st century

and the concept
of the open-source outbreak.

Dave Ng and I go way back.

He's bought me many beers over the years,
and consequently, I owe him many favors,

so when he asked me to come be
the Alumni speaker for this event,

I said, "Sure, no problem."

And he said, "Can you talk
about swine flu?"

And I said, "Absolutely.
I can fill 15 minutes of pig flu.

No problem."

But then I got stuck,

and I really didn't know
what I wanted to say about pig flu.

I mean, this is an absolutely
extraordinary event,

this TEDxTerryTalk series.

There are extraordinary speakers,
and this is an extraordinary audience,

and it really did demand
an extraordinary topic.

So I couldn't really stand up here
and give you a pedestrian lecture

for 15 minutes

on the importance of washing your hands,

which incidentally,
is extremely important.

And I didn't want to get
too scientific or too technical

and do something like

the evolutionary origins
of pandemic influenza,

which is interesting
but can be summarized very quickly -

and that pandemic influenza
comes from pigs

and it comes from birds.

So I needed a topic
that instead, you know,

was really big, was really meaningful,

that you would take away something from it
at the end of the day,

and that really span multiple disciplines.

So I thought and I thought and I thought,

and it wasn't really until Wednesday

that I came up with the topic
that I actually wanted to focus on,

a TED-worthy topic.

And the TED-worthy topic
came about on Wednesday, as I said,

and we were having a research symposium
at the Centre for Disease Control,

and one of our doctors, Richard Lester,

was talking about a project
he had led in Kenya

that harnessed that country's infatuation
with the mobile phone, the cell phone,

to ensure that HIV patients were complying
with taking their HIV medications

and that they weren't experiencing
any side-effects.

This was basically an SMS,
or a text message,

that went out from the clinic
to the patients every week,

and they would text a message back

that they were taking their medications
and that they were doing all right.

So Richard's talk
really brought up two things.

It brought up that there's a lot
of new technology that's out there,

and there's a lot of new attitudes
towards technology,

and that public health in the 21st century

really has to capitalize
on this new technology

and on these new attitudes

to become, basically, "public health 2.0,"

a new model for delivering
public health services.

So in listening to Richard's presentation,
I was reminded of something,

very similar to this thought
of this new model of delivery,

that I had written about H1N1 back in May.

So in my other life,
when I'm not doing CDC related things,

I blog for The Globe and Mail's
globecampus.ca website as "Nerd Girl."

It's a lovely photo of me
with very good lighting.

I don't normally look that good.
(Laughter)

Back in May,

when swine flu first broke
and appeared on people's radar,

I wrote a post about some of the things

that as a scientist,
I found really fascinating

about the current outbreak,

and one of those things

was the idea that this H1N1 outbreak

is really world's first
open-source outbreak.

And what I meant by that
was that data about this virus

was being generated in real time

and it was being shared
with researchers around the world freely,

with no restrictions,

and the community had really come together
as kind of a hive mind

to solve this problem
of pandemic influenza.

So reminded of this idea that I had
by Richard's talk,

I figured that this would be
my suitably TED-worthy topic.

So what I'm going to talk about today
is this idea of the open-source outbreak,

or outbreak 2.0.

You know, what are the factors
that have enabled it?

And how did it really unfold with H1N1?

And what might we see,
as far as outbreak 2.0,

for the next outbreak?

So I figure I'll give you
my take-home point right now,

no sense in waiting till the end,

and that is basically
that advances in technology,

particularly around genome sequencing

and just, you know, bioinformatics
and come from original infrastructure,

have really contributed to this,

as has a sea change
in scientists' attitudes

towards becoming more open access oriented
and more collaborative.

And really, those two things together

have brought us to the brink
of this new era in public health

where our response
to emerging infectious diseases

and emerging threats

is much more rapid
than it's ever been in the past

and much more collaborative
than it's ever been.

So we'll hop in our awesome time machine

and go back with Doc
to the heady carefree days of March 2009,

when swine flu wasn't nearly
a blip on anybody's radar.

I've got a calendar here.

There's March, April, May, June.

And mid-March,

that public health officials in Mexico
started picking up increased flu activity,

and this increased throughout
late March and early April.

And on April 15,

the first two cases of what would
ultimately be recognized as H1N1

were picked up in California,

and the screening program in California
kind of triggered a warning bell

because these were two cases

that didn't resemble
any previous human influenza

that we had seen before.

So the Californian officials
called in the CDC,

and about six days later,

the CDC was able to confirm

that this was a novel
swine-origin influenza virus.

So as soon as they issued this advisory,

public health officials,
basically the world over,

started looking for this virus
in their own population,

and sure enough, they found it.

They found it here in Canada.

They found it elsewhere in the States.

They found it in Europe.

And by the time the 29th rolled around,
just over a week later,

it had basically been found everywhere.

You know, over that week,

the number of cases
was pretty much doubling every day.

So on the 29th,

the World Health Organization
raised the pandemic threat level to five,

meaning that a pandemic was imminent.

And as we all know,

the number of cases steadily increased
throughout April and May,

and finally on June 11,

a pandemic was officially declared
by the World Health Organization.

In Canada, as of June 11,
when the pandemic was declared,

at that point, we had about 5,000
laboratory confirmed cases of swine flu.

And using a figure of 20 to 30 cases
out there in the wild

for every lab confirmed case,

that meant we were looking
at about 100 to 150 thousand people

that were infected
at the beginning of June.

So where does the idea of open-source
outbreak fit in with all of this?

Well, if we go back
to our calendar a second -

I love this shot of the sneeze,
it's so disgusting,

it's the perfect public health picture -

as people were sharing germs

and this pandemic
was spreading around the world,

so too were researchers sharing data.

And if we go back to the calendar,
we can see that on April 25,

just, like, four days after the CDC
confirmed this novel infectious agent,

we already had the first
genome sequence for swine flu,

the blueprint for making
the swine flu virus.

And the very next day -
this is so cool -

a group of international researchers,

there's 13 people representing
eight institutions across four countries,

came together

and decided that they were going
to take this genome sequence data

and apply molecular
evolutionary analysis techniques to it

to figure out where this bug had come from
and how it was evolving.

So they created a wiki site.

It was basically an open lab notebook
that was open to everybody in the world.

They could check out the analysis
that this group was doing,

and they worked 24/7
on analyzing the genome sequence data

and figuring out
what was going on with swine flu.

As a matter of fact,

that very same day, that group,
working together as a team,

had figured out the origins
of H1N1 influenza.

They realized that this virus
was made up of six parts of a pig virus

that had been circulating
in North America for a while

and two parts of a pig virus

that had been circulating
in Europe and Asia.

Only five days after the first
genome sequence was made available,

we had the first paper,
official science research paper,

published on H1N1.

This was again about
the origins of the virus.

It was a different group
from the wiki group,

but they arrived at the same conclusions.

And this is absolutely remarkable,

to go from a genome sequence
and five days later,

having a peer-reviewed and published paper
using that genome sequence.

It's the fastest turnaround
we've ever seen in science.

By about a week and a half later,

you had genomic data available
for about 69 different viral isolates,

and the wiki group used this
to do a molecular clock analysis

where they calculated that the virus
would have entered the human population

probably towards late 2008 or early 2009.

And this molecular clock analysis

was included in, really,
the first major paper about H1N1

that was submitted
to the journal Science on May 5,

and in only six days,

it had been peer-reviewed, accepted,
and made available online -

again, remarkable turnaround.

And a month after that,

by the time the pandemic
was declared on June 11,

you had over 250 papers available on H1N1.

If we go back and compare this to SARS,
which was our last big outbreak,

you'll see just how quickly we've managed
to improve our response times.

So SARS - we thought
we were pretty fast with SARS.

If you take day zero as the day
that we isolated the SARS coronavirus,

it took us 19 days
to get to a viral genome.

And I will give a gratuitous plug here.

That viral genome was done
at the BC Centre for Disease Control

with researchers
from the Genome Sciences Centre

in 19 days.

I can't take any credit.

Applause to Bob though.

Good job, Bob. (Laughs)

(Applause)

It's the reason we're all still here
and not dead of respiratory illness.

(Laughter)

In 19 days of H1N1,

we had over 100 viral genomes,

we knew where and when and how
this virus had emerged,

there were multiple papers

that had already been published
or were in press,

and we were already growing up
a seed strain for use in the vaccine.

So, you know, we thought
we were fast in 2003,

and we're really fast now.

So how is this possible?

How are we so fast at responding to this?

It really boils down to two things:

technological advances

and a shift in the attitude
of the science community.

Technology-wise,

it's never been easier, cheaper, or faster
to sequence a genome.

Back when the Human
Genome Project started in 1990,

it took 10 years to get to a draft version
of one human genome,

took another three years to complete it -

though many would argue
it's still not actually complete -

cost three billion dollars;

and took hundreds,
if not thousands, of people.

Fast-forward to spring of this year.

It took this guy four weeks,
48,000 dollars, and two of his friends

to sequence his own genome
in its entirety.

The advances in this field
had been absolutely remarkable,

and they just keep coming.

I mean, it's getting to the point
where probably within the next two years,

we're going to have a thousand-dollar
human genome available within a week.

Similarly, it's never been
easier, cheaper, or faster

to push data around the world.

We've got things like email,

collaborative workspaces
like Google Docs and Wiki,

and we've just got the network
infrastructure to support all this.

I mean, when I first got on the internet -

and I won't tell you when that was,

it was a long time ago -

I would hit the "Get messages" button
on my email program,

I would go downstairs and make a sandwich,

and then I would hope that by the time
I finished making that sandwich

and came back upstairs,

my six email messages
would have downloaded,

and sometimes they hadn't.

Similarly,

along with these technological advances,

we've had real advances
in scientists' attitudes.

And you're seeing the ranks
of academia and research

populated by people of my generation.

And for lack of a better word,
I'll call us the file-sharing generation.

So we were people that grew up

trading our Commodore 64 games
back and forth with each other -

totally dating myself.

I look young,

but I'm older than you'd think.

(Laughter)

We trade our Commodore 64 games,

we go on BBS's,

and, you know, we have the first people

to use Napster and create
Friendster accounts

and things like that.

And this really engendered
in our generation

this belief that information
and, particularly, knowledge

wants to be free and should be free

and that there should be
no barriers to accessing it.

And this has really manifested itself
in research as an open access

and, really, the adoption
of an open access mentality.

Another gratuitous plug is
that it's Open Access Week in two weeks.

UBC's got a lot
of excellent events planned,

so do check them out.

Back when I was starting graduate school,

PLOS, the Public Library of Science,
and Biomed Central -

both open access publishers -

these were just fledgling concepts,
they didn't exist yet,

but now, just really about a decade later,

these have become two of the biggest
publishing forces in science

and published some of the most
impactful journals that are out there,

and in fact, a survey
that was done earlier this year

says that 85% of scientists
support open access.

So in addition to the spirit
of wanting information to be free,

scientists these days
are also really big on collaboration.

We appreciate that, you know,
two heads are better than one,

but 20 heads are better than 2,

especially when those 20 heads
all come from different disciplines

and bring different skill sets,

and I think this is really
clearly illustrated

when you look at how authorship of papers
has changed over the last 50 years.

So I have on the left
a screenshot of Nature,

taken from an issue
published exactly 50 years ago today,

and you can see

that there's two articles on this page
whose authorship you can see.

They're both single-author papers,

and it's kind of hard to see
the references,

but they all refer to single-
or dual-author papers.

On the other hand,

you've got a paper taken from Nature,

the issue that came out on Thursday,

and this is much more typical
of what you're seeing these days,

and this is a paper with 36 authors.

So we've really embraced
collaboration and team science.

Science has moved away
from the domain of the individual

and moved into the domain
of the community.

So how is all this, you know, Genome,
Sciences, technology, data infrastructure,

and changes in attitudes

impacted our response to H1N1?

Thanks to all the data

that's being generated
and shared in real time across the world,

we know that literally as of this morning,

that the virus is still susceptible
to all the antivirals that are out there.

We know that it hasn't drifted
from the vaccine strain,

it hasn't evolved away from that,

so that the vaccine will still offer
protection when it's available.

And we know that this basically
entirely displaced seasonal influenza

and it's the only flu
that's circulating around.

What's really interesting
is that in any outbreak situation,

it's always a race

between the spread of information
and the spread of the virus.

And in the past,

the virus has always won that race.

It has always gotten a lot faster
than new info has.

But this is the first time, really,

where the spread of information
has caught up to the spread of the virus.

And I think in future outbreaks,

you're actually going to see information
travel faster than the bug can.

So what about the next outbreak?

We don't know what it's going to be,

but we know how are we
going to deal with it.

I think we are at the beginning
of an era of genome surveillance

for outbreak identification.

The way we've looked
for outbreaks in the past,

first, it was looking for spikes
in lab test volumes

for particular pathogens,

then we moved on to a system

where we look for spikes

in patient visits to clinicians
for particular complaints.

But where we think we're going now
is this idea of genome surveillance

and, basically, setting up
a sentinel system

to sequence the genomes
of viruses and bacteria

that are circulating in a population

so that we can pick up on threats

before they register as spikes
on the lab or the clinic's radar.

So the idea is that we'd go in

and just sample a population

and all of its viruses
and all of its bacteria

just to see what's out there

well in advance of it
actually causing disease.

If you go back to that molecular
clock analysis that the wiki group did,

you can see that swine flu was circulating
for months in the human population

before the first cases were reported,

and it was probably circulating in pigs
for months or even years before that.

So if we had been sequencing
influenza genomes

as part of a sentinel program
in humans and in pigs,

we might have found this really early
and have been able to nip it in the bud.

Now, how exactly do you go about

sequencing all the viruses
and all the bacteria

that are circulating in a population
at any given time?

Well, the answer is incredibly disgusting

but unique.

That's what I like to call
"sewage-nomics,"

which is a very catchy term.

And that essentially involves
sequencing people's poop -

what is in people's poop?

Because any pathogen
that enters a population

is necessarily going to manifest itself
in what exits from that population.

So if you could set up
some sort of sentinel system

where you're screening
for emerging infectious diseases

in something like sewage,

that's really capturing the microbial
health of an entire population,

you might be able
to pick up on these things

well before they register
as actual disease.

It's incredibly disgusting
and a gonzo idea,

but there's a lot of people
that are thinking about this,

and I think we're really not too far off
from a system like this.

So, you know, stay in school,
go to med school,

and you can work in poop too
in the future.

(Laughter)

Yay.

So that's basically all I wanted to say.

I wanted to leave you there
with three pieces of advice

around the sort of public health 2.0,

that I think you can take away

and apply it to any field
that you're working in.

And the first bit of advice

is to embrace new technologies
and the opportunities that they provide.

The second bit of advice

is to embrace the spirit
of open access and collaboration.

And the third bit of advice is,
this winter, for God's sakes,

don't embrace anybody
that has a fever and/or cough,

(Laughter)

because they've got the swine flu.

(Laughs)

So with that, I'd like to say
oink, oink and thank you,

and she'll move on.

(Applause)