1
00:00:00,060 --> 00:00:04,060
VO: Decades of research and technology for better forecasting of the when, where

2
00:00:04,080 --> 00:00:08,080
and how intense a hurricane will be? But what if we could

3
00:00:08,100 --> 00:00:12,100
predict a disease outbreak in the wake of a storm? That's the question

4
00:00:12,120 --> 00:00:16,120
some researchers asked about cholera in Haiti in the aftermath of Hurricane

5
00:00:16,140 --> 00:00:20,140
Matthew. Cholera is a water-borne infectious disease that

6
00:00:20,160 --> 00:00:24,170
occurs when a person ingests food or water contaminated with the Vibrio

7
00:00:24,190 --> 00:00:28,190
bacterium. Cholera causes severe diarrhea, nausea,

8
00:00:28,210 --> 00:00:32,210
vomiting and dehydration and can lead to death if untreated.

9
00:00:32,230 --> 00:00:36,250
Researchers estimate that hundreds of thousands of cases are reported each year

10
00:00:36,270 --> 00:00:40,330
worldwide.
Colwell: The bacterium is found in

11
00:00:40,350 --> 00:00:44,380
world oceans globally, especially

12
00:00:44,400 --> 00:00:48,420
in the temperate regions and in the tropics. So in the

13
00:00:48,440 --> 00:00:52,540
countries less developed with

14
00:00:52,560 --> 00:00:56,650
infrastructure that is not the equivalent, let's say, of Europe or the

15
00:00:56,670 --> 00:01:00,730
United States or Canada, then the population

16
00:01:00,750 --> 00:01:04,760
that has to rely on river water or pond water

17
00:01:04,780 --> 00:01:08,820
is at risk for cholera.
VO: In addition to water insecurity,

18
00:01:08,840 --> 00:01:12,860
high seasonal temperatures followed by extreme rainfall,

19
00:01:12,880 --> 00:01:16,900
concentrated populations and a natural disaster

20
00:01:16,920 --> 00:01:21,020
are all conditions conducive to a cholera epidemic. This was

21
00:01:21,040 --> 00:01:25,060
the case for Haiti in 2010.
Colwell: The data that we were able to

22
00:01:25,080 --> 00:01:29,080
pull together showed that in 2010

23
00:01:29,100 --> 00:01:33,090
it was the hottest summer in fifty years. And then as if

24
00:01:33,110 --> 00:01:37,110
that weren't enough, there was a hurricane that skirted

25
00:01:37,130 --> 00:01:41,130
the island, but it dumped the heaviest rainfall

26
00:01:41,150 --> 00:01:45,160
in fifty years.
Jutla: We tried to make an algorithm

27
00:01:45,180 --> 00:01:49,180
in a cohesive form to determine the risk. And then that basically

28
00:01:49,200 --> 00:01:53,210
provided us with the first clues on the risk of outbreak of cholera

29
00:01:53,230 --> 00:01:57,220
in Haiti after this earthquake.

30
00:01:57,240 --> 00:02:01,250
Then we used the same algorithm

31
00:02:01,270 --> 00:02:05,300
with improved satellite datasets from Global

32
00:02:05,320 --> 00:02:09,370
Precipitation Measurement mission after Hurricane Matthew struck

33
00:02:09,390 --> 00:02:13,390
that region again. And we were

34
00:02:13,410 --> 00:02:17,420
able to, in real time, predict the risk of cholera infection in human

35
00:02:17,440 --> 00:02:21,450
population at least four weeks in advance.

36
00:02:21,470 --> 00:02:25,500
We did the same thing for Yemen. We knew there was a mass movement of

37
00:02:25,520 --> 00:02:29,540
human population due to civil unrest in that part of the world, and then we had

38
00:02:29,560 --> 00:02:33,610
very heavy precipitation. And then we immediately started

39
00:02:33,630 --> 00:02:37,630
monitoring conditions. And that basically converged to give us a

40
00:02:37,650 --> 00:02:37,730
risk on where and when this disease will

41
00:02:37,750 --> 00:02:41,780
risk on where and when this disease will

42
00:02:41,800 --> 00:02:45,860
lock in on human population.
Colwell: I think we can predict and prevent

43
00:02:45,880 --> 00:02:49,970
and I'd like to see that happen very quickly, in the next three to

44
00:02:49,990 --> 00:02:54,150
five years, and I'd like to see the satellite system

45
00:02:54,170 --> 00:02:58,190
to be part of the regular public health tools so that

46
00:02:58,210 --> 00:03:02,230
we can do prediction as well as the

47
00:03:02,250 --> 00:03:06,250
tracking of epidemics that's done traditionally now.

48
00:03:06,270 --> 00:03:10,250
[ music ]

49
00:03:10,270 --> 00:03:11,532
[ music ]