1
00:00:06,000 --> 00:00:09,240
I'm about to break
some life-or-death news to

2
00:00:09,240 --> 00:00:11,120
the people that are gathered here.

3
00:00:11,120 --> 00:00:14,360
Now, I may look or even feel
perfectly normal,

4
00:00:14,360 --> 00:00:18,160
but I'm actually carrying
a potentially lethal disease that

5
00:00:18,160 --> 00:00:23,120
I have been spreading around this
normally quiet, leafy little town.

6
00:00:27,800 --> 00:00:31,480
And I've got to say, they all look
remarkably happy about it.

7
00:00:32,760 --> 00:00:37,760
That's probably because it's part of
the biggest experiment of its kind.

8
00:00:38,240 --> 00:00:40,880
With the help of thousands
of volunteers, we are

9
00:00:40,880 --> 00:00:45,920
about to simulate the outbreak of
a fatal contagion throughout the UK.

10
00:00:47,640 --> 00:00:51,240
That might seem like a funny thing
to want to do, but if I can

11
00:00:51,240 --> 00:00:56,200
succeed, this will save lives when
- not if - a real pandemic hits.

12
00:01:00,480 --> 00:01:03,120
Our pandemic might be simulated,

13
00:01:03,120 --> 00:01:07,800
but global outbreaks of deadly flu
are a very real threat.

14
00:01:07,800 --> 00:01:10,880
In the last hundred years,
lethal strains of flu have

15
00:01:10,880 --> 00:01:15,680
swept around the planet four times,
killing untold millions.

16
00:01:16,840 --> 00:01:21,480
The UK Government puts pandemic flu
at the top of its risk register.

17
00:01:21,480 --> 00:01:24,040
The reason for that
is it WILL happen,

18
00:01:24,040 --> 00:01:25,920
there WILL be another pandemic.

19
00:01:25,920 --> 00:01:29,800
The next virulent outbreak
could happen at any time.

20
00:01:29,800 --> 00:01:32,200
To predict the devastation
it could cause,

21
00:01:32,200 --> 00:01:35,920
researchers need to understand
how much we move around

22
00:01:35,920 --> 00:01:40,480
and how many people we meet,
and for that, they need good data.

23
00:01:42,040 --> 00:01:45,840
So I've joined forces with
Ebola doctor Javid Abdelmoneim,

24
00:01:45,840 --> 00:01:49,640
and we've launched a smartphone app
that will reveal how a lethal

25
00:01:49,640 --> 00:01:52,680
pandemic virus could strike the UK.

26
00:01:56,160 --> 00:01:59,800
Now, the infection started here,
in Haslemere.

27
00:01:59,800 --> 00:02:01,920
The pandemic started last night,
at midnight.

28
00:02:01,920 --> 00:02:04,560
I probably have contaminated
quite a few people now.

29
00:02:04,560 --> 00:02:06,080
I think there's
a really strong chance

30
00:02:06,080 --> 00:02:08,640
that I have caught the pandemic.

31
00:02:08,640 --> 00:02:11,840
We will find out
how quickly it spreads.

32
00:02:11,840 --> 00:02:14,840
Are some parts of the UK
safer than others

33
00:02:14,840 --> 00:02:18,080
when it comes to the spread
of an infectious disease?

34
00:02:18,080 --> 00:02:21,560
Will it reach all parts of the UK?

35
00:02:22,800 --> 00:02:26,680
Our BBC Pandemic experiment
will find out.

36
00:02:26,680 --> 00:02:29,720
And we'll discover
how many of us could die.

37
00:02:30,880 --> 00:02:35,120
My word! That's an enormous number.
That IS an enormous number.

38
00:02:39,160 --> 00:02:42,160
The pandemic that we've started
here in Haslemere might well be

39
00:02:42,160 --> 00:02:44,960
simulated, but the information
that it gives us

40
00:02:44,960 --> 00:02:49,160
will help to answer very real
life-and-death questions.

41
00:02:49,160 --> 00:02:53,840
This will be the first ever
life-saving pandemic.

42
00:03:10,760 --> 00:03:13,480
It seems like there's
no shortage of things to

43
00:03:13,480 --> 00:03:17,480
worry about at the moment -
global terrorism, the devastating

44
00:03:17,480 --> 00:03:21,400
effects of climate change
and volatile world leaders.

45
00:03:21,400 --> 00:03:25,040
So I hate to break it to you,
but according to the UK Government's

46
00:03:25,040 --> 00:03:28,640
official list of risks to
this country, we actually have

47
00:03:28,640 --> 00:03:32,600
another, much bigger threat
that we also need to worry about.

48
00:03:35,520 --> 00:03:38,120
Now, this isn't just
a hypothetical disaster.

49
00:03:38,120 --> 00:03:40,040
This is something that has happened

50
00:03:40,040 --> 00:03:42,560
several times
in the past hundred years.

51
00:03:42,560 --> 00:03:46,200
It's also something that we seem to
have forgotten about.

52
00:03:46,200 --> 00:03:49,320
Could I ask you one question
on camera? Sure.

53
00:03:49,320 --> 00:03:52,720
Do you know what was the disaster
that killed the most

54
00:03:52,720 --> 00:03:55,760
people in the last hundred years?
No, I don't.

55
00:03:55,760 --> 00:03:59,440
Could you have a guess?
Oh, no, don't put me on the spot!

56
00:03:59,440 --> 00:04:00,800
HANNAH LAUGHS

57
00:04:00,800 --> 00:04:05,840
Ooh... That's a question.
I've no idea! I don't know.

58
00:04:06,960 --> 00:04:10,200
Oh, God, that's a tricky one,
isn't it? Er...

59
00:04:10,200 --> 00:04:11,840
Is it CO2 emissions?

60
00:04:11,840 --> 00:04:15,440
I would imagine
the most recent tsunami.

61
00:04:15,440 --> 00:04:18,120
Tsunami?
Well, yes, the 2005 tsunami

62
00:04:18,120 --> 00:04:21,920
did kill a lot of people,
tens of thousands of people.

63
00:04:21,920 --> 00:04:25,040
Something much bigger, that killed
tens of millions of people.

64
00:04:25,040 --> 00:04:27,440
Probably World War II.

65
00:04:27,440 --> 00:04:30,320
First and Second World War.
Yeah, the First, Second World War.

66
00:04:30,320 --> 00:04:34,720
War and natural disasters. That's
all I can think of. Generally war.

67
00:04:34,720 --> 00:04:39,680
Yeah, generally war. OK. So,
the actual answer is Spanish flu,

68
00:04:40,120 --> 00:04:43,840
which killed... What?!
..up to 100 million people...

69
00:04:43,840 --> 00:04:47,960
I had no idea. ..just 100
years ago. Sicknesses. Oh, wow.

70
00:04:47,960 --> 00:04:50,400
..and killed up to
a hundred million people,

71
00:04:50,400 --> 00:04:53,120
so more than World War I
and World War II combined.

72
00:04:53,120 --> 00:04:54,760
That's incredible.

73
00:04:56,720 --> 00:05:01,800
Spanish flu was a truly global
contagion, a pandemic,

74
00:05:02,240 --> 00:05:07,200
a virus 20 times more deadly
than ordinary seasonal flu.

75
00:05:07,200 --> 00:05:09,280
It was a new strain of flu

76
00:05:09,280 --> 00:05:13,080
that overwhelmed
victims' natural immunity.

77
00:05:13,080 --> 00:05:16,720
The startling fact
is that another unknown

78
00:05:16,720 --> 00:05:21,720
and potentially deadly flu virus
could emerge at any time,

79
00:05:22,280 --> 00:05:25,960
which is why our Government
needs to be prepared.

80
00:05:32,880 --> 00:05:37,920
Dotted around the country are secret
depots housing vast stockpiles,

81
00:05:38,640 --> 00:05:42,440
all poised for
the next big outbreak -

82
00:05:42,440 --> 00:05:45,040
protective kit
so that health workers

83
00:05:45,040 --> 00:05:47,920
can carry out their jobs safely,
antiviral

84
00:05:47,920 --> 00:05:51,160
drugs that can limit symptoms
if they're taken early enough.

85
00:05:52,720 --> 00:05:56,120
It is critical that the right
supplies are in the right

86
00:05:56,120 --> 00:05:58,680
place at the right time.

87
00:05:59,800 --> 00:06:03,720
But those calculations depend on
mathematical models that

88
00:06:03,720 --> 00:06:07,000
predict how quickly
a virus will spread

89
00:06:07,000 --> 00:06:10,840
and how many of us
will become infected.

90
00:06:12,600 --> 00:06:16,800
With this kind of information, you
can make really important decisions,

91
00:06:16,800 --> 00:06:20,240
things like, have you got
enough hospital beds,

92
00:06:20,240 --> 00:06:23,240
do you need to close the airports,
have you got enough

93
00:06:23,240 --> 00:06:26,800
antivirals or how many body bags
are you going to need?

94
00:06:26,800 --> 00:06:30,320
Now, these are real
life-or-death questions,

95
00:06:30,320 --> 00:06:33,480
but you don't get to have
a trial run. But with mathematical

96
00:06:33,480 --> 00:06:35,880
simulations and computer models,
you can

97
00:06:35,880 --> 00:06:40,080
try out a few different scenarios
so that when the time comes

98
00:06:40,080 --> 00:06:43,720
you can make the right decision
and save thousands of lives.

99
00:06:46,800 --> 00:06:48,640
To create the models they need,

100
00:06:48,640 --> 00:06:53,600
mathematicians use a set of rules,
basically a series of equations.

101
00:06:55,800 --> 00:06:59,880
These equations predict
how a particular disease moves,

102
00:06:59,880 --> 00:07:04,840
from person to person
and then from place to place.

103
00:07:08,400 --> 00:07:12,760
But to do that, mathematicians
need information about us -

104
00:07:12,760 --> 00:07:15,640
how we move and who we meet.

105
00:07:15,640 --> 00:07:17,360
The more accurate the data,

106
00:07:17,360 --> 00:07:20,640
the better the prediction
that emerges from the calculations.

107
00:07:26,360 --> 00:07:29,720
But how good is the data
we already have?

108
00:07:29,720 --> 00:07:34,680
Are we as prepared as we can be
for the next deadly outbreak?

109
00:07:38,360 --> 00:07:41,600
modelling for pandemics,
Dr Adam Kucharski.

110
00:07:45,800 --> 00:07:50,000
If you want to build a model
of how infectious diseases spread,

111
00:07:50,000 --> 00:07:51,520
what do you need to know?

112
00:07:51,520 --> 00:07:54,280
There are a couple of bits of
information that are really useful.

113
00:07:54,280 --> 00:07:56,320
One is the contacts
that people have.

114
00:07:56,320 --> 00:07:58,080
For an infection like flu,
for example,

115
00:07:58,080 --> 00:08:01,040
it depends on who you interact with,
who you meet.

116
00:08:01,040 --> 00:08:02,880
So that's one kind of data
we really need.

117
00:08:02,880 --> 00:08:05,200
The second thing is where
those contacts are made -

118
00:08:05,200 --> 00:08:07,080
how much you interact
with your community,

119
00:08:07,080 --> 00:08:08,880
how much you travel
to other locations,

120
00:08:08,880 --> 00:08:10,800
and that can really help us
understand how much

121
00:08:10,800 --> 00:08:14,040
a disease will spread more broadly,
outside a single community.

122
00:08:14,040 --> 00:08:17,360
Don't you have all this data
already? We're in the 21st century,

123
00:08:17,360 --> 00:08:21,200
people are being tracked all the
time. Don't you already know this?

124
00:08:21,200 --> 00:08:23,680
There is information out there,
but it's really not

125
00:08:23,680 --> 00:08:26,320
designed for this specific question
about how diseases spread.

126
00:08:26,320 --> 00:08:28,240
So we might have data
about where people are

127
00:08:28,240 --> 00:08:31,440
but no idea of who those people are,
how many contacts they're making.

128
00:08:31,440 --> 00:08:32,680
On a much smaller scale,

129
00:08:32,680 --> 00:08:35,200
there's been surveys of
what interactions people

130
00:08:35,200 --> 00:08:38,000
have in a given day, but that's
not linked to their movements.

131
00:08:38,000 --> 00:08:40,160
It's important to link
those two together,

132
00:08:40,160 --> 00:08:42,760
to have the movements
AND the contacts in one data set.

133
00:08:42,760 --> 00:08:46,920
Those surveys, do they cover
a big portion of the population?

134
00:08:46,920 --> 00:08:49,920
Generally, because of the effort
that's required to survey people...

135
00:08:49,920 --> 00:08:52,160
If you imagine writing a diary
of all your contacts,

136
00:08:52,160 --> 00:08:54,280
it's quite an intensive effort,

137
00:08:54,280 --> 00:08:56,240
and the data sets
are generally quite small.

138
00:08:56,240 --> 00:08:59,760
We're talking maybe a few hundred,
a thousand people. Wow.

139
00:08:59,760 --> 00:09:01,880
OK, so, in a population of millions,

140
00:09:01,880 --> 00:09:04,880
that's a tiny fraction of the people
you might want to capture. Yeah.

141
00:09:04,880 --> 00:09:07,480
We can say something general
about the whole country,

142
00:09:07,480 --> 00:09:10,040
but we really can't understand
how different regions,

143
00:09:10,040 --> 00:09:13,000
for example,
interact with each other.

144
00:09:13,000 --> 00:09:17,080
And this is the big problem.
Pandemic planning relies on

145
00:09:17,080 --> 00:09:19,360
limited data sets from the UK,

146
00:09:19,360 --> 00:09:23,200
samples of just hundreds
or a few thousand people,

147
00:09:23,200 --> 00:09:26,600
which means epidemiologists
have to make their best

148
00:09:26,600 --> 00:09:29,320
assumptions about how we behave.

149
00:09:29,320 --> 00:09:33,040
But if the assumptions are wrong,
the results could be, too.

150
00:09:33,040 --> 00:09:36,600
And that could have
fatal consequences

151
00:09:36,600 --> 00:09:40,000
unless we can improve
the current data.

152
00:09:40,000 --> 00:09:43,360
So, what can I do to help?
What's the ideal situation here?

153
00:09:43,360 --> 00:09:46,600
What would you like?
What would be fantastic to have

154
00:09:46,600 --> 00:09:48,800
is information on how people move,

155
00:09:48,800 --> 00:09:51,040
and ideally not just
where they think they move

156
00:09:51,040 --> 00:09:53,840
but actually some way
of verifying where they go,

157
00:09:53,840 --> 00:09:57,000
and information on who they are -
what's their age, what's their

158
00:09:57,000 --> 00:09:59,400
professional status?
And who they interact with -

159
00:09:59,400 --> 00:10:01,080
what kind of contacts do they have,

160
00:10:01,080 --> 00:10:04,000
how many people are they meeting,
have they met these people before?

161
00:10:04,000 --> 00:10:06,880
And if we had all those things, that
would be a really useful data set.

162
00:10:06,880 --> 00:10:09,360
And that would help you work out

163
00:10:09,360 --> 00:10:11,360
how a disease might spread
in the country.

164
00:10:11,360 --> 00:10:14,280
That could give us a much better
understanding of these things.

165
00:10:14,280 --> 00:10:18,160
OK, how many people do you need me
to get on board here?

166
00:10:18,160 --> 00:10:21,720
I think if we could get about 10,000
people in this kind of data set,

167
00:10:21,720 --> 00:10:22,920
that would be fantastic.

168
00:10:22,920 --> 00:10:24,760
OK.

169
00:10:24,760 --> 00:10:26,600
No problem!

170
00:10:31,880 --> 00:10:33,760
Getting data about the travel

171
00:10:33,760 --> 00:10:38,320
and social habits of
10,000 volunteers would clearly be

172
00:10:38,320 --> 00:10:41,040
a very dramatic improvement,

173
00:10:41,040 --> 00:10:46,000
a step change in the detail
available to pandemic researchers.

174
00:10:46,000 --> 00:10:47,960
It IS a challenge,

175
00:10:47,960 --> 00:10:51,440
but the technology that can
make it happen is already out there.

176
00:10:53,320 --> 00:10:56,880
Smartphones are the perfect tool
for pandemic research.

177
00:10:58,240 --> 00:10:59,560
Forget the phone bit,

178
00:10:59,560 --> 00:11:02,120
it's the combination
of a built-in GPS tracker

179
00:11:02,120 --> 00:11:07,000
and specially designed apps that
make them a data gatherer's dream.

180
00:11:10,600 --> 00:11:12,680
OK, so here's the plan.

181
00:11:12,680 --> 00:11:16,400
Adam and his fellow mathematicians
need GPS data

182
00:11:16,400 --> 00:11:20,440
and social contact data
to be able to predict the spread

183
00:11:20,440 --> 00:11:25,440
of a pandemic, and we're going to
help him, using a smartphone.

184
00:11:25,440 --> 00:11:29,760
We have created the BBC Pandemic
app, and here's how it works.

185
00:11:32,440 --> 00:11:34,240
Users who download the app

186
00:11:34,240 --> 00:11:38,760
instantly catch our harmless
virtual pandemic virus.

187
00:11:38,760 --> 00:11:41,280
To get the detailed data we need,

188
00:11:41,280 --> 00:11:45,120
the app will track their locations
anonymously

189
00:11:45,120 --> 00:11:48,440
once an hour for 24 hours.

190
00:11:49,560 --> 00:11:53,480
Users will also be asked
their age and gender,

191
00:11:53,480 --> 00:11:56,960
and the app asks them to record
the details of every person

192
00:11:56,960 --> 00:12:01,080
they encounter during
that 24-hour period.

193
00:12:01,080 --> 00:12:05,400
The combined contacts, GPS
and age information will be

194
00:12:05,400 --> 00:12:09,000
sent to our maths team
at Cambridge University.

195
00:12:11,000 --> 00:12:15,480
Real data from 10,000 people
will fundamentally improve

196
00:12:15,480 --> 00:12:19,400
the accuracy and detail
of the current assumptions

197
00:12:19,400 --> 00:12:22,400
and so predict with
greater confidence than

198
00:12:22,400 --> 00:12:27,360
ever before how devastating
the next pandemic could be.

199
00:12:27,360 --> 00:12:31,680
This really is the very first time
that GPS data has been

200
00:12:31,680 --> 00:12:35,360
combined with data about how many
people you come into contact with

201
00:12:35,360 --> 00:12:37,200
and data about your demographics.

202
00:12:37,200 --> 00:12:40,480
This is an incredibly powerful
tool for predicting

203
00:12:40,480 --> 00:12:42,400
the spread of pandemics.

204
00:12:42,400 --> 00:12:43,760
Now, I don't know about you,

205
00:12:43,760 --> 00:12:46,160
but I think that's reason enough
to download it.

206
00:12:46,160 --> 00:12:49,560
Now all I need to do is get
a few thousand people involved.

207
00:12:54,800 --> 00:12:58,720
It is now 24 minutes past eight.

208
00:12:58,720 --> 00:13:03,680
This morning, our campaign to get
10,000 app users nationwide begins.

209
00:13:04,760 --> 00:13:07,480
If a pandemic comes,
you can't stop it.

210
00:13:07,480 --> 00:13:10,760
The only thing you can do is
be as prepared as possible. Right.

211
00:13:10,760 --> 00:13:14,040
It's a difference between having
1,000 patients a day in the NHS

212
00:13:14,040 --> 00:13:16,200
or 10,000 patients a day in the NHS.

213
00:13:16,200 --> 00:13:18,840
We've built this smartphone app

214
00:13:18,840 --> 00:13:20,960
that you can download
on all different platforms.

215
00:13:20,960 --> 00:13:23,320
It's called the BBC Pandemic.

216
00:13:25,760 --> 00:13:28,800
Our national experiment
is under way.

217
00:13:28,800 --> 00:13:31,440
But that's not all
we're setting out to do.

218
00:13:33,560 --> 00:13:34,960
HANDBELL RINGS

219
00:13:34,960 --> 00:13:37,440
Oyez!

220
00:13:37,440 --> 00:13:40,760
We're recruiting for
a second experiment

221
00:13:40,760 --> 00:13:43,120
here in Haslemere in Surrey.

222
00:13:43,120 --> 00:13:46,280
Good citizens of Haslemere!

223
00:13:46,280 --> 00:13:51,320
We want this small town to give us
a close-up view of an outbreak,

224
00:13:51,560 --> 00:13:56,120
watch how a virus spreads
from one person to the next.

225
00:13:56,120 --> 00:13:58,680
Take part today and sign up for...

226
00:13:58,680 --> 00:14:02,920
If we can spread a virtual contagion
through Haslemere, this

227
00:14:02,920 --> 00:14:05,400
unprecedented detail can be used

228
00:14:05,400 --> 00:14:09,440
to test the best ways
to stop an outbreak.

229
00:14:09,440 --> 00:14:13,800
Who should be vaccinated?
Should schools be closed

230
00:14:13,800 --> 00:14:16,080
or places where people gather?

231
00:14:18,800 --> 00:14:23,640
We've persuaded nearly 500 people
from Haslemere to take part,

232
00:14:23,640 --> 00:14:27,440
and they show up as blue dots
here on our map.

233
00:14:28,520 --> 00:14:30,840
The Haslemere app is different.

234
00:14:30,840 --> 00:14:35,200
Unlike the national experiment,
the people here start uninfected

235
00:14:35,200 --> 00:14:37,440
and will only become infected

236
00:14:37,440 --> 00:14:41,680
if they come into contact
with another infected person.

237
00:14:44,880 --> 00:14:47,240
This app works by pinpointing

238
00:14:47,240 --> 00:14:51,320
their location every five minutes
for three days.

239
00:14:52,680 --> 00:14:55,720
The GPS tracks created
by each user are sent

240
00:14:55,720 --> 00:14:57,960
to our mathematicians in Cambridge.

241
00:15:00,160 --> 00:15:03,960
The team will combine all of
the tracks to identify the moments

242
00:15:03,960 --> 00:15:06,120
when our volunteers overlap.

243
00:15:09,160 --> 00:15:13,520
The equations then calculate
whether someone becomes infected

244
00:15:13,520 --> 00:15:18,360
based on how long they remain
close to an infected person.

245
00:15:20,200 --> 00:15:22,440
Once the calculations are done,

246
00:15:22,440 --> 00:15:26,120
the results will be sent
back to the app, informing users

247
00:15:26,120 --> 00:15:28,640
whether or not
they have been infected

248
00:15:28,640 --> 00:15:32,240
and, if so, how many people
they went on to infect.

249
00:15:38,120 --> 00:15:40,520
The app collects data anonymously,

250
00:15:40,520 --> 00:15:43,600
but some users have started
filming their daily

251
00:15:43,600 --> 00:15:48,280
lives in the moments before
the BBC Pandemic hits their town.

252
00:15:49,320 --> 00:15:53,800
In a real outbreak, any one of them
could become a victim.

253
00:15:53,800 --> 00:15:57,720
Hi, my name's Lynette, I'm 61 years
old. My name's Ed. I'm 37.

254
00:15:57,720 --> 00:16:00,160
I'm a schoolteacher
here in Haslemenre.

255
00:16:00,160 --> 00:16:04,120
My name is Faye Walton.
I'm 41 years old.

256
00:16:04,120 --> 00:16:08,440
My name's Pippa Hollins.
I'm a nurse in Haslemere.

257
00:16:08,440 --> 00:16:12,560
I'm just about to have my breakfast
with John, my husband.

258
00:16:12,560 --> 00:16:15,080
There's John. Say good morning,
John. Good morning!

259
00:16:15,080 --> 00:16:20,120
My name is Justine Charman.
I live and work in Haslemere.

260
00:16:20,280 --> 00:16:23,360
So, it's a nice, quiet morning
in the Apple Tree.

261
00:16:26,200 --> 00:16:30,240
Everyone with the app in Haslemere
is susceptible to getting infected,

262
00:16:30,240 --> 00:16:32,000
but our outbreak can only start

263
00:16:32,000 --> 00:16:37,080
when someone who is already infected
comes to town, a patient zero.

264
00:16:42,520 --> 00:16:47,560
The term "patient zero" is commonly
used to describe an individual

265
00:16:47,800 --> 00:16:52,400
who first introduces a contagious
disease to a community.

266
00:16:52,400 --> 00:16:54,520
Now, sometimes things
really are that simple,

267
00:16:54,520 --> 00:16:58,280
one person who somehow or other
starts everything off.

268
00:16:58,280 --> 00:17:03,320
In 2014, when Ebola hit West Africa,
epidemiologists tracked

269
00:17:03,920 --> 00:17:08,120
patient zero back to
a two-year-old toddler in Guinea.

270
00:17:08,120 --> 00:17:11,920
And in 2009,
when swine flu hit the UK,

271
00:17:11,920 --> 00:17:15,200
it was a plane with six
infected people travelling

272
00:17:15,200 --> 00:17:18,560
back from Mexico that brought it
to this country.

273
00:17:18,560 --> 00:17:20,760
Eventually,
from that handful of cases,

274
00:17:20,760 --> 00:17:23,400
half a million people
would go on to be affected.

275
00:17:24,640 --> 00:17:28,920
So, let's imagine something
similar to that last flu pandemic.

276
00:17:28,920 --> 00:17:32,280
I've just come back from my holiday,
and I'm feeling a bit tired,

277
00:17:32,280 --> 00:17:34,040
but it's probably just jet lag.

278
00:17:34,040 --> 00:17:37,640
And I've got a few errands
that I need to run about town.

279
00:17:37,640 --> 00:17:39,680
Let the BBC Pandemic begin.

280
00:17:45,240 --> 00:17:50,240
I'm now actively infectious, and the
app is tracking my every move today.

281
00:17:53,640 --> 00:17:56,520
First, to the local gym
for a spot of yoga.

282
00:17:59,760 --> 00:18:03,040
Anyone else with the app
is also being tracked.

283
00:18:11,040 --> 00:18:16,040
In Cambridge, our maths team enter
all the GPS tracks into their model.

284
00:18:17,840 --> 00:18:19,840
To create a detailed map

285
00:18:19,840 --> 00:18:23,040
that will reveal
how the virus actually spreads,

286
00:18:23,040 --> 00:18:27,160
we want to infect as many people
as possible, so the team has set

287
00:18:27,160 --> 00:18:31,840
mathematical rules that mean
no-one is immune to the virus.

288
00:18:31,840 --> 00:18:34,200
And it's highly infectious...

289
00:18:38,080 --> 00:18:41,760
..which means I could infect
someone up to 20 metres away.

290
00:18:41,760 --> 00:18:44,720
But the longer and closer
we are to each other,

291
00:18:44,720 --> 00:18:48,360
the higher the probability
of infection.

292
00:18:48,360 --> 00:18:52,400
All I need to do now
is to come in contact with as many

293
00:18:52,400 --> 00:18:54,280
people as possible.

294
00:18:57,040 --> 00:18:58,920
So, time to move on.

295
00:19:01,000 --> 00:19:04,640
A spot of shopping, I think.

296
00:19:04,640 --> 00:19:07,720
I don't want to attract
too much attention to myself today

297
00:19:07,720 --> 00:19:10,440
and risk affecting
other people's behaviour,

298
00:19:10,440 --> 00:19:13,360
so I'm wearing a small camera
as I go shopping.

299
00:19:13,360 --> 00:19:16,960
Can you recommend me something?
What do you like?

300
00:19:16,960 --> 00:19:18,520
Hmm, science!

301
00:19:21,280 --> 00:19:23,040
Where next?

302
00:19:34,000 --> 00:19:37,840
I work in a shop just over the road
from where I live.

303
00:19:37,840 --> 00:19:40,560
I've worked there for
about six years in all now.

304
00:19:40,560 --> 00:19:43,200
Have you got washers in here?
Yes. Where are they?

305
00:19:44,640 --> 00:19:49,680
Our pandemic is just a simulation,
so obviously I don't feel sick,

306
00:19:49,880 --> 00:19:52,800
but it is entirely possible
that someone with a genuine

307
00:19:52,800 --> 00:19:57,400
virus would be well enough
to walk around like this.

308
00:19:57,400 --> 00:19:59,680
All viruses are different.

309
00:19:59,680 --> 00:20:02,440
It's almost as though they've got
different personalities which

310
00:20:02,440 --> 00:20:04,640
will affect how they spread.

311
00:20:04,640 --> 00:20:08,960
So something like HIV will
lurk around inside your body

312
00:20:08,960 --> 00:20:13,960
for years before finally making
itself known when you become ill.

313
00:20:13,960 --> 00:20:18,920
And even something like flu, you can
be infectious for a full day without

314
00:20:19,040 --> 00:20:23,280
feeling ill, meaning that you can
carry on about your normal,

315
00:20:23,280 --> 00:20:27,480
everyday business without realising
that you're spreading a disease.

316
00:20:33,840 --> 00:20:36,200
So, today I popped in
for a coffee with my friend,

317
00:20:36,200 --> 00:20:37,840
and that was really nice.

318
00:20:37,840 --> 00:20:41,200
Thank you very much. And your food
order's been taken, yeah? Thank you.

319
00:20:41,200 --> 00:20:42,680
We're on Haslemere High Street.

320
00:20:42,680 --> 00:20:46,320
We've just left the studio for the
day. Everything's all cleared up.

321
00:20:49,320 --> 00:20:54,280
Hello, I'm Mary. This is Mary. OK,
and we run the checkouts together.

322
00:21:00,320 --> 00:21:04,640
Right, time for a pint, I think,
and my last stop here for the day.

323
00:21:04,640 --> 00:21:07,320
Hopefully, there'll be a few
people in here with the app.

324
00:21:07,320 --> 00:21:09,400
Thanking you!

325
00:21:09,400 --> 00:21:13,640
But it's going to take a few days
to be sure of how far the virtual

326
00:21:13,640 --> 00:21:16,640
BBC virus has spread.

327
00:21:16,640 --> 00:21:19,640
And all the while,
our volunteers are blithely

328
00:21:19,640 --> 00:21:22,480
carrying on with
their normal routine.

329
00:21:22,480 --> 00:21:25,440
It is the end of the day for me,
and it's my favourite

330
00:21:25,440 --> 00:21:28,200
part of the day, because we're
going to go and pick Daisy up.

331
00:21:28,200 --> 00:21:30,920
It's been a really, really long day.
It's about half past nine.

332
00:21:30,920 --> 00:21:33,280
I've come into contact
with loads of kids today -

333
00:21:33,280 --> 00:21:35,440
a big assembly,
I've taken all of Year 7 out.

334
00:21:35,440 --> 00:21:38,960
So, it's 20 past 11, Thursday night.

335
00:21:38,960 --> 00:21:42,160
I've just finished work
after a very, very long day.

336
00:21:46,880 --> 00:21:48,720
Over the next 72 hours,

337
00:21:48,720 --> 00:21:53,480
our maths team in Cambridge will be
working on the data day and night,

338
00:21:53,480 --> 00:21:56,240
calculating how many of
the good people of Haslemere

339
00:21:56,240 --> 00:21:57,760
- or their phones, at least -

340
00:21:57,760 --> 00:22:01,400
have caught the BBC Pandemic
from my patient zero walk.

341
00:22:12,760 --> 00:22:15,160
JAVID: Spanish flu was by far

342
00:22:15,160 --> 00:22:18,680
the deadliest pandemic
in the last century.

343
00:22:18,680 --> 00:22:23,080
At its peak, around 15,000 people
died in England

344
00:22:23,080 --> 00:22:26,080
and Wales in a single week.

345
00:22:26,080 --> 00:22:30,040
It was so virulent,
it made people's lungs bleed.

346
00:22:30,040 --> 00:22:32,760
Death followed within hours.

347
00:22:32,760 --> 00:22:37,640
It was a terrifying variant of
the seasonal flu we see each winter.

348
00:22:40,240 --> 00:22:43,960
But what makes the radical
difference between seasonal

349
00:22:43,960 --> 00:22:46,480
and pandemic flu?

350
00:22:46,480 --> 00:22:50,520
I'm meeting one of the country's
leading experts to find out

351
00:22:50,520 --> 00:22:54,440
how a flu virus can transform
into a killer that can

352
00:22:54,440 --> 00:22:56,600
threaten our nation at any moment.

353
00:22:59,200 --> 00:23:01,080
If we think about
what one of these viruses

354
00:23:01,080 --> 00:23:03,640
looks like to our body,
here it comes.

355
00:23:03,640 --> 00:23:07,920
It's basically a bag with all
its genetic material in the middle,

356
00:23:07,920 --> 00:23:12,600
and on the outside it's got
all these little spikes.

357
00:23:12,600 --> 00:23:15,320
Now, all these spikes
are what's sticking out

358
00:23:15,320 --> 00:23:20,320
and being presented to our immune
system, and it makes antibodies.

359
00:23:20,400 --> 00:23:24,560
The antibodies we can draw like
little Y-shaped things, and they

360
00:23:24,560 --> 00:23:29,240
perfectly recognise the exact shape
of the spike of this influenza

361
00:23:29,240 --> 00:23:33,760
virus, and the person is protected
against that invading virus.

362
00:23:33,760 --> 00:23:36,760
These antibodies are very specific
to that flu virus,

363
00:23:36,760 --> 00:23:39,600
so we're immune to flu, surely.

364
00:23:39,600 --> 00:23:41,800
Yes, that would be the case

365
00:23:41,800 --> 00:23:46,200
if everything just stayed the same
year on year, but it doesn't,

366
00:23:46,200 --> 00:23:50,520
and the reason for that
is viruses create thousands

367
00:23:50,520 --> 00:23:55,520
and millions of copies of themselves
in a person as they replicate.

368
00:23:55,520 --> 00:24:00,520
Now, in copying themselves,
they sometimes make mistakes,

369
00:24:01,240 --> 00:24:05,960
so if this virus is replicating
and it makes a mistake in the gene

370
00:24:05,960 --> 00:24:09,480
that encodes this spike,
now the spike looks different.

371
00:24:09,480 --> 00:24:11,240
It might be a different shape.

372
00:24:11,240 --> 00:24:14,640
I'm going to actually draw it
as a different colour here.

373
00:24:14,640 --> 00:24:19,640
So now the spikes are all
coloured yellow. OK?

374
00:24:20,240 --> 00:24:24,120
It's a bit like
you go outside in your coat

375
00:24:24,120 --> 00:24:26,880
and then you change it and put on
a different-coloured coat.

376
00:24:26,880 --> 00:24:29,800
You now look different
to everybody outside.

377
00:24:29,800 --> 00:24:33,880
This virus looks different to
all the antibodies that had been

378
00:24:33,880 --> 00:24:35,960
induced in the first place
to the white virus.

379
00:24:35,960 --> 00:24:38,520
So even though you had all those
white antibodies, they don't

380
00:24:38,520 --> 00:24:41,160
do you any good whatsoever
against this yellow virus.

381
00:24:45,480 --> 00:24:49,880
It's this ability of the flu virus
to transform itself that can

382
00:24:49,880 --> 00:24:52,160
make it so dangerous.

383
00:24:57,120 --> 00:25:00,840
When the outer coat of proteins
changes in only a small way,

384
00:25:00,840 --> 00:25:05,720
that's called antigenic drift.
It's still a familiar virus.

385
00:25:05,720 --> 00:25:07,680
That's what happens
with seasonal flu.

386
00:25:09,200 --> 00:25:13,520
These relatively minor changes
in the coat of seasonal flu

387
00:25:13,520 --> 00:25:17,560
viruses mean our immune system
gives us some protection.

388
00:25:19,120 --> 00:25:23,760
But the real danger is that
a new, completely unrecognisable

389
00:25:23,760 --> 00:25:28,800
flu virus will emerge, a virus
with pandemic potential

390
00:25:29,040 --> 00:25:32,080
which our bodies have
no immunity to fight off

391
00:25:32,080 --> 00:25:34,160
and can cause far more damage.

392
00:25:36,040 --> 00:25:38,720
Pandemic flu viruses emerge

393
00:25:38,720 --> 00:25:42,160
because humans are not the only
creatures that get flu.

394
00:25:43,520 --> 00:25:48,440
Pandemics are influenza viruses
that come across from other species

395
00:25:48,440 --> 00:25:51,040
of animals into humans,
usually from birds,

396
00:25:51,040 --> 00:25:54,320
wild birds like ducks, geese, swans.

397
00:25:54,320 --> 00:25:56,240
And the thing about those viruses is

398
00:25:56,240 --> 00:25:58,240
although they're also
still flu viruses -

399
00:25:58,240 --> 00:26:01,960
they look just the same
on the inside - their coats,

400
00:26:01,960 --> 00:26:05,320
their spikes that are on the outside
are completely different.

401
00:26:13,200 --> 00:26:17,240
Because these viruses have got genes
from the animal population,

402
00:26:17,240 --> 00:26:20,480
it can mean that their coat
is dramatically different.

403
00:26:20,480 --> 00:26:22,760
That's called antigenic shift,

404
00:26:22,760 --> 00:26:27,680
and it can mean the human immune
system is entirely unprepared.

405
00:26:29,560 --> 00:26:33,480
Therefore everybody gets infected,
the outbreak is explosive

406
00:26:33,480 --> 00:26:36,520
because everybody is susceptible,

407
00:26:36,520 --> 00:26:40,480
and we see it ripping round the
world in a matter of days and weeks.

408
00:26:42,640 --> 00:26:46,440
Right now, there are several
particularly worrying flu

409
00:26:46,440 --> 00:26:49,320
viruses circulating
in bird populations.

410
00:26:51,160 --> 00:26:54,000
So far, those viruses
haven't mutated enough

411
00:26:54,000 --> 00:26:56,800
to transmit easily between humans.

412
00:26:58,800 --> 00:27:01,520
But the fear is they can, which is

413
00:27:01,520 --> 00:27:04,360
why we are conducting
our experiment.

414
00:27:04,360 --> 00:27:08,840
We need as much data as possible
before the next pandemic erupts.

415
00:27:13,600 --> 00:27:16,760
In Haslemere, we've been using
our local app to track

416
00:27:16,760 --> 00:27:19,520
nearly 500 people for three days.

417
00:27:20,640 --> 00:27:25,000
The aim here is to see
in exceptional detail how

418
00:27:25,000 --> 00:27:29,120
a virus moves from person to person
as an outbreak unfolds.

419
00:27:30,480 --> 00:27:34,400
If we can see what drives
an infection through this community,

420
00:27:34,400 --> 00:27:39,400
it could help save the lives of
these people come a real pandemic.

421
00:27:39,640 --> 00:27:43,680
Hi! It's Friday,
the shop is super-busy.

422
00:27:43,680 --> 00:27:45,600
We've had an absolutely mental day.

423
00:27:45,600 --> 00:27:48,440
I know a lot of people are using

424
00:27:48,440 --> 00:27:52,920
this pandemic app
that are in the pub.

425
00:27:52,920 --> 00:27:56,040
I probably have contaminated
quite a few people now.

426
00:27:56,040 --> 00:27:59,480
Or, if any of them have got it,
they probably have given it to me.

427
00:27:59,480 --> 00:28:02,120
This is my Friday evening.
My son's got a friend coming round,

428
00:28:02,120 --> 00:28:03,840
and I'm about to
cook them both dinner.

429
00:28:03,840 --> 00:28:07,040
So if I am contaminated,
that will pass on.

430
00:28:07,040 --> 00:28:10,680
I hope this information is helpful
to you. Have a good evening. Cheers!

431
00:28:18,040 --> 00:28:22,160
All the volunteers in Haslemere
are about to discover their fate.

432
00:28:23,360 --> 00:28:26,240
Only the computer model
knows which, if any,

433
00:28:26,240 --> 00:28:28,920
of the people turning up here
are infected.

434
00:28:28,920 --> 00:28:32,680
I need to give them a PIN number
to reveal their individual results.

435
00:28:33,760 --> 00:28:38,000
There's been no dry run. Nothing
like this has been attempted before.

436
00:28:39,040 --> 00:28:44,080
I'm about to find out if my walk
around town had any effect at all.

437
00:28:44,880 --> 00:28:47,120
To be honest, it's all
a bit nerve-racking.

438
00:28:49,520 --> 00:28:53,960
To infect you all, there had to
first be a patient zero,

439
00:28:53,960 --> 00:28:56,640
which you may have guessed
was actually me.

440
00:28:57,720 --> 00:29:02,760
We will see if I actually managed
to successfully infect any of you.

441
00:29:04,320 --> 00:29:06,840
OK, so if you could get
your phones out, I'm going

442
00:29:06,840 --> 00:29:10,160
to give two different PIN numbers.

443
00:29:10,160 --> 00:29:13,160
When you put in that PIN number,
it will tell you

444
00:29:13,160 --> 00:29:15,200
whether or not you've been infected.

445
00:29:15,200 --> 00:29:17,720
If you have been infected,
please move over to this side,

446
00:29:17,720 --> 00:29:21,440
and if you haven't been infected,
move over to that side.

447
00:29:21,440 --> 00:29:23,920
OK, so the two PIN numbers are...

448
00:29:23,920 --> 00:29:25,320
Drum roll!

449
00:29:26,360 --> 00:29:31,400
..for Android,
please type in 8165...

450
00:29:32,320 --> 00:29:37,200
..and for Apple, type in 8195.

451
00:29:39,720 --> 00:29:43,160
OK, if you could move to this side
if you've been infected

452
00:29:43,160 --> 00:29:45,400
and that side if you
have not been infected...

453
00:29:55,200 --> 00:29:58,520
Look at that! OK.

454
00:29:58,520 --> 00:30:02,840
I have to confess, I think
we've got a good number of you...

455
00:30:02,840 --> 00:30:05,120
We managed to infect
a good number of you.

456
00:30:05,120 --> 00:30:06,960
VOICEOVER: This is a great result.

457
00:30:06,960 --> 00:30:10,840
72 people in this small gathering
alone have been infected.

458
00:30:13,040 --> 00:30:16,200
Yes, I was infected. I'm really
surprised at the results.

459
00:30:16,200 --> 00:30:17,880
I've infected one person.

460
00:30:17,880 --> 00:30:21,120
I know exactly where I got infected
and what time I got infected,

461
00:30:21,120 --> 00:30:22,920
just purely because
I'm on annual leave

462
00:30:22,920 --> 00:30:25,160
so I haven't done much,
just pottered around the house.

463
00:30:25,160 --> 00:30:27,520
So the one time I did go out
I actually got infected.

464
00:30:30,160 --> 00:30:35,200
But this is just a fraction of
all our volunteers in Haslemere.

465
00:30:36,560 --> 00:30:39,560
To see how the infection
moved from person to person,

466
00:30:39,560 --> 00:30:42,280
we need to look
at the bigger picture.

467
00:30:47,000 --> 00:30:49,920
Our mathematicians from
the University of Cambridge,

468
00:30:49,920 --> 00:30:52,080
led by Professor Julia Gog,

469
00:30:52,080 --> 00:30:54,920
have finished crunching the data.

470
00:30:56,520 --> 00:31:00,760
The first thing I want to show you
is the patient zero track,

471
00:31:00,760 --> 00:31:02,400
which is you!

472
00:31:02,400 --> 00:31:05,600
That's the High Street there.
Mm-hm. The gym there.

473
00:31:05,600 --> 00:31:08,640
That's where I go first, that's
where I go to yoga. Here I come.

474
00:31:08,640 --> 00:31:11,560
The blue dot there is one of
the users, the other users,

475
00:31:11,560 --> 00:31:14,640
and they're uninfected, because
everyone else starts uninfected. OK.

476
00:31:14,640 --> 00:31:16,760
It's only you that started infected.

477
00:31:16,760 --> 00:31:19,840
And let's see what happens
when you cross paths with them.

478
00:31:19,840 --> 00:31:24,560
Ooh...! And here I come,
ready to infect them.

479
00:31:24,560 --> 00:31:27,040
So, I'm looping round, because
I'm doing some filming. And...

480
00:31:27,040 --> 00:31:31,200
Ah! So that's them getting infected.
Now they're infected.

481
00:31:31,200 --> 00:31:33,120
And they're off up the High Street.

482
00:31:33,120 --> 00:31:36,200
So if they cross paths
with someone else,

483
00:31:36,200 --> 00:31:38,800
there's a chance that other person
will get infected. Exactly.

484
00:31:38,800 --> 00:31:41,600
So it's a chance. And the closer
they are, the longer they are,

485
00:31:41,600 --> 00:31:43,040
infection could happen.

486
00:31:53,600 --> 00:31:57,480
That's one person, then.
That's one infection.

487
00:31:57,480 --> 00:32:00,520
And what happens next?
Ah, well, let's have a look.

488
00:32:02,760 --> 00:32:06,000
So, this is looking at, zoomed out,
on the High Street. This is still me

489
00:32:06,000 --> 00:32:07,840
as I'm wandering round.
Yeah.

490
00:32:07,840 --> 00:32:12,280
And now there's loads of dots,
loads of people who are infected.

491
00:32:12,280 --> 00:32:14,520
The blue dots are people
who haven't been infected yet,

492
00:32:14,520 --> 00:32:16,760
and you can see the orange or red
with the circle around.

493
00:32:16,760 --> 00:32:20,840
And you can see you because you've
got the extra dotted halo around.

494
00:32:20,840 --> 00:32:23,880
And actually, often I come
into contact with people or people

495
00:32:23,880 --> 00:32:27,160
are coming into contact with each
other and not getting infected.

496
00:32:27,160 --> 00:32:29,640
So it's not every single time.
No.

497
00:32:29,640 --> 00:32:32,720
It depends on proximity,
length of time.

498
00:32:32,720 --> 00:32:34,440
So that's the whole of Haslemere.

499
00:32:34,440 --> 00:32:36,760
So now this is when I went off
to the supermarket.

500
00:32:36,760 --> 00:32:39,000
And in the pub, of course,
at the end of the day!

501
00:32:39,000 --> 00:32:40,600
A vital way to end a day!

502
00:32:40,600 --> 00:32:43,360
We're now at
six o'clock in the evening.

503
00:32:43,360 --> 00:32:44,920
I got there about half ten or so.

504
00:32:44,920 --> 00:32:48,120
How many people by this point

505
00:32:48,120 --> 00:32:49,920
are infected overall?

506
00:32:49,920 --> 00:32:54,840
There were 77. 77?! Yes.
Just in one day? That's right.

507
00:32:54,840 --> 00:32:59,880
Just from me arriving. OK.
How many people did I infect, then?

508
00:33:02,920 --> 00:33:06,560
more got infected,
more got infected... Mm-hm. Wow.

509
00:33:06,560 --> 00:33:11,640
With no immunity to our virus,
by Day 3 the infection has

510
00:33:12,240 --> 00:33:15,160
spread out of control
through our volunteers.

511
00:33:16,280 --> 00:33:21,360
Loads of people have this
all over the map. Oh, my goodness!

512
00:33:21,920 --> 00:33:25,760
What have we done?!
SHE LAUGHS

513
00:33:25,760 --> 00:33:28,600
Wow, it's everywhere.
It's not even...

514
00:33:28,600 --> 00:33:30,720
You don't even just have
little clusters any more,

515
00:33:30,720 --> 00:33:32,320
it's literally everywhere.

516
00:33:34,000 --> 00:33:37,400
And people are still
getting infected. Oh, my goodness.

517
00:33:39,760 --> 00:33:43,480
That is an extraordinary number
of people to be infected from just

518
00:33:43,480 --> 00:33:46,720
one person walking
down the High Street. Yeah.

519
00:33:46,720 --> 00:33:48,600
VOICEOVER: In just three days,

520
00:33:48,600 --> 00:33:53,240
86% of all people who took part
were infected.

521
00:33:53,240 --> 00:33:58,320
Our pandemic virus spread like
wildfire because no-one was immune.

522
00:33:58,720 --> 00:34:00,600
If this had been a real virus,

523
00:34:00,600 --> 00:34:03,240
the town would have
come to a standstill.

524
00:34:07,320 --> 00:34:10,320
Instead, it's created
an extraordinary

525
00:34:10,320 --> 00:34:15,320
simulation of how an infection
would spread from person to person.

526
00:34:15,400 --> 00:34:19,560
Even better, our app also reveals
information that will help

527
00:34:19,560 --> 00:34:23,120
test how we can stop
an infection spreading.

528
00:34:23,120 --> 00:34:25,480
For those of you who were infected,

529
00:34:25,480 --> 00:34:30,400
your apps should also tell you how
many people you went on to infect.

530
00:34:30,400 --> 00:34:32,560
How many people
infected no-one else?

531
00:34:33,680 --> 00:34:36,360
How many people infected
one or more?

532
00:34:37,400 --> 00:34:39,560
Two or more?

533
00:34:39,560 --> 00:34:41,360
Three or more?

534
00:34:41,360 --> 00:34:43,640
Four or more?

535
00:34:43,640 --> 00:34:45,480
Five or more?

536
00:34:45,480 --> 00:34:47,200
How many did you get, sir?

537
00:34:47,200 --> 00:34:48,800
Six. Six.
And, madam?

538
00:34:48,800 --> 00:34:51,560
Five. Five. OK. These are
our super-spreaders, then!

539
00:34:52,920 --> 00:34:55,280
You got six people.
I got six people.

540
00:34:55,280 --> 00:34:57,480
I'm feeling very guilty about it!

541
00:34:57,480 --> 00:34:59,680
And I already know
I need to apologise to

542
00:34:59,680 --> 00:35:01,480
a lot of people at the coffee shop!

543
00:35:01,480 --> 00:35:03,960
Oh, is that where you think it was,
where you think it happened?

544
00:35:03,960 --> 00:35:07,120
Friday morning. You were
sitting in the coffee shop? I was.

545
00:35:07,120 --> 00:35:09,480
But actually, I've just noticed,

546
00:35:09,480 --> 00:35:13,080
I spoke to you
on my patient zero walk.

547
00:35:13,080 --> 00:35:14,520
You were in the hardware shop.

548
00:35:14,520 --> 00:35:18,080
That's right. Is that where
you work? I do. Did I infect you?

549
00:35:18,080 --> 00:35:21,240
Yeah, I assume so! And how many
people did you then go on to infect?

550
00:35:21,240 --> 00:35:25,040
Eight. Eight people?
That makes you the champion.

551
00:35:26,440 --> 00:35:29,840
And it's not a great
championship to win. No.

552
00:35:29,840 --> 00:35:32,520
Do you think it was just people that
you were speaking to in the shop?

553
00:35:32,520 --> 00:35:34,040
I think it's just the high volume

554
00:35:34,040 --> 00:35:36,240
of customers that we have,
really, in the shop.

555
00:35:40,120 --> 00:35:42,760
I'm quite impressed, actually!

556
00:35:42,760 --> 00:35:46,400
I don't really know!
It's quite shocking, really, eight.

557
00:35:48,320 --> 00:35:52,480
As our data reveals, when it comes
to infectious diseases,

558
00:35:52,480 --> 00:35:55,520
clearly we aren't all equal.

559
00:35:55,520 --> 00:35:59,240
Pandemic researchers
call people like Justine

560
00:35:59,240 --> 00:36:02,520
social super-spreaders, and they're
thought to be key players

561
00:36:02,520 --> 00:36:04,440
in the spread of a real infection.

562
00:36:05,720 --> 00:36:09,560
It may seem obvious that someone
whose job or lifestyle

563
00:36:09,560 --> 00:36:14,000
brings them into contact with a lot
of people could spread more disease,

564
00:36:14,000 --> 00:36:18,400
but this is the first time they've
actually been identified like this.

565
00:36:19,920 --> 00:36:24,920
Thanks to the app, we can now test
a potentially life-saving idea.

566
00:36:26,480 --> 00:36:29,200
If we can understand
who the super-spreaders are,

567
00:36:29,200 --> 00:36:31,080
then it might help with control,

568
00:36:31,080 --> 00:36:33,960
for example a situation
where we've got a vaccine

569
00:36:33,960 --> 00:36:37,280
but either it's in short supply
or we've got a very limited time

570
00:36:37,280 --> 00:36:39,720
to deploy it so we can't
just vaccinate everyone.

571
00:36:39,720 --> 00:36:42,680
You've got to choose who you're
going to vaccinate. Be selective.

572
00:36:42,680 --> 00:36:45,680
Exactly, and rather than choosing
at random, you can be really

573
00:36:45,680 --> 00:36:48,960
strategic and choose those who are
going to infect a lot of other

574
00:36:48,960 --> 00:36:51,560
people or would have
infected other people.

575
00:36:51,560 --> 00:36:54,120
So on the left is going to be
our original Haslemere epidemic,

576
00:36:54,120 --> 00:36:56,920
and on the right is going to be
our simulation of what would have

577
00:36:56,920 --> 00:36:59,280
happened if we vaccinate
the super-spreaders.

578
00:36:59,280 --> 00:37:00,480
How do you define that?

579
00:37:00,480 --> 00:37:03,600
Yeah, so, for this one we chose
the people who actually infected

580
00:37:03,600 --> 00:37:05,680
more than three other
people...three or more.

581
00:37:05,680 --> 00:37:08,080
So if you take those out...
Yeah.

582
00:37:08,080 --> 00:37:10,920
..vaccinate them so they're
immune from the virus... Yep.

583
00:37:10,920 --> 00:37:14,760
..and see how it changes. Yes.
OK. And that was only 46 people,

584
00:37:14,760 --> 00:37:17,360
so it's a pretty small
vaccine roll-out.

585
00:37:17,360 --> 00:37:20,280
And what you see is the dots
are in the same place. Mm-hm.

586
00:37:20,280 --> 00:37:25,200
But... So, this is me,
my little walk on the first day.

587
00:37:25,200 --> 00:37:28,200
A few infected. Quite a lot going on
on this one.

588
00:37:28,200 --> 00:37:30,520
This one we've seen before.
Much less, actually.

589
00:37:30,520 --> 00:37:33,640
We're in Day 2 now.
Far less! Oh, my gosh!

590
00:37:34,720 --> 00:37:38,320
Oh, my gosh. So just vaccinating
10% of the population.

591
00:37:38,320 --> 00:37:40,240
That's a much slower epidemic.

592
00:37:48,400 --> 00:37:51,720
We vaccinated 46.
Stephen, you ran this one.

593
00:37:51,720 --> 00:37:55,720
How many fewer people were there
infected in the super-spreaders?

594
00:37:55,720 --> 00:37:57,280
There were...

595
00:37:58,320 --> 00:38:01,080
Sorry... 112 fewer.
OK. 112.

596
00:38:01,080 --> 00:38:04,280
So that's people
who weren't vaccinated

597
00:38:04,280 --> 00:38:08,080
but ended up not getting infected
just by virtue

598
00:38:08,080 --> 00:38:13,120
of the fact that the super-spreaders
were protected away from the virus.

599
00:38:14,000 --> 00:38:17,400
That's exactly right. Gosh, it does
make quite a big difference, then.

600
00:38:20,400 --> 00:38:25,080
We simulated the vaccination
of only 10% of our app users,

601
00:38:25,080 --> 00:38:27,240
but targeting these super-spreaders

602
00:38:27,240 --> 00:38:30,400
reduced the number
of infections by 40%.

603
00:38:32,680 --> 00:38:37,600
Our app allowed us to identify
super-spreaders in Haslemere, but in

604
00:38:37,600 --> 00:38:41,760
the real world, people who work
in busy places like schools

605
00:38:41,760 --> 00:38:44,120
could also be super-spreaders.

606
00:38:45,320 --> 00:38:50,320
Deciding who to vaccinate first
with greatest effect is problematic,

607
00:38:51,840 --> 00:38:55,480
but there could be a way
of identifying individual

608
00:38:55,480 --> 00:38:58,360
super-spreaders BEFORE a contagion,

609
00:38:58,360 --> 00:39:02,560
and that could mean a revolution
in pandemic prevention.

610
00:39:06,680 --> 00:39:10,560
That's the theory being tested
in a unique experiment.

611
00:39:11,840 --> 00:39:15,000
Researchers at Imperial College
want to know if some

612
00:39:15,000 --> 00:39:18,480
people are biological
super-spreaders.

613
00:39:18,480 --> 00:39:20,720
Is there something about their genes

614
00:39:20,720 --> 00:39:23,400
that makes them become
highly infectious?

615
00:39:24,760 --> 00:39:29,040
If there is a biological reason,
there could be a test to identify

616
00:39:29,040 --> 00:39:33,120
super-spreaders before
an outbreak hits, and that's

617
00:39:33,120 --> 00:39:37,520
the revolution in pandemic control
that could be starting here today.

618
00:39:39,520 --> 00:39:44,400
It's about 7am, and this looks like
any ordinary hospital wing.

619
00:39:44,400 --> 00:39:46,920
In fact, it's a research facility.

620
00:39:46,920 --> 00:39:50,000
The patients through there
are perfectly healthy...

621
00:39:51,440 --> 00:39:52,840
..for the time being.

622
00:39:54,760 --> 00:39:58,240
Technically, Joydeep Dutta
isn't a patient.

623
00:39:58,240 --> 00:40:03,040
He's one of ten healthy volunteers
being quarantined in this ward

624
00:40:03,040 --> 00:40:07,840
about to be infected with a large
dose of purified swine flu,

625
00:40:07,840 --> 00:40:11,560
because researchers are hoping
they can pinpoint what makes

626
00:40:11,560 --> 00:40:15,280
people react differently
to the same virus.

627
00:40:15,280 --> 00:40:18,280
Dr Chris Chiu is running
the experiment.

628
00:40:20,280 --> 00:40:24,360
There's so much that we don't know
about flu infection -

629
00:40:24,360 --> 00:40:27,160
why some people get
more sick than others.

630
00:40:27,160 --> 00:40:31,800
And in fact, we don't really know
what contagiousness is.

631
00:40:33,880 --> 00:40:38,120
To find out what might make
someone particularly contagious,

632
00:40:38,120 --> 00:40:42,240
the team will dose Joydeep
and the other volunteers with flu

633
00:40:42,240 --> 00:40:44,080
then compare how they respond.

634
00:40:47,720 --> 00:40:49,800
Each of these vials is packed with

635
00:40:49,800 --> 00:40:53,320
five million
swine flu virus particles,

636
00:40:53,320 --> 00:40:57,200
the same strain that caused
the 2009 pandemic.

637
00:40:59,880 --> 00:41:03,120
Meanwhile, Joydeep is
calmly waiting next door.

638
00:41:04,080 --> 00:41:06,360
What made you want to do this?

639
00:41:06,360 --> 00:41:11,360
I had a family member who had
the swine flu in 2009, and she was

640
00:41:11,360 --> 00:41:15,440
quite affected by it, and it changed
her life, to a certain extent.

641
00:41:22,400 --> 00:41:27,120
What happens next is down to
a battle between the five million

642
00:41:27,120 --> 00:41:30,360
virus particles and
Joydeep's personal biology.

643
00:41:31,920 --> 00:41:35,760
At the moment, there's no way
of predicting how he will react.

644
00:41:37,440 --> 00:41:41,320
Over the following days,
his samples, along with those from

645
00:41:41,320 --> 00:41:45,240
the other volunteers, are tested
to see how infected they've become.

646
00:41:48,240 --> 00:41:51,600
Everyone was inoculated
with the same dose,

647
00:41:51,600 --> 00:41:54,920
which makes the difference
in their results astonishing.

648
00:41:54,920 --> 00:41:58,720
Chris first shows me a sample
from one of the other volunteers.

649
00:42:00,320 --> 00:42:04,360
What you can see here is that
every spot that's green is a cell

650
00:42:04,360 --> 00:42:06,040
that's infected with flu.

651
00:42:06,040 --> 00:42:10,000
So you're physically seeing
swine flu virus right there,

652
00:42:10,000 --> 00:42:12,160
and it's showing up green.

653
00:42:12,160 --> 00:42:13,960
Exactly.

654
00:42:13,960 --> 00:42:17,040
VOICEOVER: But Joydeep's samples
look radically different.

655
00:42:18,880 --> 00:42:23,920
Joydeep has made a lot more virus
in his nose than the other person.

656
00:42:24,160 --> 00:42:28,200
We can zoom in on one of those areas
where it's particularly dense, and

657
00:42:28,200 --> 00:42:33,200
here there's one which looks like
it's swollen up and falling apart.

658
00:42:34,000 --> 00:42:36,960
Oh, wow! That literally does
look like a sort of...

659
00:42:36,960 --> 00:42:38,760
..almost a puff of smoke.

660
00:42:38,760 --> 00:42:43,000
So that's a cell that's
bursting open in death, so to speak,

661
00:42:43,000 --> 00:42:45,360
overwhelmed by virus.
That's right, yeah.

662
00:42:45,360 --> 00:42:47,400
That's exactly what's happening.
Gosh.

663
00:42:51,920 --> 00:42:56,360
It's a very early result, but it
suggests there is something specific

664
00:42:56,360 --> 00:43:01,120
about Joydeep's biology that could
make him particularly infectious.

665
00:43:02,200 --> 00:43:06,800
Do you think you actually might be
able to find a genetic test

666
00:43:06,800 --> 00:43:10,600
to predetermine who might become
really contagious with flu?

667
00:43:10,600 --> 00:43:13,640
Yeah, so, that would be
the ultimate hope. Wow.

668
00:43:17,080 --> 00:43:22,000
There is a way to go before
this research turns into a reality,

669
00:43:22,000 --> 00:43:25,800
but once we have the ability
to identify super-spreaders,

670
00:43:25,800 --> 00:43:28,560
our data shows
vaccinating them could be

671
00:43:28,560 --> 00:43:31,880
an effective way
of reducing a contagion.

672
00:43:33,920 --> 00:43:37,280
What's so exciting about
the data we have gathered is that

673
00:43:37,280 --> 00:43:42,320
mathematicians can now use it
to test a range of interventions.

674
00:43:43,480 --> 00:43:46,160
What if we closed schools?

675
00:43:46,160 --> 00:43:49,800
How would that impact
the number of people infected?

676
00:43:49,800 --> 00:43:52,560
And shutting down places
where people gather?

677
00:43:55,160 --> 00:43:57,640
Or asking them to stay at home?

678
00:43:59,360 --> 00:44:03,280
Now we have this data, there are
all manner of possibilities

679
00:44:03,280 --> 00:44:05,360
that could be tested to reveal how

680
00:44:05,360 --> 00:44:08,360
communities like Haslemere
could be saved...

681
00:44:08,360 --> 00:44:10,400
..but for one snag.

682
00:44:11,560 --> 00:44:14,440
Haslemere is not
closed off from the world.

683
00:44:15,640 --> 00:44:18,920
I've been down to
Christchurch in Dorset.

684
00:44:18,920 --> 00:44:21,840
I'm now down on the Bakerloo line,
at Marylebone.

685
00:44:21,840 --> 00:44:24,760
I'm just going to take a trip on
the Tube and head back to Waterloo.

686
00:44:26,520 --> 00:44:29,560
Viruses travel
wherever we take them,

687
00:44:29,560 --> 00:44:32,960
from town to town
or village or city...

688
00:44:34,120 --> 00:44:38,240
..which is why three weeks ago
we also launched our national

689
00:44:38,240 --> 00:44:43,240
experiment in the hope of getting
10,000 people to take part.

690
00:44:47,160 --> 00:44:50,000
The second phase of our experiment
is going to use the

691
00:44:50,000 --> 00:44:53,240
BBC Pandemic app to collect
data from volunteers

692
00:44:53,240 --> 00:44:56,880
all around the country,
and our mathematicians are going to

693
00:44:56,880 --> 00:45:01,880
calculate how quickly a
Spanish-flu-like virus might spread.

694
00:45:02,080 --> 00:45:05,520
We'll find out how many of us
could get infected

695
00:45:05,520 --> 00:45:07,600
and how many of us could die.

696
00:45:12,480 --> 00:45:14,960
In the national experiment,
users will

697
00:45:14,960 --> 00:45:20,000
"catch" our harmless virtual virus
as soon as they start the app.

698
00:45:20,520 --> 00:45:25,560
This time, they will be tracked
once an hour for 24 hours.

699
00:45:26,360 --> 00:45:31,400
The aim now is to track journeys
around the entire country,

700
00:45:31,400 --> 00:45:35,400
across cities and between
towns and countryside.

701
00:45:36,800 --> 00:45:41,600
The app also asks users for
their age and gender and to report

702
00:45:41,600 --> 00:45:46,400
the age and number of people
they encounter during that 24 hours.

703
00:45:48,640 --> 00:45:52,280
The aim is to collect data
that will improve our understanding

704
00:45:52,280 --> 00:45:55,000
of how much we actually
move and mix.

705
00:45:55,000 --> 00:45:59,080
For example, do some age groups
travel more than others?

706
00:46:00,280 --> 00:46:04,200
How isolated are older people?

707
00:46:04,200 --> 00:46:08,360
If we can recruit 10,000 users,
it will be the largest,

708
00:46:08,360 --> 00:46:13,360
most detailed UK data set ever
collected and it will predict how

709
00:46:13,840 --> 00:46:18,840
far and how fast a lethal pandemic
could spread across the country.

710
00:46:26,560 --> 00:46:30,840
If a virus that was carrying
a very high case fatality became

711
00:46:30,840 --> 00:46:33,880
transmissible through the air
between people,

712
00:46:33,880 --> 00:46:36,800
we could have a devastating
pandemic on our hands.

713
00:46:36,800 --> 00:46:39,320
Millions of people
could die around the world.

714
00:46:41,560 --> 00:46:45,960
We know that a pandemic will come.
There's no doubt about that.

715
00:46:45,960 --> 00:46:49,440
These are regular occurrences.

716
00:46:49,440 --> 00:46:54,040
What we saw from the last
flu pandemic in 2009, which was

717
00:46:54,040 --> 00:46:56,480
a relatively mild disease,

718
00:46:56,480 --> 00:47:00,920
was that our health service
was almost unable to cope.

719
00:47:08,000 --> 00:47:13,040
In my mission to get the best data,
I need to involve people from

720
00:47:13,600 --> 00:47:15,600
all ages but especially the ones

721
00:47:15,600 --> 00:47:17,640
we know least about -
young people.

722
00:47:19,680 --> 00:47:23,000
Which is why I'm bringing in
some big guns to help spread

723
00:47:23,000 --> 00:47:25,760
word about the BBC Pandemic app...

724
00:47:27,080 --> 00:47:31,520
..masters of new media,
social media.

725
00:47:31,520 --> 00:47:34,800
These guys are a new breed
of marketers.

726
00:47:34,800 --> 00:47:36,360
They've got Twitter, Facebook

727
00:47:36,360 --> 00:47:39,200
and Instagram wrapped around
their little finger,

728
00:47:39,200 --> 00:47:44,120
and hopefully they're going to help
me make the BBC Pandemic go viral.

729
00:47:44,120 --> 00:47:46,720
CHATTER

730
00:47:46,720 --> 00:47:50,160
If everyone in this room looks a bit
on the young side, it's because,

731
00:47:50,160 --> 00:47:54,600
I'm told, the average age
of people here is 23.

732
00:47:54,600 --> 00:47:59,400
Now, that means that a lot of these
people were born after 1995 and

733
00:47:59,400 --> 00:48:04,040
really are the first generation that
was raised in the smartphone era.

734
00:48:04,040 --> 00:48:07,040
Now, given that this is the group
that I need to target,

735
00:48:07,040 --> 00:48:09,240
who better to ask for help?

736
00:48:10,800 --> 00:48:15,040
The team here are fabulously
well connected on Facebook,

737
00:48:15,040 --> 00:48:19,400
with about three million young adult
followers in the UK alone.

738
00:48:19,400 --> 00:48:22,040
Hi, guys. We're now here
in the city centre today trying out

739
00:48:22,040 --> 00:48:24,640
the BBC's new Pandemic app,
which could potentially...

740
00:48:24,640 --> 00:48:27,280
A specially crafted video
targeting this group has been

741
00:48:27,280 --> 00:48:30,920
sitting on their Facebook page
for the last five hours.

742
00:48:30,920 --> 00:48:33,320
But has anyone bitten?

743
00:48:33,320 --> 00:48:37,680
So, on Facebook we've reached
1,015,750 people.

744
00:48:37,680 --> 00:48:40,200
OK, so what does that mean
by "reached"?

745
00:48:40,200 --> 00:48:42,400
Does that mean it came up
in their timelines?

746
00:48:42,400 --> 00:48:46,360
That means that 1,015,750 people
have actually seen that video.

747
00:48:46,360 --> 00:48:48,040
Not necessarily watched
the whole video,

748
00:48:48,040 --> 00:48:50,160
but they have seen that video
on their timeline.

749
00:48:50,160 --> 00:48:53,520
So, how many people watched it?
How many people watched it?

750
00:48:53,520 --> 00:48:55,400
So, how many people
actually watched it?

751
00:48:55,400 --> 00:48:59,840
That is 263,885 video views.

752
00:48:59,840 --> 00:49:02,080
VOICEOVER: Watching the video
is one thing,

753
00:49:02,080 --> 00:49:05,400
but how many people actually
followed the link to the app?

754
00:49:05,400 --> 00:49:08,760
So, just on Android devices...
OK.

755
00:49:08,760 --> 00:49:13,640
..we've got 1,473.
But then on Apple... Mm-hm...

756
00:49:13,640 --> 00:49:15,400
..we've got...

757
00:49:15,400 --> 00:49:19,480
..a lot higher, 4,340.

758
00:49:19,480 --> 00:49:23,200
That's amazing! So we have got
pushing 6,000 people on there...

759
00:49:23,200 --> 00:49:27,600
Pushing 6,000. Yeah, yeah.
..in five hours. Five hours, yeah.

760
00:49:27,600 --> 00:49:30,600
I am actually deeply impressed.
SHE LAUGHS

761
00:49:30,600 --> 00:49:33,640
I'm deeply impressed.
That is genuinely amazing.

762
00:49:33,640 --> 00:49:35,800
ON VIDEO: It sounds amazing!
That's a gamechanger.

763
00:49:35,800 --> 00:49:39,200
The more people that download it and
use it, the more effective it is.

764
00:49:41,680 --> 00:49:45,920
Social media and maths
are unlikely weapons,

765
00:49:45,920 --> 00:49:49,720
but they are essential
in the fight against a pandemic.

766
00:49:49,720 --> 00:49:52,400
Despite all the wonders
of modern medicine,

767
00:49:52,400 --> 00:49:57,400
another Spanish-flu-like virus
could still bring society to a halt.

768
00:49:59,600 --> 00:50:02,560
You may have thought that the huge
number of fatalities seen

769
00:50:02,560 --> 00:50:05,240
during the Spanish flu pandemic
is something that we could

770
00:50:05,240 --> 00:50:06,880
relegate to a bygone era.

771
00:50:06,880 --> 00:50:08,680
And that's partially true,

772
00:50:08,680 --> 00:50:11,440
with modern medicines
such as antibiotics.

773
00:50:11,440 --> 00:50:15,160
Antibiotics would be useful
for the secondary bacterial

774
00:50:15,160 --> 00:50:18,880
infections that flu victims often
suffer and that often kill them,

775
00:50:18,880 --> 00:50:21,880
ones like bacterial pneumonia.

776
00:50:21,880 --> 00:50:25,520
But antibiotics
would not be the cure-all,

777
00:50:25,520 --> 00:50:28,280
because they do not work
against viruses.

778
00:50:31,320 --> 00:50:35,720
Antibiotics treat bacterial
infections by interfering

779
00:50:35,720 --> 00:50:39,280
with the processes
that keep bacteria alive.

780
00:50:41,400 --> 00:50:44,600
But viruses aren't alive.

781
00:50:44,600 --> 00:50:47,560
They're just bags
containing genetic material...

782
00:50:51,560 --> 00:50:55,560
..which means antibiotics
are powerless against viruses.

783
00:50:55,560 --> 00:50:57,640
What you need are vaccines.

784
00:51:00,520 --> 00:51:03,640
Every autumn, thousands of us

785
00:51:03,640 --> 00:51:07,880
line up to get a seasonal flu jab
that's tailor-made each year.

786
00:51:09,320 --> 00:51:13,360
The vaccine is designed
to produce antibodies that block

787
00:51:13,360 --> 00:51:17,280
the spikes of the specific
flu viruses circulating that winter.

788
00:51:19,040 --> 00:51:23,120
But the problem with a pandemic
flu virus is you only know about it

789
00:51:23,120 --> 00:51:26,080
when it's started making people ill.

790
00:51:26,080 --> 00:51:28,520
And designing a vaccine takes time.

791
00:51:32,640 --> 00:51:35,840
I need to find out how much time
so we can factor

792
00:51:35,840 --> 00:51:39,440
that into our predictions about
the spread of the BBC Pandemic.

793
00:51:42,120 --> 00:51:45,920
So I've come to one of only
two factories in the UK that

794
00:51:45,920 --> 00:51:47,560
produce flu vaccine.

795
00:51:49,600 --> 00:51:54,280
At the moment, it's busy pumping out
millions of doses of seasonal

796
00:51:54,280 --> 00:51:56,200
flu vaccine needed each year.

797
00:51:56,200 --> 00:51:58,760
So, I'm going to take you down
this area here.

798
00:51:58,760 --> 00:52:00,880
All of them rely on these.

799
00:52:00,880 --> 00:52:03,240
This is where the eggs
come in from the farms.

800
00:52:04,440 --> 00:52:07,840
Dr Laura O'Brien
is in charge of this site.

801
00:52:13,400 --> 00:52:16,240
Laura, why are we using eggs
to make vaccines?

802
00:52:16,240 --> 00:52:20,280
So, we've used eggs for decades
to manufacture influenza vaccine.

803
00:52:20,280 --> 00:52:24,200
Viruses need living cells to grow.
And if you think about a single egg,

804
00:52:24,200 --> 00:52:26,680
there's lots of living cells
in there,

805
00:52:26,680 --> 00:52:29,600
it's protected by a shell,
so it's almost a clean environment,

806
00:52:29,600 --> 00:52:34,520
and it's just the perfect place
for a virus to propagate and grow.

807
00:52:34,520 --> 00:52:36,680
There are a huge number
of eggs here.

808
00:52:36,680 --> 00:52:38,400
How many do you handle in a day?

809
00:52:38,400 --> 00:52:41,280
Every single day, you know,
Monday through to Sunday,

810
00:52:41,280 --> 00:52:43,040
here in the plant we're putting

811
00:52:43,040 --> 00:52:46,280
over half a million eggs
through the factory.

812
00:52:46,280 --> 00:52:47,960
So, the eggs have been incubated...

813
00:52:47,960 --> 00:52:52,840
Right now, this factory is making
only seasonal flu vaccine.

814
00:52:52,840 --> 00:52:55,880
Each fertilised egg
is inoculated with flu virus,

815
00:52:55,880 --> 00:52:58,280
which then replicates.

816
00:52:58,280 --> 00:53:03,360
After just a few days, the eggs
are packed with the virus particles.

817
00:53:03,960 --> 00:53:08,240
Then all those virus particles
are inactivated to make the vaccine,

818
00:53:08,240 --> 00:53:10,560
so they're powerless to cause flu

819
00:53:10,560 --> 00:53:14,520
but can still fool the human body
into making antibodies.

820
00:53:14,520 --> 00:53:16,760
We put them in different incubators

821
00:53:16,760 --> 00:53:19,720
so not all the one day's batch
goes the same.

822
00:53:19,720 --> 00:53:23,440
When it comes to making
flu vaccine, you first need to

823
00:53:23,440 --> 00:53:27,000
know which strain of the flu virus
to grow inside the eggs.

824
00:53:29,320 --> 00:53:33,640
A global monitoring system run by
the World Health Organization

825
00:53:33,640 --> 00:53:36,760
constantly watches out
for animal flu viruses

826
00:53:36,760 --> 00:53:39,560
that have hopped into humans.

827
00:53:39,560 --> 00:53:41,640
It's the WHO that declares

828
00:53:41,640 --> 00:53:44,960
if one of these pandemic viruses
is on the loose.

829
00:53:49,640 --> 00:53:54,680
Laura, presumably when a pandemic
is declared, production changes.

830
00:53:55,000 --> 00:53:57,600
When the strain's declared,
that's the Big Bang.

831
00:53:57,600 --> 00:54:00,480
When the pandemic's declared,
here at the site

832
00:54:00,480 --> 00:54:04,080
we have a pandemic response plan
that we immediately enact.

833
00:54:06,040 --> 00:54:11,040
It's a race against time
to isolate and identify the pandemic

834
00:54:11,200 --> 00:54:15,280
virus so that a vaccine can be
created and put into production.

835
00:54:16,920 --> 00:54:20,360
It really is all hands on deck,
then. There's urgency.

836
00:54:20,360 --> 00:54:24,800
How long can you expect
from that get-go moment

837
00:54:24,800 --> 00:54:26,960
to produce your first
batch of vaccines?

838
00:54:26,960 --> 00:54:28,840
What we see is,
for a pandemic strain,

839
00:54:28,840 --> 00:54:32,560
generally in industry it takes about
four months from that strain to

840
00:54:32,560 --> 00:54:37,080
actually getting those first doses
out there to patients worldwide.

841
00:54:39,000 --> 00:54:43,560
At least four months
before a pandemic vaccine

842
00:54:43,560 --> 00:54:47,800
is tested and available.
That's a sobering thought.

843
00:54:49,240 --> 00:54:51,880
So just how many of us
could become infected

844
00:54:51,880 --> 00:54:54,800
before a vaccine is available?

845
00:54:54,800 --> 00:54:58,320
We can use the results of
our national experiment to find out.

846
00:55:00,000 --> 00:55:03,840
We launched the nationwide
BBC Pandemic app a little over

847
00:55:03,840 --> 00:55:06,280
two months ago.

848
00:55:06,280 --> 00:55:09,120
We needed thousands of people
of different ages in different

849
00:55:09,120 --> 00:55:11,760
locations to catch our virus

850
00:55:11,760 --> 00:55:15,440
to get real detail on how people
move around the country

851
00:55:15,440 --> 00:55:17,280
and who they come into contact with.

852
00:55:22,320 --> 00:55:23,880
So, how did we do?

853
00:55:24,960 --> 00:55:29,240
Adam, when we met a couple of
months ago, you set me a challenge

854
00:55:29,240 --> 00:55:31,640
to go out and get
a data set for you.

855
00:55:31,640 --> 00:55:34,480
Just remind us of what it was
you were looking for.

856
00:55:34,480 --> 00:55:37,920
The aim was to get data on
social interactions and movements.

857
00:55:37,920 --> 00:55:40,480
We know these kind of things
are important for diseases,

858
00:55:40,480 --> 00:55:43,640
but, really, to have those things
combined and have a large number

859
00:55:43,640 --> 00:55:46,840
of participants would be really
valuable for predicting epidemics.

860
00:55:46,840 --> 00:55:48,920
And, Julia, you've got
the results, I know,

861
00:55:48,920 --> 00:55:52,360
of how many we actually managed.
We needed 10,000, right?

862
00:55:52,360 --> 00:55:55,200
That was the number that you set.
Are you ready? Yeah.

863
00:55:57,520 --> 00:56:02,520
He-hey! That's brilliant! So, almost
29,000. Wow, that is impressive.

864
00:56:03,120 --> 00:56:05,760
And that's proper, full, rich,

865
00:56:05,760 --> 00:56:08,000
glorious data for every single one
of those people.

866
00:56:08,000 --> 00:56:10,560
Yep, these are all users
who've used the app, and we've got

867
00:56:10,560 --> 00:56:13,960
information on movements,
locations, the contact survey,

868
00:56:13,960 --> 00:56:15,440
all the bits of it together.

869
00:56:15,440 --> 00:56:18,640
Is that data spread across different
age groups and different areas?

870
00:56:18,640 --> 00:56:21,360
Yeah. This is one of the important
things you wanted, wasn't it?

871
00:56:21,360 --> 00:56:24,440
So, I can show you
how it's distributed by age.

872
00:56:24,440 --> 00:56:26,280
And here we go.

873
00:56:30,000 --> 00:56:32,960
This is fantastic. We've got
the younger age groups, we've got

874
00:56:32,960 --> 00:56:35,760
the kind of young adults,
right up to much older people.

875
00:56:35,760 --> 00:56:40,280
The previous data set was mainly
people over 35, is that right? Yeah.

876
00:56:40,280 --> 00:56:42,560
One of the challenges with
particularly survey data is

877
00:56:42,560 --> 00:56:44,840
you get much more participation
in older age groups.

878
00:56:44,840 --> 00:56:48,200
To have these young groups, young
professionals, is just fantastic.

879
00:56:48,200 --> 00:56:50,880
We don't have this
in this scale anywhere else.

880
00:56:52,640 --> 00:56:57,200
Our app has generated a massive
set of data about the movements

881
00:56:57,200 --> 00:57:01,480
of men and women of different ages,
living in cities, towns

882
00:57:01,480 --> 00:57:04,520
and rural areas
all over the country

883
00:57:04,520 --> 00:57:09,560
and how much they mix together.
It's revealed new information.

884
00:57:10,280 --> 00:57:14,760
We now know the over-60s
make more, longer journeys

885
00:57:14,760 --> 00:57:16,680
than people under 24.

886
00:57:17,680 --> 00:57:20,120
And country folk spend more time

887
00:57:20,120 --> 00:57:22,840
further away from home
than city dwellers.

888
00:57:24,760 --> 00:57:27,880
All this information
improves the calculations

889
00:57:27,880 --> 00:57:30,560
and the accuracy
of pandemic predictions.

890
00:57:31,800 --> 00:57:35,960
But before we predict the spread
of a deadly virus through the UK,

891
00:57:35,960 --> 00:57:40,400
we still need one more piece
of information for the model -

892
00:57:40,400 --> 00:57:44,080
how easily the flu virus can spread.

893
00:57:44,080 --> 00:57:47,560
One of the ways we can tailor models
to a specific infection is

894
00:57:47,560 --> 00:57:52,520
to use a value known as R0,
and this measures, for a specific

895
00:57:52,800 --> 00:57:56,600
infectious person, on average
how many infections they generate.

896
00:57:56,600 --> 00:57:59,760
So how many people
one person goes on to infect.

897
00:57:59,760 --> 00:58:02,760
Exactly. If you drop one person
in a susceptible population,

898
00:58:02,760 --> 00:58:04,880
how many new infections
will they create?

899
00:58:04,880 --> 00:58:07,280
And that's different for
lots of different diseases.

900
00:58:07,280 --> 00:58:10,360
So, you know, measles, Ebola
are ones I've got in my head, but...

901
00:58:10,360 --> 00:58:13,760
Exactly. So for Ebola
in West Africa in the early stages,

902
00:58:13,760 --> 00:58:16,200
it was around two in some areas.

903
00:58:16,200 --> 00:58:19,240
So each case on average was
generating two additional cases.

904
00:58:19,240 --> 00:58:23,240
But for measles it's much higher.
It can be 15 or 20 in some...

905
00:58:23,240 --> 00:58:26,240
So one person can infect
20 people before...?

906
00:58:26,240 --> 00:58:28,800
If it hits a susceptible
population, exactly, yeah.

907
00:58:28,800 --> 00:58:31,200
That's why vaccination's
important in that one.

908
00:58:31,200 --> 00:58:34,360
So, what number did you
actually use for flu?

909
00:58:34,360 --> 00:58:38,600
We looked at past pandemics,
and the typical range is 1.5 to two.

910
00:58:38,600 --> 00:58:42,240
And what we've done for the BBC
Pandemic is we've got an R0 of 1.8.

911
00:58:42,240 --> 00:58:45,040
OK. And that's not massively high.

912
00:58:45,040 --> 00:58:48,280
That's not crazy high
for a pandemic.

913
00:58:48,280 --> 00:58:51,040
It's a reasonable scenario
to be planning for.

914
00:58:53,880 --> 00:58:56,600
This is what we've been waiting for.

915
00:58:56,600 --> 00:58:59,440
Thanks to nearly 30,000 app users,

916
00:58:59,440 --> 00:59:03,080
we've now got the best data
ever available for our team

917
00:59:03,080 --> 00:59:06,040
to use in their calculations.

918
00:59:06,040 --> 00:59:10,200
How many people in the UK does
the mathematical model predict

919
00:59:10,200 --> 00:59:12,000
could become infected

920
00:59:12,000 --> 00:59:16,960
if a Spanish-flu-like pandemic
began in Haslemere today?

921
00:59:16,960 --> 00:59:18,840
So, Haslemere is
glowing in red here,

922
00:59:18,840 --> 00:59:21,520
just so we can see where it is.
As well as seeing the spread,

923
00:59:21,520 --> 00:59:25,280
which will be the dots going
towards red as the cases increase,

924
00:59:25,280 --> 00:59:29,200
we'll also see a bar with the total
number of infections. OK. Ready?

925
00:59:32,680 --> 00:59:35,200
So it hits London
very quickly, then.

926
00:59:36,200 --> 00:59:38,840
And then we need to zoom out.
And we're already zooming out.

927
00:59:38,840 --> 00:59:40,920
Look at it,
literally sweeping through.

928
00:59:40,920 --> 00:59:44,560
We had the Glasgow and Belfast area
go off before Edinburgh.

929
00:59:52,040 --> 00:59:55,360
43 million people?!

930
00:59:55,360 --> 00:59:57,280
And that number is extraordinary.

931
00:59:59,400 --> 01:00:04,160
43.3 million, which is about
just over two thirds of the country.

932
01:00:04,160 --> 01:00:07,240
Based on data from
thousands of app users,

933
01:00:07,240 --> 01:00:12,000
the model predicts very few parts of
the country will escape infection.

934
01:00:12,000 --> 01:00:15,240
Only the most remote corners
remain blue.

935
01:00:16,280 --> 01:00:20,280
But the prediction of the number
of deaths is even more shocking.

936
01:00:24,400 --> 01:00:28,640
By the end of May 1919,
the death toll from Spanish flu

937
01:00:28,640 --> 01:00:33,680
had risen to over 150,000
in England and Wales alone.

938
01:00:33,920 --> 01:00:38,560
Globally, it's thought at least
2% of people affected died.

939
01:00:40,520 --> 01:00:45,560
Based on the worst-case scenario of
43 million infected nationwide and

940
01:00:46,200 --> 01:00:51,200
then falling sick, a 2% death rate
today would be utterly catastrophic.

941
01:00:52,640 --> 01:00:57,040
My word! That's an enormous number.
That IS an enormous number.

942
01:00:57,040 --> 01:01:00,760
So, in the worst, worst case,
that's the number

943
01:01:00,760 --> 01:01:05,800
of deaths that we could expect based
on our data in a future pandemic.

944
01:01:06,840 --> 01:01:11,880
VOICEOVER: Over 800,000 dead,
millions more infected.

945
01:01:12,280 --> 01:01:14,880
And there's one more piece
of shocking information that

946
01:01:14,880 --> 01:01:19,840
our calculations can reveal - how
quickly those deaths will happen.

947
01:01:19,840 --> 01:01:23,400
Over what period are all these
deaths going to be seen?

948
01:01:23,400 --> 01:01:26,280
Yeah, so, we can have
a look at the timescale,

949
01:01:26,280 --> 01:01:28,360
and I can show you that here.

950
01:01:28,360 --> 01:01:30,120
We're going to go through, and we're

951
01:01:30,120 --> 01:01:33,960
going to see with the time up there,
so you've got the day up there.

952
01:01:33,960 --> 01:01:36,080
So we're back to Day 1,
back to Haslemere.

953
01:01:38,360 --> 01:01:40,640
So, this is one month in.
Gosh. London.

954
01:01:42,760 --> 01:01:44,360
Two months in, you can see

955
01:01:44,360 --> 01:01:47,680
it's made it up to Manchester,
Edinburgh, Belfast.

956
01:01:47,680 --> 01:01:51,640
You can kind of see it moving. But
there's a few long-distance numbers.

957
01:01:51,640 --> 01:01:53,920
Yeah, it jumps over
to Northern Ireland.

958
01:01:53,920 --> 01:01:56,760
Yeah, I mean, that was just by
about Day 60, two months, wasn't it,

959
01:01:56,760 --> 01:01:58,800
even well before it got
to Aberdeen there?

960
01:01:58,800 --> 01:02:01,640
You don't have very long, do you?
No. That's really frightening.

961
01:02:01,640 --> 01:02:04,480
But presumably... I mean,
you've got a timescale there,

962
01:02:04,480 --> 01:02:07,960
but presumably pandemic flu
isn't going to arise for the first

963
01:02:07,960 --> 01:02:10,160
time on this planet...
it's not likely to be Haslemere.

964
01:02:10,160 --> 01:02:11,880
It'll be across the globe somewhere.

965
01:02:11,880 --> 01:02:14,520
Yeah, chances are it will start
somewhere else in the world.

966
01:02:14,520 --> 01:02:16,680
Based on what we know
about these kind of viruses,

967
01:02:16,680 --> 01:02:19,520
it may well be somewhere in
rural South-East Asia, for example.

968
01:02:19,520 --> 01:02:21,560
And what sort of lead time
might we have on that,

969
01:02:21,560 --> 01:02:23,240
that sort of knowledge coming to us?

970
01:02:23,240 --> 01:02:25,560
Potentially, if it emerges there,
we'd get, say,

971
01:02:25,560 --> 01:02:28,400
two to four weeks before it
actually arrived in the UK.

972
01:02:28,400 --> 01:02:29,840
OK, let's call that a month.

973
01:02:29,840 --> 01:02:32,760
And then we've just seen it takes
just about two months to get

974
01:02:32,760 --> 01:02:36,760
from right near the south coast all
the way up to Aberdeen and Belfast.

975
01:02:36,760 --> 01:02:40,560
That's three months.
And I learnt that it takes at least

976
01:02:40,560 --> 01:02:44,520
four months to get
pandemic flu vaccine out.

977
01:02:44,520 --> 01:02:46,320
So there's a real gap there.

978
01:02:46,320 --> 01:02:49,800
Our data revealed
an alarming reality.

979
01:02:50,920 --> 01:02:54,200
We forecast over 800,000 could die

980
01:02:54,200 --> 01:02:57,440
if a pandemic spreads unchecked
throughout the UK.

981
01:02:57,440 --> 01:03:02,440
43 million could become infected
before a vaccine is available.

982
01:03:03,800 --> 01:03:07,280
Until the precise nature
of the virus shows itself,

983
01:03:07,280 --> 01:03:11,000
we won't know how many
will actually fall sick.

984
01:03:11,000 --> 01:03:15,000
But government plans assume
half of us will be affected.

985
01:03:16,640 --> 01:03:20,480
It is truly frightening.
What would any of us do faced with

986
01:03:20,480 --> 01:03:24,960
the prospect of a pandemic virus
sweeping towards our hometown?

987
01:03:27,200 --> 01:03:32,240
The gravity of the situation
has made our volunteers reflective.

988
01:03:32,520 --> 01:03:36,400
So, if we caught a deadly virus,
what would we do differently?

989
01:03:36,400 --> 01:03:39,400
Do you think we should all come
to this place? No. What would we do?

990
01:03:39,400 --> 01:03:43,240
Stay at home? Stay at home?
Why would we stay at home?

991
01:03:43,240 --> 01:03:44,960
So no-one else catches the virus.

992
01:03:44,960 --> 01:03:49,080
John and I are just
driving up to north London.

993
01:03:49,080 --> 01:03:50,920
We were just thinking
about whether...

994
01:03:52,200 --> 01:03:55,440
..if there was a pandemic, whether
we'd be doing this, and the answer,

995
01:03:55,440 --> 01:03:59,480
we both agree, is no.
We'd be staying at home.

996
01:04:00,480 --> 01:04:02,520
If you thought it was
that infectious that it could

997
01:04:02,520 --> 01:04:04,520
travel really quickly,
you'd want to stay put,

998
01:04:04,520 --> 01:04:06,680
because you wouldn't
want to give it to anyone else.

999
01:04:06,680 --> 01:04:09,120
But you wouldn't want to be around
anyone that might have it.

1000
01:04:09,120 --> 01:04:12,000
If you found out now,
it's happened, it's happening now,

1001
01:04:12,000 --> 01:04:14,120
was the first thing to do
to get back to your kids?

1002
01:04:14,120 --> 01:04:18,120
Yes, on the one hand, but then,
turn that on the head,

1003
01:04:18,120 --> 01:04:21,640
if there's a risk that
you're infected and they're not...

1004
01:04:21,640 --> 01:04:26,160
Yeah, so you're like patient zero.
..do you go home? You tell me. Mm.

1005
01:04:28,640 --> 01:04:33,640
When a pandemic arrives in the UK,
the NHS will be on the front line.

1006
01:04:35,280 --> 01:04:40,040
Dr Chloe Sellwood leads
pandemic planning for NHS England.

1007
01:04:42,720 --> 01:04:45,760
We do plan for the reasonable
worst-case scenario,

1008
01:04:45,760 --> 01:04:49,840
so we look at what 1918 was like
and we look at those numbers,

1009
01:04:49,840 --> 01:04:54,120
and we translate them across to now
and we think about how

1010
01:04:54,120 --> 01:04:57,000
that could impact on the NHS, and
that's where modelling comes in.

1011
01:04:57,000 --> 01:04:59,080
That gives us some numbers.

1012
01:04:59,080 --> 01:05:03,480
And that helps us to think about
how the NHS might be affected

1013
01:05:03,480 --> 01:05:06,560
and how the NHS might
need to respond differently.

1014
01:05:06,560 --> 01:05:10,160
So if you think about 50%
of the population being

1015
01:05:10,160 --> 01:05:13,520
affected in the reasonable
worst-case scenario,

1016
01:05:13,520 --> 01:05:17,080
if all of those cases happened
in one week, that's a really big,

1017
01:05:17,080 --> 01:05:19,520
short, sharp peak, lots of cases.

1018
01:05:19,520 --> 01:05:24,280
If we can slow the spread so they
happen over a much longer period,

1019
01:05:24,280 --> 01:05:28,800
that reduces the impact on the NHS
and makes it much more manageable.

1020
01:05:28,800 --> 01:05:32,600
Yes, we're seeing a pandemic
for a longer period of time,

1021
01:05:32,600 --> 01:05:34,560
but it's fewer patients
at any one time.

1022
01:05:34,560 --> 01:05:37,520
And of course, therefore,
then increase the likelihood

1023
01:05:37,520 --> 01:05:40,040
of the NHS being able to cope.
Yes, exactly.

1024
01:05:44,160 --> 01:05:49,160
We need ways to slow down a pandemic
when one inevitably hits,

1025
01:05:49,560 --> 01:05:53,560
and this is where
mathematical modelling can come in.

1026
01:05:53,560 --> 01:05:57,680
It can provide hard numbers
to help guide decisions,

1027
01:05:57,680 --> 01:06:01,560
because the long-term closure
of schools or public transport

1028
01:06:01,560 --> 01:06:04,840
would have an enormous
impact on society.

1029
01:06:04,840 --> 01:06:08,600
Interventions like that
can't be taken lightly.

1030
01:06:08,600 --> 01:06:12,720
But our data can be used
to test a far simpler idea, too.

1031
01:06:14,560 --> 01:06:18,800
The Government have a key message
during any major outbreak.

1032
01:06:18,800 --> 01:06:21,680
They tell us to wash our hands.

1033
01:06:21,680 --> 01:06:24,680
But what difference
would it make if we did?

1034
01:06:24,680 --> 01:06:27,680
Could something as simple
as hand-washing make any real

1035
01:06:27,680 --> 01:06:29,360
dent on the death toll?

1036
01:06:30,600 --> 01:06:35,520
We've asked Julia to try something
that has never been done before.

1037
01:06:35,520 --> 01:06:37,280
She's used our data to simulate

1038
01:06:37,280 --> 01:06:41,520
the impact hand-washing could have
in a major flu outbreak.

1039
01:06:45,880 --> 01:06:49,960
What we can model here is the effect
of EVERYBODY - and again, we're

1040
01:06:49,960 --> 01:06:53,920
assuming the best-case scenario -
everybody takes extra precautions.

1041
01:06:53,920 --> 01:06:56,480
That means washing hands more often,

1042
01:06:56,480 --> 01:06:59,640
certainly after meeting infectious
people, but maybe washing your

1043
01:06:59,640 --> 01:07:02,960
hands, like, five to ten times
more than normal per day,

1044
01:07:02,960 --> 01:07:05,160
and everyone does it
for the full period.

1045
01:07:05,160 --> 01:07:07,880
And we can look at the effect
that has on transmission. OK.

1046
01:07:10,120 --> 01:07:13,760
So, on the left is the one
we've seen before. Right.

1047
01:07:13,760 --> 01:07:15,640
There's nothing on the right yet!

1048
01:07:15,640 --> 01:07:16,880
Look at the numbers!

1049
01:07:17,920 --> 01:07:21,480
So now, by two months, actually,

1050
01:07:21,480 --> 01:07:23,920
only London is really affected.

1051
01:07:23,920 --> 01:07:26,200
Do you know... Do you think

1052
01:07:26,200 --> 01:07:29,360
it's going to reach the final same
number but it's just slower?

1053
01:07:29,360 --> 01:07:31,720
No? Overall. Well,
it doesn't look like...

1054
01:07:31,720 --> 01:07:34,480
I mean, it's slowing down.
That's a good question.

1055
01:07:34,480 --> 01:07:35,920
So, two things happen there.

1056
01:07:35,920 --> 01:07:37,800
It's definitely slower in spread,

1057
01:07:37,800 --> 01:07:40,880
and the total number is...what?

1058
01:07:40,880 --> 01:07:43,440
Lower.
It's only...three quarters?

1059
01:07:43,440 --> 01:07:45,960
Yeah, three quarters
of what you had before.

1060
01:07:45,960 --> 01:07:49,800
That's 13 million people
who don't have the virus.

1061
01:07:49,800 --> 01:07:53,440
I mean, that's quite a big,
persuasive message there, isn't it,

1062
01:07:53,440 --> 01:07:54,880
just washing your hands?

1063
01:07:54,880 --> 01:07:59,280
But, I mean, we're talking about
genuinely millions of people

1064
01:07:59,280 --> 01:08:02,440
and potentially
hundreds of thousands of lives.

1065
01:08:02,440 --> 01:08:05,480
That's right.
There's 13 million cases averted.

1066
01:08:05,480 --> 01:08:07,000
That's a lot of lives.

1067
01:08:09,080 --> 01:08:12,240
If we all commit
to changing our behaviour,

1068
01:08:12,240 --> 01:08:16,120
there ARE ways to slow
the spread of a pandemic.

1069
01:08:16,120 --> 01:08:19,520
But the real hope
for the future is prevention,

1070
01:08:19,520 --> 01:08:23,320
and that's something that's being
worked on in labs around the world.

1071
01:08:26,760 --> 01:08:30,360
One of those labs
is at University College London.

1072
01:08:30,360 --> 01:08:34,640
It's run by biochemical engineer
Dr Tarit Mukhapadhyay.

1073
01:08:40,160 --> 01:08:44,680
Researchers like Tarit are working
on radically new ways for making

1074
01:08:44,680 --> 01:08:47,520
enormous quantities
of flu vaccine quickly.

1075
01:08:48,760 --> 01:08:53,600
Instead of eggs, Tarit
is growing vaccine in yeast cells

1076
01:08:53,600 --> 01:08:56,640
in a steel vat called a bioreactor.

1077
01:08:56,640 --> 01:08:58,960
The bigger the vat, the more quickly

1078
01:08:58,960 --> 01:09:02,520
mass doses of vaccine
can be produced.

1079
01:09:02,520 --> 01:09:05,560
But there's an even bigger
revolution in the offing.

1080
01:09:06,680 --> 01:09:10,240
Well, the Holy Grail is actually try
and create a universal flu vaccine.

1081
01:09:10,240 --> 01:09:13,600
Right. And so this is a vaccine
you don't need to constantly update

1082
01:09:13,600 --> 01:09:16,440
every year, or indeed
wait for a pandemic to emerge,

1083
01:09:16,440 --> 01:09:18,240
before you can create a vaccine.

1084
01:09:19,840 --> 01:09:23,440
The coats of flu viruses
constantly change,

1085
01:09:23,440 --> 01:09:27,920
meaning a new vaccine
is needed for each one.

1086
01:09:27,920 --> 01:09:32,920
But we now know some bits of the
flu virus don't change so easily.

1087
01:09:33,080 --> 01:09:37,040
Make a vaccine that targets
those stable bits of the virus

1088
01:09:37,040 --> 01:09:40,040
and you've got some protection
ready and waiting

1089
01:09:40,040 --> 01:09:42,840
when a pandemic virus emerges.

1090
01:09:42,840 --> 01:09:45,880
Rather than constantly having to
update the vaccine,

1091
01:09:45,880 --> 01:09:50,040
a vaccine that is based upon
the universal system may be good

1092
01:09:50,040 --> 01:09:53,040
enough for five years, and I guess
the Holy Grail is making

1093
01:09:53,040 --> 01:09:54,960
a vaccine that's
good enough for ten years.

1094
01:09:54,960 --> 01:09:57,680
I mean, we will have to
update it at some stage,

1095
01:09:57,680 --> 01:10:00,520
but it gives us a little more
breathing space and a little

1096
01:10:00,520 --> 01:10:04,160
more flexibility in trying to
deal with future pandemics.

1097
01:10:04,160 --> 01:10:07,360
That really does sound
incredible to achieve,

1098
01:10:07,360 --> 01:10:10,080
but how close are we to doing that?

1099
01:10:10,080 --> 01:10:13,440
Well, I guess in terms of getting a
universal vaccine out there, I mean,

1100
01:10:13,440 --> 01:10:17,920
we're still at least one decade
if not more years away.

1101
01:10:17,920 --> 01:10:21,560
I think getting a commercial
universal flu vaccine out there

1102
01:10:21,560 --> 01:10:23,080
is still a long way away.

1103
01:10:24,960 --> 01:10:28,680
Until we get there, our data
will put pandemic researchers

1104
01:10:28,680 --> 01:10:31,960
in the best position possible.

1105
01:10:31,960 --> 01:10:36,800
As well as their long-term
forecasts, good GPS and contact data

1106
01:10:36,800 --> 01:10:41,520
will be essential for them to
model accurately during a pandemic.

1107
01:10:44,720 --> 01:10:46,160
This would be incredibly useful,

1108
01:10:46,160 --> 01:10:49,600
just having this scale of data
but also having it by different

1109
01:10:49,600 --> 01:10:51,720
age group, by different
geographical locations.

1110
01:10:51,720 --> 01:10:54,160
It means we can really look at
things in much more detail

1111
01:10:54,160 --> 01:10:55,840
than has been possible before.

1112
01:10:55,840 --> 01:10:59,160
To have this kind of data
for, what, 20,000 people,

1113
01:10:59,160 --> 01:11:01,960
that's just incredible.
We dream of that kind of data!

1114
01:11:01,960 --> 01:11:03,760
You're welcome!

1115
01:11:08,640 --> 01:11:11,800
It began by recruiting a community.

1116
01:11:11,800 --> 01:11:14,880
..and that side, if
you have not been infected.

1117
01:11:14,880 --> 01:11:16,800
It's turned into the biggest

1118
01:11:16,800 --> 01:11:20,120
citizen science experiment
in pandemic research.

1119
01:11:20,120 --> 01:11:21,720
Yes, you heard that right,

1120
01:11:21,720 --> 01:11:24,520
an app could help save lives.
And here's how.

1121
01:11:24,520 --> 01:11:27,560
It's brought some harsh realities
into sharp focus.

1122
01:11:27,560 --> 01:11:31,160
My word! That's the number of deaths
that we could expect,

1123
01:11:31,160 --> 01:11:34,160
based on our data,
in a future pandemic.

1124
01:11:35,200 --> 01:11:40,200
The BBC Four Pandemic experiment
has created an invaluable legacy.

1125
01:11:45,440 --> 01:11:49,440
Thanks to tens of thousands
of you across the country,

1126
01:11:49,440 --> 01:11:54,440
we now have a new gold-standard
data set for pandemic research,

1127
01:11:54,800 --> 01:11:58,960
a vast amount of information that
will be shared with mathematicians

1128
01:11:58,960 --> 01:12:03,280
from the Department of Health
and pandemic researchers nationwide.

1129
01:12:06,800 --> 01:12:11,120
But the experiment isn't over yet.

1130
01:12:12,280 --> 01:12:16,160
The app is still available,
if you would like to join in.

1131
01:12:16,160 --> 01:12:20,360
And we still need
your movement and contact data,

1132
01:12:20,360 --> 01:12:25,320
which will be passed anonymously
to our maths team in Cambridge.

1133
01:12:28,520 --> 01:12:33,560
Better data equals better maths,
which equals better models,

1134
01:12:33,960 --> 01:12:37,000
and better models mean
better health care for all of us.

1135
01:12:38,280 --> 01:12:41,920
So, with that in mind, the maths
team have come up with a few

1136
01:12:41,920 --> 01:12:44,160
areas of our data set that could

1137
01:12:44,160 --> 01:12:47,000
still do with a little bit
of improvement.

1138
01:12:48,960 --> 01:12:52,640
They want more teenagers and
over-60s to download the app,

1139
01:12:52,640 --> 01:12:55,560
so please spread the word.

1140
01:12:55,560 --> 01:12:59,000
We also need to see
how our behaviour at weekends

1141
01:12:59,000 --> 01:13:02,040
compared to weekdays
changes the predictions,

1142
01:13:02,040 --> 01:13:06,320
so you can feel free to use the app
as many times as you like,

1143
01:13:06,320 --> 01:13:11,400
weekends AND weekdays, no matter
how mundane your plans are.

1144
01:13:12,000 --> 01:13:16,040
In fact, you can all still take part
in the biggest citizen science

1145
01:13:16,040 --> 01:13:21,120
experiment of its kind, unleashing
the life-saving power of mathematics

1146
01:13:22,240 --> 01:13:27,000
on what the Government considers to
be the greatest risk to our society.