What really happens when you mix medications? | Russ Altman

188,917 views ใƒป 2016-03-23

TED


ืื ื ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ืœืžื˜ื” ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ.

ืžืชืจื’ื: Michael Coslovsky ืžื‘ืงืจ: Zeeva Livshitz
00:12
So you go to the doctor and get some tests.
0
12811
3321
ืืชื ื”ื•ืœื›ื™ื ืœืจื•ืคื ื•ื”ื•ื ืจื•ืฉื ืœื›ื ื›ืžื” ื‘ื“ื™ืงื•ืช.
00:16
The doctor determines that you have high cholesterol
1
16674
2620
ื”ืจื•ืคื ืงื•ื‘ืข ืฉื™ืฉ ืœื›ื ื›ื•ืœืกื˜ืจื•ืœ ื’ื‘ื•ื”
00:19
and you would benefit from medication to treat it.
2
19318
3171
ื•ืฉื›ื“ืื™ ืœื›ื ืœืงื—ืช ืชืจื•ืคื” ื‘ืฉื‘ื™ืœ ื–ื”.
00:22
So you get a pillbox.
3
22981
1556
ืื– ืืชื ืžืงื‘ืœื™ื ืงื•ืคืกืช ื’ืœื•ืœื•ืช.
00:25
You have some confidence,
4
25505
1199
ื™ืฉ ืœื›ื ื‘ื™ื˜ื—ื•ืŸ ืžืกื•ื™ื,
00:26
your physician has some confidence that this is going to work.
5
26728
2937
ืœืจื•ืคื ืฉืœื›ื ื™ืฉ ื‘ื™ื˜ื—ื•ืŸ ืžืกื•ื™ื ืฉื–ื” ื™ืขืฉื” ืืช ื”ืขื‘ื•ื“ื”.
00:29
The company that invented it did a lot of studies, submitted it to the FDA.
6
29689
3553
ื”ื—ื‘ืจื” ืฉืคื™ืชื—ื” ืืช ื–ื” ืขืจื›ื” ืžื—ืงืจื™ื ืจื‘ื™ื,
ื”ื’ื™ืฉื” ืืช ื–ื” ืœืžื™ื ื”ืœ ื”ืžื–ื•ืŸ ื•ื”ืชืจื•ืคื•ืช,
ื”ื ื‘ื—ื ื• ืืช ื–ื” ื‘ื™ืกื•ื“ื™ื•ืช, ื‘ืกืคืงื ื•ืช. ื”ื ืื™ืฉืจื• ืืช ื–ื”.
00:33
They studied it very carefully, skeptically, they approved it.
7
33266
3107
00:36
They have a rough idea of how it works,
8
36397
1889
ื™ืฉ ืœื”ื ืžื•ืฉื’ ื›ืœืœื™ ืื™ืš ื–ื” ืขื•ื‘ื“,
00:38
they have a rough idea of what the side effects are.
9
38310
2453
ื™ืฉ ืœื”ื ืžื•ืฉื’ ื›ืœืœื™ ืžื” ื”ืŸ ืชื•ืคืขื•ืช ื”ืœื•ื•ืื™.
00:40
It should be OK.
10
40787
1150
ื–ื” ืฆืจื™ืš ืœื”ื™ื•ืช ื‘ืกื“ืจ.
00:42
You have a little more of a conversation with your physician
11
42864
2818
ืืชื ืžื ื”ืœื™ื ืฉื™ื—ื” ืงืฆืช ื™ื•ืชืจ ืžืขืžื™ืงื” ืขื ื”ืจื•ืคื ืฉืœื›ื
00:45
and the physician is a little worried because you've been blue,
12
45706
2963
ื•ื”ืจื•ืคื ืงืฆืช ืžื•ื“ืื’ ื›ื™ ืœืื—ืจื•ื ื” ื”ื™ื™ืชื ื‘ืžืฆื‘ ืจื•ื— ืจืข,
00:48
haven't felt like yourself,
13
48693
1293
ืœื ื”ืจื’ืฉืชื ื›ืจื’ื™ืœ,
ืœื ื”ืฆืœื—ืชื ืœื™ื”ื ื•ืช ืžื”ื“ื‘ืจื™ื ื‘ื—ื™ื™ื ื›ืžื• ื‘ื“ืจืš ื›ืœืœ.
00:50
you haven't been able to enjoy things in life quite as much as you usually do.
14
50010
3731
00:53
Your physician says, "You know, I think you have some depression.
15
53765
3186
ื”ืจื•ืคื ืื•ืžืจ, "ืืชื ื™ื•ื“ืขื™ื, ืื ื™ ื—ื•ืฉื‘ ืฉืืชื ืงืฆืช ื‘ื“ื™ื›ืื•ืŸ.
00:57
I'm going to have to give you another pill."
16
57792
2315
"ืื ื™ ืืฆื˜ืจืš ืœืชืช ืœื›ื ืขื•ื“ ื›ื“ื•ืจ."
01:00
So now we're talking about two medications.
17
60934
2483
ืขื›ืฉื™ื• ืื ื—ื ื• ื›ื‘ืจ ืžื“ื‘ืจื™ื ืขืœ ืฉืชื™ ืชืจื•ืคื•ืช.
01:03
This pill also -- millions of people have taken it,
18
63441
3104
ื’ื ื”ื›ื“ื•ืจ ื”ื–ื”-- ืžื™ืœื™ื•ื ื™ ืื ืฉื™ื ืœืงื—ื• ืื•ืชื•,
01:06
the company did studies, the FDA looked at it -- all good.
19
66569
3631
ื”ื—ื‘ืจื” ืขืฉืชื” ืžื—ืงืจื™ื, ื”ืžื™ื ื”ืœ ื‘ื—ืŸ ืื•ืชื• -- ื”ื›ืœ ื˜ื•ื‘.
01:10
Think things should go OK.
20
70823
2057
ืกื‘ื™ืจ ืฉื”ื›ืœ ื™ื”ื™ื” ื‘ืกื“ืจ.
01:12
Think things should go OK.
21
72904
2197
ืกื‘ื™ืจ ืฉื”ื›ืœ ื™ื”ื™ื” ื‘ืกื“ืจ.
01:15
Well, wait a minute.
22
75125
1439
ืื‘ืœ, ืจืง ืจื’ืข.
01:16
How much have we studied these two together?
23
76588
3517
ื›ืžื” ื—ืงืจื ื• ืืช ืฉืชื™ ืืœื• ื‘ื™ื—ื“?
01:20
Well, it's very hard to do that.
24
80630
2300
ืืช ื–ื” ืงืฉื” ืžืื“ ืœืขืฉื•ืช.
01:22
In fact, it's not traditionally done.
25
82954
2130
ืœืžืขืฉื”, ืœื ื ื•ื”ื’ื™ื ืœืขืฉื•ืช ื–ืืช.
01:25
We totally depend on what we call "post-marketing surveillance,"
26
85108
5518
ืื ื—ื ื• ืžืกืชืžื›ื™ื ื‘ืื•ืคืŸ ืžื•ื—ืœื˜ ืขืœ "ืžืขืงื‘ ืœืื—ืจ ื”ืฉื™ื•ื•ืง",
01:30
after the drugs hit the market.
27
90650
1880
ืื—ืจื™ ืฉื”ืชืจื•ืคื•ืช ื”ื•ืคืฆื• ืœืžื›ื™ืจื”.
01:32
How can we figure out if bad things are happening
28
92996
2848
ืื™ืš ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื’ืœื•ืช ืื ื“ื‘ืจื™ื ืœื ื˜ื•ื‘ื™ื ืงื•ืจื™ื
01:35
between two medications?
29
95868
1357
ื‘ื™ืŸ ืฉืชื™ ืชืจื•ืคื•ืช?
01:37
Three? Five? Seven?
30
97249
2030
ืฉืœื•ืฉ? ื—ืžืฉ? ืฉื‘ืข ืชืจื•ืคื•ืช?
01:39
Ask your favorite person who has several diagnoses
31
99708
2415
ืฉืืœื• ืืช ื”ืื“ื ื”ืื”ื•ื‘ ืขืœื™ื›ื ืฉื™ืฉ ืœื• ื›ืžื” ืื‘ื—ื•ื ื™ื
01:42
how many medications they're on.
32
102147
1834
ื›ืžื” ืชืจื•ืคื•ืช ื”ื•ื ืœื•ืงื— ื›ืจื’ืข.
01:44
Why do I care about this problem?
33
104530
1580
ืœืžื” ืื™ื›ืคืช ืœื™ ืžื”ื‘ืขื™ื” ื”ื–ื•? ืื™ื›ืคืช ืœื™ ืžืžื ื” ืžืื“.
01:46
I care about it deeply.
34
106134
1157
ืื ื™ ืžื™ื“ืขืŸ ื•ื—ื•ืงืจ ื ืชื•ื ื™ื, ื•ืœืžืขืŸ ื”ืืžืช, ืœื“ืขืชื™,
01:47
I'm an informatics and data science guy and really, in my opinion,
35
107315
4304
01:51
the only hope -- only hope -- to understand these interactions
36
111643
3745
ื”ืชืงื•ื•ื” ื”ื™ื—ื™ื“ื” -- ื”ื™ื—ื™ื“ื” -- ืœื”ื‘ื™ืŸ ืืช ื”ืื™ื ื˜ืจืืงืฆื™ื•ืช ื”ืœืœื•
01:55
is to leverage lots of different sources of data
37
115412
3056
ื”ื™ื ืœื ืฆืœ ื”ืจื‘ื” ืžืงื•ืจื•ืช ืฉืœ ื ืชื•ื ื™ื
01:58
in order to figure out when drugs can be used together safely
38
118492
3556
ื›ื“ื™ ืœืžืฆื•ื ืžืชื™ ื ื™ืชืŸ ืœื”ืฉืชืžืฉ ื‘ืชืจื•ืคื•ืช ื‘ื•-ื–ืžื ื™ืช ื‘ืื•ืคืŸ ื‘ื˜ื•ื—
02:02
and when it's not so safe.
39
122072
1777
ื•ืžืชื™ ื–ื” ืœื ื›ืœ ื›ืš ื‘ื˜ื•ื—.
02:04
So let me tell you a data science story.
40
124615
2051
ืื– ืชืจืฉื• ืœื™ ืœืกืคืจ ืœื›ื ืกื™ืคื•ืจ ืžืžื“ืขื™ ื”ื ืชื•ื ื™ื.
02:06
And it begins with my student Nick.
41
126690
2154
ื•ื”ื•ื ืžืชื—ื™ืœ ืขื ื”ืกื˜ื•ื“ื ื˜ ืฉืœื™, ื ื™ืง.
02:08
Let's call him "Nick," because that's his name.
42
128868
2380
ื ืงืจื ืœื• "ื ื™ืง", ื›ื™ ื–ื” ื”ืฉื ืฉืœื•.
02:11
(Laughter)
43
131272
1592
(ืฆื—ื•ืง)
02:12
Nick was a young student.
44
132888
1201
ื ื™ืง ื”ื™ื” ืกื˜ื•ื“ื ื˜ ืฆืขื™ืจ.
02:14
I said, "You know, Nick, we have to understand how drugs work
45
134113
3079
ืืžืจืชื™, "ืืชื” ื™ื•ื“ืข, ื ื™ืง, ืื ื—ื ื• ืฆืจื™ื›ื™ื ืœื”ื‘ื™ืŸ ืื™ืš ืชืจื•ืคื•ืช ืขื•ื‘ื“ื•ืช
02:17
and how they work together and how they work separately,
46
137216
2626
"ืื™ืš ื”ืŸ ืขื•ื‘ื“ื•ืช ื‘ื™ื—ื“, ื•ืื™ืš ื”ืŸ ืขื•ื‘ื“ื•ืช ื‘ื ืคืจื“,
02:19
and we don't have a great understanding.
47
139866
1922
"ื•ืื™ืŸ ืœื ื• ื”ื‘ื ื” ืžื™ ื™ื•ื“ืข ืžื”."
02:21
But the FDA has made available an amazing database.
48
141812
2405
ืื‘ืœ ืžื™ื ื”ืœ ื”ืžื–ื•ืŸ ื•ื”ืชืจื•ืคื•ืช ื”ื ื’ื™ืฉ ืžืื’ืจ ืžื™ื“ืข ืžื“ื”ื™ื.
02:24
It's a database of adverse events.
49
144241
1699
ื–ื” ืžืื’ืจ ืžื™ื“ืข ืฉืœ ืชื•ืคืขื•ืช ืฉืœื™ืœื™ื•ืช.
02:26
They literally put on the web --
50
146321
1642
ื”ื ืžืžืฉ ืžืขืœื™ื ืœืื™ื ื˜ืจื ื˜ --
02:27
publicly available, you could all download it right now --
51
147987
3119
ื ื’ื™ืฉ ืœืฆื™ื‘ื•ืจ, ื›ื•ืœื›ื ื™ื›ื•ืœื™ื ืœื”ื•ืจื™ื“ ืืช ื–ื” ืžืžืฉ ืขื›ืฉื™ื• --
02:31
hundreds of thousands of adverse event reports
52
151130
3627
ืžืื•ืช ืืœืคื™ ื“ื•"ื—ื•ืช ืฉืœ ืื™ืจื•ืขื™ ืชื•ืคืขื•ืช ืฉืœื™ืœื™ื•ืช
02:34
from patients, doctors, companies, pharmacists.
53
154781
3760
ืžืžื˜ื•ืคืœื™ื, ืจื•ืคืื™ื, ื—ื‘ืจื•ืช, ืจื•ืงื—ื™ื.
02:38
And these reports are pretty simple:
54
158565
1749
ื•ื”ื“ื•"ื—ื•ืช ื”ืืœื” ื”ื ืคืฉื•ื˜ื™ื ืœืžื“ื™:
02:40
it has all the diseases that the patient has,
55
160338
2658
ื›ืชื•ื‘ื•ืช ื‘ื”ื ื›ืœ ื”ืžื—ืœื•ืช ืžื”ืŸ ืกื•ื‘ืœ ื”ืžื˜ื•ืคืœ,
02:43
all the drugs that they're on,
56
163020
1767
ื›ืœ ื”ืชืจื•ืคื•ืช ืฉื”ื•ื ืœื•ืงื—,
02:44
and all the adverse events, or side effects, that they experience.
57
164811
3818
ื•ื›ืœ ื”ืชื•ืคืขื•ืช ื”ืฉืœื™ืœื™ื•ืช, ืื• ืชื•ืคืขื•ืช ื”ืœื•ื•ืื™, ืฉื”ื ื—ื•ื•ื™ื.
02:48
It is not all of the adverse events that are occurring in America today,
58
168653
3436
ืืœื• ืœื ื›ืœ ื”ืชื•ืคืขื•ืช ื”ืฉืœื™ืœื™ื•ืช ืฉืžืชืจื—ืฉื•ืช ื‘ืืžืจื™ืงื” ื›ื™ื•ื,
02:52
but it's hundreds and hundreds of thousands of drugs.
59
172113
2578
ืื‘ืœ ืืœื• ืžืื•ืช ื•ืžืื•ืช ืืœืคื™ ืชืจื•ืคื•ืช.
02:54
So I said to Nick,
60
174715
1299
ืื– ืืžืจืชื™ ืœื ื™ืง, "ื‘ื•ื ื ื—ืฉื•ื‘ ืขืœ ื’ืœื•ืงื•ื–.
02:56
"Let's think about glucose.
61
176038
1826
02:57
Glucose is very important, and we know it's involved with diabetes.
62
177888
3567
"ื”ื’ืœื•ืงื•ื– ื—ืฉื•ื‘ ืžืื“, ื•ืื ื—ื ื• ื™ื•ื“ืขื™ื ืฉื”ื•ื ืงืฉื•ืจ ื‘ืกื›ืจืช.
03:01
Let's see if we can understand glucose response.
63
181479
3970
"ื‘ื•ื ื ืจืื” ืื ื ื•ื›ืœ ืœื”ื‘ื™ืŸ ืืช ื”ืชื’ื•ื‘ื” ืœื’ืœื•ืงื•ื–."
03:05
I sent Nick off. Nick came back.
64
185473
2458
ืฉืœื—ืชื™ ืืช ื ื™ืง ืœื“ืจื›ื•. ื ื™ืง ื—ื–ืจ.
03:08
"Russ," he said,
65
188248
1786
"ืจืืก", ื”ื•ื ืืžืจ,
03:10
"I've created a classifier that can look at the side effects of a drug
66
190351
5112
"ื™ืฆืจืชื™ ื›ืœื™ ืžื™ื•ืŸ ืฉื™ื›ื•ืœ ืœื‘ื—ื•ืŸ ืืช ืชื•ืคืขื•ืช ื”ืœื•ื•ืื™ ืฉืœ ืชืจื•ืคื”
03:15
based on looking at this database,
67
195487
2051
"ื‘ื”ืชื‘ืกืก ืขืœ ืžืื’ืจ ื”ืžื™ื“ืข ื”ื–ื”,
03:17
and can tell you whether that drug is likely to change glucose or not."
68
197562
4271
"ืฉื™ื›ื•ืœ ืœื•ืžืจ ืœืš ืื ืกื‘ื™ืจ ืฉื”ืชืจื•ืคื” ื”ื–ื• ืชืฉื ื” ืืช ืจืžืช ื”ื’ืœื•ืงื•ื–."
03:21
He did it. It was very simple, in a way.
69
201857
2016
ื”ื•ื ืขืฉื” ื–ืืช. ื–ื” ืžืื“ ืคืฉื•ื˜, ืื™ื›ืฉื”ื•.
03:23
He took all the drugs that were known to change glucose
70
203897
2635
ื”ื•ื ืœืงื— ืืช ื›ืœ ื”ืชืจื•ืคื•ืช ืฉื”ื™ื” ื™ื“ื•ืข ืฉื”ืŸ ืžืฉื ื•ืช ืืช ื”ื’ืœื•ืงื•ื–
03:26
and a bunch of drugs that don't change glucose,
71
206556
2389
ื•ื›ืžื” ืชืจื•ืคื•ืช ืฉืœื ืžืฉื ื•ืช ืืช ื”ื’ืœื•ืงื•ื–,
03:28
and said, "What's the difference in their side effects?
72
208969
2888
ื•ืืžืจ, "ืžื” ื”ื”ื‘ื“ืœ ื‘ืชื•ืคืขื•ืช ื”ืœื•ื•ืื™ ืฉืœื”ืŸ?
03:31
Differences in fatigue? In appetite? In urination habits?"
73
211881
4852
"ื”ื”ื‘ื“ืœ ื‘ืชื—ื•ืฉืช ื”ืชืฉื™ืฉื•ืช? ื‘ืชื™ืื‘ื•ืŸ? ื‘ื”ืจื’ืœื™ ืžืชืŸ ื”ืฉืชืŸ?"
03:36
All those things conspired to give him a really good predictor.
74
216757
2960
ื›ืœ ื”ื“ื‘ืจื™ื ื”ืœืœื• ื™ื—ื“ ื ืชื ื• ืœื• ื›ืœื™ ื‘ืขืœ ื™ื›ื•ืœื•ืช ื ื™ื‘ื•ื™ ืžืžืฉ ื˜ื•ื‘ื•ืช.
03:39
He said, "Russ, I can predict with 93 percent accuracy
75
219741
2548
ื”ื•ื ืืžืจ, "ืจืืก, ืื ื™ ื™ื›ื•ืœ ืœื ื‘ื, ื‘ื“ื™ื•ืง ืฉืœ 93 ืื—ื•ื–ื™ื
03:42
when a drug will change glucose."
76
222313
1572
"ืžืชื™ ืชืจื•ืคื” ืชืฉื ื” ืืช ื”ื’ืœื•ืงื•ื–."
03:43
I said, "Nick, that's great."
77
223909
1416
ืืžืจืชื™, "ื ื™ืง, ื–ื” ื ื”ื“ืจ."
ื”ื•ื ืกื˜ื•ื“ื ื˜ ืฆืขื™ืจ ื•ืฆืจื™ืš ืœืขื•ื“ื“ ืืช ื”ื‘ื˜ื—ื•ืŸ ื”ืขืฆืžื™ ืฉืœื•.
03:45
He's a young student, you have to build his confidence.
78
225349
2896
"ืื‘ืœ ื ื™ืง, ื™ืฉ ื‘ืขื™ื”:
03:48
"But Nick, there's a problem.
79
228269
1390
03:49
It's that every physician in the world knows all the drugs that change glucose,
80
229683
3960
"ื›ืœ ืจื•ืคื ื‘ืขื•ืœื ืžื›ื™ืจ ืืช ื›ืœ ื”ืชืจื•ืคื•ืช ืฉืžืฉื ื•ืช ืืช ื”ื’ืœื•ืงื•ื–,
03:53
because it's core to our practice.
81
233667
2038
"ื›ื™ ื–ื” ื—ืฉื•ื‘ ืžืื“ ื‘ืคืจืงื˜ื™ืงื” ืฉืœื ื•.
03:55
So it's great, good job, but not really that interesting,
82
235729
3722
"ืื– ื ื”ื“ืจ, ืขื‘ื•ื“ื” ื˜ื•ื‘ื”, ืื‘ืœ ืœื ื‘ืืžืช ื›ืœ ื›ืš ืžืขื ื™ื™ืŸ,
03:59
definitely not publishable."
83
239475
1531
"ื•ื‘ื”ื—ืœื˜ ืœื ืžืฉื”ื• ืฉืืคืฉืจ ืœืคืจืกื."
04:01
(Laughter)
84
241030
1014
(ืฆื—ื•ืง)
04:02
He said, "I know, Russ. I thought you might say that."
85
242068
2550
ื”ื•ื ืืžืจ, "ืื ื™ ื™ื•ื“ืข, ืจืืก. ื—ืฉื‘ืชื™ ืฉื–ื” ืžื” ืฉืชื’ื™ื“."
04:04
Nick is smart.
86
244642
1152
ื ื™ืง ื‘ื—ื•ืจ ืคื™ืงื—.
04:06
"I thought you might say that, so I did one other experiment.
87
246149
2874
"ื—ืฉื‘ืชื™ ืฉื–ื” ืžื” ืฉืชื’ื™ื“, ืื– ืขืฉื™ืชื™ ืขื•ื“ ื ื™ืกื•ื™ ืื—ื“.
"ื‘ื“ืงืชื™ ืื ืฉื™ื ื‘ืžืื’ืจ ื”ืžื™ื“ืข ื”ื–ื” ืฉืœืงื—ื• ืฉืชื™ ืชืจื•ืคื•ืช ื‘ื• ื–ืžื ื™ืช,
04:09
I looked at people in this database who were on two drugs,
88
249047
2928
04:11
and I looked for signals similar, glucose-changing signals,
89
251999
4422
"ื•ื—ื™ืคืฉืชื™ ืกื™ืžื ื™ื ื“ื•ืžื™ื, ืกื™ืžื ื™ ืฉื™ื ื•ื™ ืจืžืช ื’ืœื•ืงื•ื–,
04:16
for people taking two drugs,
90
256445
1624
"ืืฆืœ ืื ืฉื™ื ืฉืœื•ืงื—ื™ื ืฉืชื™ ืชืจื•ืคื•ืช,
04:18
where each drug alone did not change glucose,
91
258093
5569
"ื›ืืฉืจ ื›ืœ ืื—ืช ืžื”ืชืจื•ืคื•ืช ื‘ืคื ื™ ืขืฆืžื” ืœื ืžืฉื ื” ืืช ืจืžืช ื”ื’ืœื•ืงื•ื–,
04:23
but together I saw a strong signal."
92
263686
2460
"ืื‘ืœ ื™ื—ื“, ื”ืŸ ื ืชื ื• ืกื™ืžืŸ ื—ื–ืง."
04:26
And I said, "Oh! You're clever. Good idea. Show me the list."
93
266170
3149
ื•ืื ื™ ืืžืจืชื™, "ืื•! ืืชื” ืคื™ืงื—. ืจืขื™ื•ืŸ ื˜ื•ื‘. ืชืจืื” ืœื™ ืืช ื”ืจืฉื™ืžื”."
04:29
And there's a bunch of drugs, not very exciting.
94
269343
2254
ื•ื™ืฉ ืงื‘ื•ืฆื” ืฉืœ ืชืจื•ืคื•ืช, ืœื ืžืื•ื“ ืžืจื’ืฉ.
04:31
But what caught my eye was, on the list there were two drugs:
95
271621
3932
ืื‘ืœ ืžื” ืฉืžืฉืš ืืช ืชืฉื•ืžืช ืœื‘ื™ ื”ื™ื”, ืฉื‘ืจืฉื™ืžื” ื”ื™ื• ืฉืชื™ ืชืจื•ืคื•ืช:
04:35
paroxetine, or Paxil, an antidepressant;
96
275577
3393
ืคืืจื•ืงืกื˜ื™ืŸ, ืื• ืคืงืกื™ืœ, ื ื•ื’ื“ ื“ื™ื›ืื•ืŸ;
04:39
and pravastatin, or Pravachol, a cholesterol medication.
97
279756
3570
ื•ืคืจื‘ืกื˜ืื˜ื™ืŸ, ืื• ืคืจืื‘ืื›ื•ืœ, ืชืจื•ืคื” ืœื›ื•ืœืกื˜ืจื•ืœ.
04:43
And I said, "Huh. There are millions of Americans on those two drugs."
98
283936
4283
ื•ืื ื™ ืืžืจืชื™, "ืื”ื”! ืžื™ืœื™ื•ื ื™ ืืžืจื™ืงืื™ื ืœื•ืงื—ื™ื ืืช ืฉืชื™ ื”ืชืจื•ืคื•ืช ื”ืืœื”."
04:48
In fact, we learned later,
99
288243
1246
ื‘ืคื•ืขืœ ื’ื™ืœื™ื ื• ืื—ืจ ื›ืš,
04:49
15 million Americans on paroxetine at the time, 15 million on pravastatin,
100
289513
6032
ืฉ-15 ืžื™ืœื™ื•ืŸ ืืžืจื™ืงืื™ื ืœื•ืงื—ื™ื ืคืืจื•ืงืกื˜ื™ืŸ ื‘ื›ืœ ืจื’ืข ื ืชื•ืŸ, 15 ืžื™ืœื™ื•ืŸ ืขืœ ืคืจื‘ืกื˜ื˜ื™ืŸ,
04:55
and a million, we estimated, on both.
101
295569
2817
ื•ื›ืžื™ืœื™ื•ืŸ, ื”ืขืจื›ื ื•, ืœื•ืงื—ื™ื ืืช ืฉืชื™ื”ืŸ.
04:58
So that's a million people
102
298767
1254
ืื– ืžื“ื•ื‘ืจ ื‘ืžื™ืœื™ื•ืŸ ืื ืฉื™ื ืฉืื•ืœื™ ืกื•ื‘ืœื™ื ืžื‘ืขื™ื•ืช ืขื ื”ื’ืœื•ืงื•ื– ืฉืœื”ื
05:00
who might be having some problems with their glucose
103
300045
2453
05:02
if this machine-learning mumbo jumbo that he did in the FDA database
104
302522
3206
ืื ื”ื”ื•ืงื•ืก ืคื•ืงื•ืก ืฉืœ ื”ืœืžื™ื“ืช-ืžื›ื•ื ื” ืฉืœื• ื‘ืžืื’ืจ ื”ืžื™ื“ืข ืฉืœ ืžื™ื ื”ืœ ื”ืžื–ื•ืŸ ื•ื”ืชืจื•ืคื•ืช
05:05
actually holds up.
105
305752
1254
ื™ืขืžื•ื“ ื‘ืžื‘ื—ืŸ.
05:07
But I said, "It's still not publishable,
106
307030
1927
ืื‘ืœ ืืžืจืชื™, "ื–ื” ืขื“ื™ื™ืŸ ืœื ืžืกืคื™ืง ืœืคื™ืจืกื•ื,
05:08
because I love what you did with the mumbo jumbo,
107
308981
2296
"ื›ื™ ืื ื™ ืื•ื”ื‘ ืืช ืžื” ืฉืขืฉื™ืช ืขื ื”ื”ื•ืงื•ืก-ืคื•ืงื•ืก ื•ืœืžื™ื“ืช ื”ืžื›ื•ื ื”,
05:11
with the machine learning,
108
311301
1246
05:12
but it's not really standard-of-proof evidence that we have."
109
312571
3864
"ืื‘ืœ ืžื” ืฉื™ืฉ ืœื ื• ืื™ื ื ื• ื‘ื“ื™ื•ืง ืจืื™ื•ืช ื‘ืœืชื™ ืžืขื•ืจืขืจื•ืช."
05:17
So we have to do something else.
110
317618
1589
ืื– ืื ื—ื ื• ืฆืจื™ื›ื™ื ืœืขืฉื•ืช ืžืฉื”ื• ืื—ืจ.
05:19
Let's go into the Stanford electronic medical record.
111
319231
2876
ื‘ื•ื ื ื™ื›ื ืก ืœืžืื’ืจ ื”ืจืคื•ืื™ ื”ืืœืงื˜ืจื•ื ื™ ืฉืœ ืกื˜ื ืคื•ืจื“.
05:22
We have a copy of it that's OK for research,
112
322131
2064
ื™ืฉ ืœื ื• ืขื•ืชืง ืฉืœื• ืฉืžืชืื™ื ืœืžื—ืงืจ,
ื”ื•ืฆืื ื• ืžื™ื“ืข ืžื–ื”ื”.
05:24
we removed identifying information.
113
324219
2046
05:26
And I said, "Let's see if people on these two drugs
114
326581
2503
ื•ืื ื™ ืืžืจืชื™, "ื‘ื•ื ื ืจืื” ืื ืœืื ืฉื™ื ืฉืœื•ืงื—ื™ื ืืช ืฉืชื™ ื”ืชืจื•ืคื•ืช ื”ืืœื”
05:29
have problems with their glucose."
115
329108
1769
"ื™ืฉ ื‘ืขื™ื” ืขื ื”ื’ืœื•ืงื•ื– ืฉืœื”ื."
05:31
Now there are thousands and thousands of people
116
331242
2207
ื™ืฉ ืืœืคื™ื ืจื‘ื™ื ืฉืœ ืื ืฉื™ื
05:33
in the Stanford medical records that take paroxetine and pravastatin.
117
333473
3459
ื‘ืžืื’ืจ ื”ืจืคื•ืื™ ืฉืœ ืกื˜ื ืคื•ืจื“ ืฉืœื•ืงื—ื™ื ืคืจื•ืงืกื˜ื™ืŸ ื•ืคืจื‘ืกื˜ื˜ื™ืŸ.
05:36
But we needed special patients.
118
336956
1799
ืื‘ืœ ื”ื™ื™ื ื• ื–ืงื•ืงื™ื ืœืžื˜ื•ืคืœื™ื ืžื™ื•ื—ื“ื™ื.
05:38
We needed patients who were on one of them and had a glucose measurement,
119
338779
4597
ื”ื™ื™ื ื• ื–ืงื•ืงื™ื ืœืžื˜ื•ืคืœื™ื ืฉืœื•ืงื—ื™ื ืื—ืช ืžื”ืชืจื•ืคื•ืช ื”ืืœื” ื•ืฉื™ืฉ ืœื”ื ืžื“ื™ื“ื•ืช ืฉืœ ื’ืœื•ืงื•ื–,
05:43
then got the second one and had another glucose measurement,
120
343400
3449
ื•ืื– ื”ืชื—ื™ืœื• ืขื ื”ืฉื ื™ื™ื” ื•ื”ื™ืชื” ืœื”ื ืžื“ื™ื“ืช ื’ืœื•ืงื•ื– ื ื•ืกืคืช,
05:46
all within a reasonable period of time -- something like two months.
121
346873
3615
ื›ืœ ื–ื” ื‘ืชืงื•ืคื” ืœื ืืจื•ื›ื” ืžื“ื™ -- ืžืฉื”ื• ื›ืžื• ื—ื•ื“ืฉื™ื™ื.
05:50
And when we did that, we found 10 patients.
122
350512
3159
ื•ื›ืฉืขืฉื™ื ื• ืืช ื–ื”, ืžืฆืื ื• 10 ืžื˜ื•ืคืœื™ื.
05:54
However, eight out of the 10 had a bump in their glucose
123
354592
4538
ืื‘ืœ, ืœ-8 ืžืชื•ืš ื”-10 ื”ื™ืชื” ืงืคื™ืฆื” ื‘ืจืžืช ื”ื’ืœื•ืงื•ื–
ื›ืฉื”ื ืงื™ื‘ืœื• ืืช ื”-"ืค" ื”ืฉื ื™ื™ื” -- ืื ื—ื ื• ืงื•ืจืื™ื ืœื–ื” "ืค" ื•-"ืค" --
05:59
when they got the second P -- we call this P and P --
124
359154
2645
06:01
when they got the second P.
125
361823
1310
ื›ืฉื”ื ืงื™ื‘ืœื• ืืช ื”-"ืค" ื”ืฉื ื™ื™ื”.
06:03
Either one could be first, the second one comes up,
126
363157
2562
ื–ื” ืœื ืžืฉื ื” ืื™ื–ื” ื”ื™ืชื” ืงื•ื“ื, ื‘ืจื’ืข ืฉื‘ืื” ื”ืฉื ื™ื™ื”,
06:05
glucose went up 20 milligrams per deciliter.
127
365743
2847
ืจืžืช ื”ื’ืœื•ืงื•ื– ืขืœืชื” ื‘-20 ืžื™ืœื™ื’ืจื ืœื“ืฆื™ืœื™ื˜ืจ.
06:08
Just as a reminder,
128
368614
1158
ืจืง ื›ืชื–ื›ื•ืจืช,
06:09
you walk around normally, if you're not diabetic,
129
369796
2325
ื‘ืื•ืคืŸ ื˜ื‘ืขื™ ื™ืฉ ืœื›ื, ืื ืืชื ืœื ื—ื•ืœื™ ืกื›ืจืช,
ืจืžืช ื’ืœื•ืงื•ื– ื‘ืกื‘ื™ื‘ื•ืช ื”-90.
06:12
with a glucose of around 90.
130
372145
1359
06:13
And if it gets up to 120, 125,
131
373528
2076
ื•ืื ื–ื” ืขื•ืœื” ืœ-120, 125,
06:15
your doctor begins to think about a potential diagnosis of diabetes.
132
375628
3450
ื”ืจื•ืคื ืฉืœื›ื ืžืชื—ื™ืœ ืœื—ืฉื•ื‘ ืขืœ ืื‘ื—ื•ืŸ ืฉืœ ืกื›ืจืช ื‘ืคื•ื˜ื ืฆื™ื”.
06:19
So a 20 bump -- pretty significant.
133
379102
2991
ืื– ืงืคื™ืฆื” ืฉืœ 20 -- ื“ื™ ืžืฉืžืขื•ืชื™.
06:22
I said, "Nick, this is very cool.
134
382601
1904
ืืžืจืชื™, "ื ื™ืง, ื–ื” ืžืžืฉ ืžื“ืœื™ืง.
06:25
But, I'm sorry, we still don't have a paper,
135
385616
2053
"ืื‘ืœ ืœืฆืขืจื™ ืขื“ื™ื™ืŸ ืื™ืŸ ืœื ื• ืžืืžืจ, "ื›ื™ ื–ื” 10 ืžื˜ื•ืคืœื™ื, ื‘ื—ื™ื™ืš --
06:27
because this is 10 patients and -- give me a break --
136
387693
2579
"ื–ื” ืœื ืžืกืคื™ืง ืžื˜ื•ืคืœื™ื."
06:30
it's not enough patients."
137
390296
1245
06:31
So we said, what can we do?
138
391565
1306
ืื– ืืžืจื ื•, ืžื” ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœืขืฉื•ืช?
06:32
And we said, let's call our friends at Harvard and Vanderbilt,
139
392895
2976
ื•ืืžืจื ื•, ื‘ื•ื ื ืงืจื ืœื—ื‘ืจื™ื ืฉืœื ื• ื‘ื”ืจื•ื•ืจื“ ื•ื‘ื•ื•ื ื“ืจื‘ื™ืœื˜,
06:35
who also -- Harvard in Boston, Vanderbilt in Nashville,
140
395895
2587
ืฉื’ื -- ืื•ื ื™ื‘ืจืกื™ื˜ืื•ืช ื”ืจื•ื•ืจื“ ื‘ื‘ื•ืกื˜ื•ืŸ, ื•-ื•ื•ื ื“ืจื‘ื™ืœื˜ ื‘ื ืฉื•ื•ื™ืœ,
06:38
who also have electronic medical records similar to ours.
141
398506
2821
ืฉื’ื ืœื”ื ื™ืฉ ืžืื’ืจื™ ืžื™ื“ืข ืจืคื•ืื™ ืืœืงื˜ืจื•ื ื™ื™ื ื“ื•ืžื™ื ืœืฉืœื ื•,
06:41
Let's see if they can find similar patients
142
401351
2020
ื•ื ืจืื” ืื ื”ื ื™ื›ื•ืœื™ื ืœืžืฆื•ื ืœื ื• ืžื˜ื•ืคืœื™ื ื“ื•ืžื™ื
06:43
with the one P, the other P, the glucose measurements
143
403395
3276
ืขื ื”-"ืค" ื”ืื—ืช ื•ืขื ื”-"ืค" ื”ืฉื ื™ื”, ื•ืขื ืžื“ื™ื“ื•ืช ื”ื’ืœื•ืงื•ื–
06:46
in that range that we need.
144
406695
1600
ื‘ืชื—ื•ื ืฉืื ื—ื ื• ืฆืจื™ื›ื™ื.
06:48
God bless them, Vanderbilt in one week found 40 such patients,
145
408787
4955
ื•ื•ื ื“ืจื‘ื™ืœื˜, ื™ื‘ื•ืจื›ื•, ืžืฆืื• ื‘ืฉื‘ื•ืข ืื—ื“ 40 ืžื˜ื•ืคืœื™ื ื›ืืœื•
06:53
same trend.
146
413766
1189
ืขื ืื•ืชื” ื”ืžื’ืžื”.
06:55
Harvard found 100 patients, same trend.
147
415804
3620
ื”ืจื•ื•ืจื“ ืžืฆืื• 100 ืžื˜ื•ืคืœื™ื. ืื•ืชื” ื”ืžื’ืžื”.
06:59
So at the end, we had 150 patients from three diverse medical centers
148
419448
4281
ื›ืš ืฉืœื‘ืกื•ืฃ, ื”ื™ื• ืœื ื• 150 ืžื˜ื•ืคืœื™ื ืžืฉืœื•ืฉื” ืžืจื›ื–ื™ื ืจืคื•ืื™ื™ื
07:03
that were telling us that patients getting these two drugs
149
423753
3297
ืฉื”ืจืื• ืœื ื• ืฉืžื˜ื•ืคืœื™ื ืฉืžืงื‘ืœื™ื ืืช ืฉืชื™ ื”ืชืจื•ืคื•ืช ื”ืœืœื•
07:07
were having their glucose bump somewhat significantly.
150
427074
2703
ืงื™ื‘ืœื• ืงืคื™ืฆืช ื’ืœื•ืงื•ื– ืžืฉืžืขื•ืชื™ืช.
07:10
More interestingly, we had left out diabetics,
151
430317
2810
ื•ื™ื•ืชืจ ืžืขื ื™ื™ืŸ, ืœื ื”ื›ืœืœื ื• ืืช ื—ื•ืœื™ ื”ืกื›ืจืช,
07:13
because diabetics already have messed up glucose.
152
433151
2317
ื›ื™ ืืฆืœ ื—ื•ืœื™ ืกื›ืจืช, ื”ื’ืœื•ืงื•ื– ื›ื‘ืจ ืžืฉื•ื‘ืฉ.
07:15
When we looked at the glucose of diabetics,
153
435492
2238
ื›ืฉื”ืกืชื›ืœื ื• ืขืœ ื”ื’ืœื•ืงื•ื– ืฉืœ ื—ื•ืœื™ ืกื›ืจืช,
07:17
it was going up 60 milligrams per deciliter, not just 20.
154
437754
3435
ื”ื•ื ืขืœื” ื‘-60 ืž"ื’ ืœื“ืฆื™ืœื™ื˜ืจ, ืœื ืจืง 20.
07:21
This was a big deal, and we said, "We've got to publish this."
155
441760
3452
ื–ื” ื”ื™ื” ืžืฉื”ื• ืจืฆื™ื ื™, ื•ืืžืจื ื•, "ืื ื—ื ื• ื—ื™ื™ื‘ื™ื ืœืคืจืกื ืืช ื–ื”."
07:25
We submitted the paper.
156
445236
1179
ื”ื’ืฉื ื• ืืช ื”ืžืืžืจ. ื”ื›ืœ ื”ื™ื” ืจืื™ื•ืช ืžื ืชื•ื ื™ื,
07:26
It was all data evidence,
157
446439
2111
07:28
data from the FDA, data from Stanford,
158
448574
2483
ื ืชื•ื ื™ื ืžืžื™ื ื”ืœ ื”ืžื–ื•ืŸ ื•ื”ืชืจื•ืคื•ืช, ื ืชื•ื ื™ื ืžืกื˜ื ืคื•ืจื“,
ื ืชื•ื ื™ื ืžื•ื•ื ื“ืจื‘ื™ืœื˜, ื ืชื•ื ื™ื ืžื”ืจื•ื•ืจื“.
07:31
data from Vanderbilt, data from Harvard.
159
451081
1946
ืœื ืขืฉื™ื ื• ืืคื™ืœื• ื ื™ืกื•ื™ ืืžื™ืชื™ ืื—ื“.
07:33
We had not done a single real experiment.
160
453051
2396
07:36
But we were nervous.
161
456495
1296
ืื‘ืœ ื”ื™ื™ื ื• ืžืชื•ื—ื™ื.
07:38
So Nick, while the paper was in review, went to the lab.
162
458201
3730
ืื– ื ื™ืง, ื‘ื–ืžืŸ ืฉื”ืžืืžืจ ืขื‘ืจ ื‘ื™ืงื•ืจืช, ื”ืœืš ืœืžืขื‘ื“ื”.
07:41
We found somebody who knew about lab stuff.
163
461955
2462
ืžืฆืื ื• ืžื™ืฉื”ื• ืฉืžืชืžืฆื ื‘ืžืขื‘ื“ื•ืช.
07:44
I don't do that.
164
464441
1393
ืื ื™ ืœื ืžืชืขืกืง ื‘ื–ื”.
07:45
I take care of patients, but I don't do pipettes.
165
465858
2417
ืื ื™ ืžื˜ืคืœ ื‘ื—ื•ืœื™ื, ืื‘ืœ ืื ื™ ืœื ืขื•ืฉื” ืคื™ืคื˜ื•ืช.
07:49
They taught us how to feed mice drugs.
166
469420
3053
ื”ื ืœื™ืžื“ื• ืื•ืชื ื• ืœื”ืื›ื™ืœ ืขื›ื‘ืจื™ื ื‘ืชืจื•ืคื•ืช.
07:52
We took mice and we gave them one P, paroxetine.
167
472864
2414
ืœืงื—ื ื• ืขื›ื‘ืจื™ื ื•ื ืชื ื• ืœื”ื "ืค" ืื—ืช, ืคืจื•ืงืกื˜ื™ืŸ.
07:55
We gave some other mice pravastatin.
168
475302
2508
ืœื›ืžื” ืขื›ื‘ืจื™ื ืื—ืจื™ื ื ืชื ื• ืคืจื‘ืกื˜ื˜ื™ืŸ.
07:57
And we gave a third group of mice both of them.
169
477834
3595
ื•ืœืงื‘ื•ืฆื” ืฉืœื™ืฉื™ืช ืฉืœ ืขื›ื‘ืจื™ื ื ืชื ื• ืืช ืฉื ื™ื”ื.
08:01
And lo and behold, glucose went up 20 to 60 milligrams per deciliter
170
481893
3946
ื•ื”ืคืœื ื•ืคืœื, ืจืžื•ืช ื”ื’ืœื•ืงื•ื– ืขืœื• ื‘ืขื›ื‘ืจื™ื ื‘-20 ืขื“ 60 ืž"ื’ ืœื“ืฆื™ืœื™ื˜ืจ.
08:05
in the mice.
171
485863
1171
08:07
So the paper was accepted based on the informatics evidence alone,
172
487058
3158
ืื– ื”ืžืืžืจ ื”ืชืงื‘ืœ ื‘ื”ืชื‘ืกืก ืขืœ ื”ืจืื™ื•ืช ืžื”ื ืชื•ื ื™ื ื‘ืœื‘ื“,
08:10
but we added a little note at the end,
173
490240
1894
ืื‘ืœ ื”ื•ืกืคื ื• ื”ืขืจื” ืงื˜ื ื” ื‘ืกื•ืฃ:
"ืื”, ื“ืจืš ืื’ื‘, ืื ื ื•ืชื ื™ื ืืช ื–ื” ืœืขื›ื‘ืจื™ื, ื–ื” ืขื•ืœื”."
08:12
saying, oh by the way, if you give these to mice, it goes up.
174
492158
2899
ื–ื” ื”ื™ื” ื ื”ื“ืจ, ื•ื”ืกื™ืคื•ืจ ื™ื›ื•ืœ ื”ื™ื” ืœื”ืกืชื™ื™ื ืฉื.
08:15
That was great, and the story could have ended there.
175
495081
2508
08:17
But I still have six and a half minutes.
176
497613
1997
ืื‘ืœ ื™ืฉ ืœื™ ืขื•ื“ ืฉืฉ ื•ื—ืฆื™ ื“ืงื•ืช.
08:19
(Laughter)
177
499634
2807
(ืฆื—ื•ืง)
08:22
So we were sitting around thinking about all of this,
178
502465
2815
ืื– ืื ื—ื ื• ื™ื•ืฉื‘ื™ื ืœื ื• ื•ื—ื•ืฉื‘ื™ื ืขืœ ื›ืœ ื–ื”,
08:25
and I don't remember who thought of it, but somebody said,
179
505304
2735
ื•ืื ื™ ืœื ื–ื•ื›ืจ ืžื™ ื—ืฉื‘ ืขืœ ื–ื”, ืื‘ืœ ืžื™ืฉื”ื• ืืžืจ,
"ืžืขื ื™ื™ืŸ ืื ืžื˜ื•ืคืœื™ื ืฉืœื•ืงื—ื™ื ืืช ืฉืชื™ ื”ืชืจื•ืคื•ืช ื”ืœืœื•
08:28
"I wonder if patients who are taking these two drugs
180
508063
3201
08:31
are noticing side effects of hyperglycemia.
181
511288
3553
"ืžื‘ื—ื™ื ื™ื ื‘ืชื•ืคืขื•ืช ืœื•ื•ืื™ ืฉืœ ื”ื™ืคืจ-ื’ืœื™ืงืžื™ื”.
08:34
They could and they should.
182
514865
1496
"ื”ื ืขืœื•ืœื™ื ื•ืฆืคื•ื™ื™ื ืœื—ื•ื•ืช ื–ืืช."
08:36
How would we ever determine that?"
183
516761
1877
ืื™ืš ื ื•ื›ืœ ืœื‘ื“ื•ืง ืืช ื–ื”?
08:39
We said, well, what do you do?
184
519551
1443
ืืžืจื ื•, ื˜ื•ื‘, ืžื” ืขื•ืฉื™ื?
ืื ืืชื ืœื•ืงื—ื™ื ืชืจื•ืคื”, ืชืจื•ืคื” ื—ื“ืฉื” ืื—ืช ืื• ืฉืชื™ื™ื,
08:41
You're taking a medication, one new medication or two,
185
521018
2580
08:43
and you get a funny feeling.
186
523622
1538
ื•ื™ืฉ ืœื›ื ื”ืจื’ืฉื” ืžื•ื–ืจื”.
08:45
What do you do?
187
525184
1151
ืžื” ืืชื ืขื•ืฉื™ื?
08:46
You go to Google
188
526359
1151
ืืชื ื ื›ื ืกื™ื ืœื’ื•ื’ืœ
08:47
and type in the two drugs you're taking or the one drug you're taking,
189
527534
3349
ื•ืžืงืœื™ื“ื™ื ืืช ืฉืชื™ ื”ืชืจื•ืคื•ืช ืฉืืชื ืœื•ืงื—ื™ื ืื• ืืช ื”ืื—ืช ืฉืืชื ืœื•ืงื—ื™ื,
08:50
and you type in "side effects."
190
530907
1603
ื•ืืชื ืžืงืœื™ื“ื™ื "ืชื•ืคืขื•ืช ืœื•ื•ืื™."
08:52
What are you experiencing?
191
532534
1356
ืืช ืžื” ืฉืืชื ื—ื•ื•ื™ื.
08:54
So we said OK,
192
534239
1151
ืื– ืืžืจื ื• "ื‘ืกื“ืจ,
08:55
let's ask Google if they will share their search logs with us,
193
535414
3056
"ื ืฉืืœ ืืช ื’ื•ื’ืœ ืื ื”ื ื™ืฉืชืคื• ืื•ืชื ื• ื‘ืจื™ืฉื•ืžื™ ื”ื—ื™ืคื•ืฉื™ื ืฉืœื”ื,
08:58
so that we can look at the search logs
194
538494
1833
"ื›ืš ืฉื ื•ื›ืœ ืœื‘ื—ื•ืŸ ืืช ืจื™ืฉื•ืžื™ ื”ื—ื™ืคื•ืฉ
09:00
and see if patients are doing these kinds of searches.
195
540351
2565
"ื•ืœื‘ื“ื•ืง ืื ืžื˜ื•ืคืœื™ื ืขื•ืจื›ื™ื ื—ื™ืคื•ืฉื™ื ื›ืืœื”."
09:02
Google, I am sorry to say, denied our request.
196
542940
3275
ื’ื•ื’ืœ, ืฆืจ ืœื™ ืœื•ืžืจ, ืกื™ืจื‘ื• ืœื‘ืงืฉื” ืฉืœื ื•.
09:06
So I was bummed.
197
546819
1151
ืื– ื”ื™ื™ืชื™ ื‘ื‘ืืกื”.
09:07
I was at a dinner with a colleague who works at Microsoft Research
198
547994
3166
ื”ื™ื™ืชื™ ื‘ืืจื•ื—ืช ืขืจื‘ ืขื ื—ื‘ืจ ืฉืขื•ื‘ื“ ื‘"ืžื™ืงืจื•ืกื•ืคื˜ ืžื—ืงืจ"
ื•ืืžืจืชื™, "ืจืฆื™ื ื• ืœืขืฉื•ืช ืื™ื–ื” ืžื—ืงืจ,
09:11
and I said, "We wanted to do this study,
199
551184
1941
"ื’ื•ื’ืœ ืกื™ืจื‘ื•, ื–ื” ื“ื™ ื‘ืืกื”."
09:13
Google said no, it's kind of a bummer."
200
553149
1880
ื”ื•ื ืืžืจ, "ื˜ื•ื‘, ืœื ื• ื™ืฉ ืืช ื”ื—ื™ืคื•ืฉื™ื ืžื‘ื™ื ื’."
09:15
He said, "Well, we have the Bing searches."
201
555053
2086
09:18
(Laughter)
202
558195
3483
(ืฆื—ื•ืง)
09:22
Yeah.
203
562805
1267
ื›ืŸ.
09:24
That's great.
204
564096
1151
ื ื”ื“ืจ.
09:25
Now I felt like I was --
205
565271
1151
ืขื›ืฉื™ื• ื”ืจื’ืฉืชื™ ื›ืื™ืœื• ืื ื™ -
09:26
(Laughter)
206
566446
1000
(ืฆื—ื•ืง)
09:27
I felt like I was talking to Nick again.
207
567470
2412
ื”ืจื’ืฉืชื™ ื›ืื™ืœื• ืื ื™ ืžื“ื‘ืจ ืืœ ื ื™ืง ืฉื•ื‘.
09:30
He works for one of the largest companies in the world,
208
570437
2624
ื”ื•ื ืขื•ื‘ื“ ื‘ืื—ืช ื”ื—ื‘ืจื•ืช ื”ื’ื“ื•ืœื•ืช ื‘ืขื•ืœื,
ื•ืื ื™ ืขื•ื“ ืžื ืกื” ืœื’ืจื•ื ืœื• ืœื”ืจื’ื™ืฉ ื™ื•ืชืจ ื˜ื•ื‘.
09:33
and I'm already trying to make him feel better.
209
573085
2206
09:35
But he said, "No, Russ -- you might not understand.
210
575315
2445
ืื‘ืœ ื”ื•ื ืืžืจ, "ืœื, ืจืืก, ืื•ืœื™ ืื™ื ืš ืžื‘ื™ืŸ.
"ืœื ืจืง ืฉื™ืฉ ืœื ื• ื”ื—ื™ืคื•ืฉื™ื ืฉืœ ื‘ื™ื ื’:
09:37
We not only have Bing searches,
211
577784
1500
"ืื ืืชื” ืžืฉืชืžืฉ ื‘ืื™ื ื˜ืจื ื˜ ืืงืกืคืœื•ืจืจ ื›ื“ื™ ืœืขืฉื•ืช ื—ื™ืคื•ืฉื™ื ื‘ื’ื•ื’ืœ,
09:39
but if you use Internet Explorer to do searches at Google,
212
579308
3340
09:42
Yahoo, Bing, any ...
213
582672
1891
"ื™ืื”ื•, ื‘ื™ื ื’, ืžื” ืฉืœื ื™ื”ื™ื”...
09:44
Then, for 18 months, we keep that data for research purposes only."
214
584587
3643
"ืื– ืœืžืฉืš ืฉื ื” ื•ื—ืฆื™ ืื ื• ืฉื•ืžืจื™ื ืืช ื”ืžื™ื“ืข ื”ื–ื” ืœ'ืžื˜ืจื•ืช ืžื—ืงืจ' ื‘ืœื‘ื“."
09:48
I said, "Now you're talking!"
215
588254
1936
ืืžืจืชื™, "ืืœื• ื“ื™ื‘ื•ืจื™ื!"
09:50
This was Eric Horvitz, my friend at Microsoft.
216
590214
2198
ื–ื” ื”ื™ื” ืืจื™ืง ื”ื•ืจื•ื‘ื™ืฅ, ื”ื—ื‘ืจ ืฉืœื™ ื‘ืžื™ืงืจื•ืกื•ืคื˜.
09:52
So we did a study
217
592436
1695
ืื– ืขืฉื™ื ื• ืžื—ืงืจ
09:54
where we defined 50 words that a regular person might type in
218
594155
4619
ื‘ื• ื”ื’ื“ืจื ื• 50 ืžื™ืœื™ื ืฉืื“ื ืจื’ื™ืœ ืขืฉื•ื™ ืœื”ืงืœื™ื“
09:58
if they're having hyperglycemia,
219
598798
1602
ืื ื™ืฉ ืœื• ื”ื™ืคืจื’ืœื™ืงืžื™ื”,
10:00
like "fatigue," "loss of appetite," "urinating a lot," "peeing a lot" --
220
600424
4762
ื›ื’ื•ืŸ "ืชืฉื™ืฉื•ืช", "ื—ื•ืกืจ ืชื™ืื‘ื•ืŸ", "ื ื•ืชืŸ ื”ืจื‘ื” ืฉืชืŸ", "ืžืฉืชื™ืŸ ื”ืจื‘ื”" --
10:05
forgive me, but that's one of the things you might type in.
221
605210
2767
ืกืœื—ื• ืœื™, ืื‘ืœ ื–ื” ืื—ื“ ื”ื“ื‘ืจื™ื ืฉื”ื•ื ืขืฉื•ื™ ืœื”ืงืœื™ื“.
ืื– ื™ืฉ ืœื ื• 50 ื‘ื™ื˜ื•ื™ื™ื ืฉื›ื™ื ื™ื ื•: "ืžื™ืœื•ืช ื”ืกื›ืจืช."
10:08
So we had 50 phrases that we called the "diabetes words."
222
608001
2790
10:10
And we did first a baseline.
223
610815
2063
ื•ืงื•ื“ื ื™ืฆืจื ื• ื‘ืกื™ืก.
10:12
And it turns out that about .5 to one percent
224
612902
2704
ื•ืžืกืชื‘ืจ ืฉื‘ืขืจืš ื—ืฆื™ ืื—ื•ื– ืขื“ ืื—ื•ื– ืื—ื“
10:15
of all searches on the Internet involve one of those words.
225
615630
2982
ืžื›ืœ ื”ื—ื™ืคื•ืฉื™ื ื‘ืื™ื ื˜ืจื ื˜ ืžื›ื™ืœ ืืช ืื—ืช ืžื”ืžื™ืœื™ื ื”ืœืœื•.
10:18
So that's our baseline rate.
226
618636
1742
ื–ืืช ื”ืจืžื” ื”ื‘ืกื™ืกื™ืช ืฉืœื ื•.
10:20
If people type in "paroxetine" or "Paxil" -- those are synonyms --
227
620402
4143
ืื ืื ืฉื™ื ืžืงืœื™ื“ื™ื "ืคืจื•ืงืกื˜ื™ืŸ" ืื• "ืคืงืกื™ืœ" -- ืืœื” ืžื™ืœื™ื ื ืจื“ืคื•ืช --
10:24
and one of those words,
228
624569
1215
ื•ืืช ืื—ืช ื”ืžื™ืœื™ื ื”ืœืœื•,
10:25
the rate goes up to about two percent of diabetes-type words,
229
625808
4890
ืื– ืชื“ื™ืจื•ืช ื”ื•ืคืขืช ืžื™ืœื•ืช ื”ืกื›ืจืช ืขื•ืœื” ืžืขืœ ืฉื ื™ ืื—ื•ื–ื™ื,
10:30
if you already know that there's that "paroxetine" word.
230
630722
3044
ืื ื™ื•ื“ืขื™ื ืฉื”ืžื™ืœื” "ืคืจื•ืงืกื˜ื™ืŸ" ื›ื‘ืจ ืงื™ื™ืžืช.
10:34
If it's "pravastatin," the rate goes up to about three percent from the baseline.
231
634191
4547
ืื ื–ื” "ืคืจื‘ืกื˜ื˜ื™ืŸ", ื”ืชื“ื™ืจื•ืช ืขื•ืœื” ืœื›-3% ืžืจืžืช ื”ื‘ืกื™ืก.
10:39
If both "paroxetine" and "pravastatin" are present in the query,
232
639171
4390
ืื "ืคืจื•ืงืกื˜ื™ืŸ" ื•"ืคืจื‘ืกื˜ื˜ื™ืŸ" ืžื•ืคื™ืขื•ืช ืฉืชื™ื”ืŸ ื‘ืฉืื™ืœืชื,
10:43
it goes up to 10 percent,
233
643585
1669
ื–ื” ืขื•ืœื” ืขื“ ืœ-10 ืื—ื•ื–ื™ื,
10:45
a huge three- to four-fold increase
234
645278
3461
ืขืœื™ื” ืขื ืงื™ืช ืฉืœ ืคื™ ืฉืœื•ืฉ ืื• ืืจื‘ืข
10:48
in those searches with the two drugs that we were interested in,
235
648763
3389
ื‘ื—ื™ืคื•ืฉื™ื ื”ืœืœื• ืขื ืฉืชื™ ื”ืชืจื•ืคื•ืช ืฉืžืขื ื™ื™ื ื•ืช ืื•ืชื ื•,
10:52
and diabetes-type words or hyperglycemia-type words.
236
652176
3566
ื•ืžื™ืœื•ืช ื”ืกื›ืจืช ืื• ืžื™ืœื•ืช ื”ื™ืคืจื’ืœื™ืงืžื™ื”.
10:56
We published this,
237
656216
1265
ืคืจืกืžื ื• ืืช ื–ื”,
10:57
and it got some attention.
238
657505
1466
ื•ื–ื” ื–ื›ื” ืœืงืฆืช ืชืฉื•ืžืช ืœื‘.
10:58
The reason it deserves attention
239
658995
1778
ื”ืกื™ื‘ื” ืฉื–ื” ืจืื•ื™ ืœืชืฉื•ืžืช ืœื‘
11:00
is that patients are telling us their side effects indirectly
240
660797
4312
ื”ื™ื ืฉื›ืš ื”ืžื˜ื•ืคืœื™ื ืžืกืคืจื™ื ืœื ื• ื‘ืขืงื™ืคื™ืŸ ืขืœ ืชื•ืคืขื•ืช ื”ืœื•ื•ืื™ ืฉืœื”ื
11:05
through their searches.
241
665133
1156
ื“ืจืš ื”ื—ื™ืคื•ืฉื™ื ืฉืœื”ื.
11:06
We brought this to the attention of the FDA.
242
666313
2138
ื”ืกื‘ื ื• ืœื›ืš ืืช ืชืฉื•ืžืช ื”ืœื‘ ืฉืœ ืžื™ื ื”ืœ ื”ืžื–ื•ืŸ ื•ื”ืชืจื•ืคื•ืช.
11:08
They were interested.
243
668475
1269
ื–ื” ืกื™ืงืจืŸ ืื•ืชื.
11:09
They have set up social media surveillance programs
244
669768
3606
ื”ื ื”ืชืงื™ื ื• ืชื•ื›ื ื™ื•ืช ืžืขืงื‘ ืื—ืจื™ ืจืฉืชื•ืช ื—ื‘ืจืชื™ื•ืช
11:13
to collaborate with Microsoft,
245
673398
1751
ื‘ืฉื™ืชื•ืฃ ืคืขื•ืœื” ืขื ืžื™ืงืจื•ืกื•ืคื˜,
11:15
which had a nice infrastructure for doing this, and others,
246
675173
2794
ืฉื™ืฉ ืœื” ืชืฉืชื™ืช ื™ืคื” ืœืฉื ื›ืš, ื•ืขื ืื—ืจื™ื,
11:17
to look at Twitter feeds,
247
677991
1282
ืœื‘ื—ื•ืŸ ืฆื™ื•ืฆื™ื ื‘ื˜ื•ื•ื™ื˜ืจ,
11:19
to look at Facebook feeds,
248
679297
1716
ืœื‘ื—ื•ืŸ ืคืจืกื•ืžื™ื ื‘ืคื™ื™ืกื‘ื•ืง,
11:21
to look at search logs,
249
681037
1311
ืœื‘ื—ื•ืŸ ืจื™ืฉื•ืžื™ื ืžืžื ื•ืขื™ ื—ื™ืคื•ืฉ,
11:22
to try to see early signs that drugs, either individually or together,
250
682372
4909
ื›ื“ื™ ืœื ืกื•ืช ื•ืœืืชืจ ืกื™ืžื ื™ื ืžื•ืงื“ืžื™ื ืœื›ืš ืฉืชืจื•ืคื•ืช, ื‘ื ืคืจื“ ืื• ื‘ื™ื—ื“,
11:27
are causing problems.
251
687305
1589
ื™ื•ืฆืจื•ืช ื‘ืขื™ื•ืช.
11:28
What do I take from this? Why tell this story?
252
688918
2174
ืžื” ื”ืœืงื— ืฉืœื™ ื‘ื›ืœ ื–ื”? ืœืžื” ืœืกืคืจ ืืช ื”ืกื™ืคื•ืจ?
ื•ื‘ื›ืŸ, ืงื•ื“ื ื›ืœ,
11:31
Well, first of all,
253
691116
1207
11:32
we have now the promise of big data and medium-sized data
254
692347
4037
ื™ืฉ ืœื ื• ื›ืขืช ืืช ื”ื”ื‘ื˜ื—ื” ื”ื’ืœื•ืžื” ื‘ื ืชื•ื ื™ื ืจื‘ื™ื ื•ื‘ื™ื ื•ื ื™ื™ื
11:36
to help us understand drug interactions
255
696408
2918
ืฉื™ืืคืฉืจื• ืœื ื• ืœื”ื‘ื™ืŸ ืื™ื ื˜ืจืืงืฆื™ื•ืช ื‘ื™ืŸ ืชืจื•ืคื•ืช
11:39
and really, fundamentally, drug actions.
256
699350
2420
ื•ื‘ืขืฆื, ื‘ืื•ืคืŸ ื‘ืกื™ืกื™, ืคืขื™ืœื•ืช ืฉืœ ืชืจื•ืคื•ืช.
11:41
How do drugs work?
257
701794
1413
ืื™ืš ืขื•ื‘ื“ื•ืช ืชืจื•ืคื•ืช?
11:43
This will create and has created a new ecosystem
258
703231
2836
ื–ื” ื™ื™ืฆื•ืจ, ื•ื›ื‘ืจ ื™ืฆืจ, ืžืขืจื›ืช ืืงื•ืœื•ื’ื™ืช ื—ื“ืฉื”
11:46
for understanding how drugs work and to optimize their use.
259
706091
3267
ื‘ืฉื‘ื™ืœ ืœื”ื‘ื™ืŸ ืื™ืš ืชืจื•ืคื•ืช ืขื•ื‘ื“ื•ืช ื•ืœืžื˜ื‘ ืืช ื”ืฉื™ืžื•ืฉ ื‘ื”ืŸ.
11:50
Nick went on; he's a professor at Columbia now.
260
710303
2659
ื ื™ืง ื”ืชืงื“ื ื”ืœืื”; ืขื›ืฉื™ื• ื”ื•ื ืคืจื•ืคืกื•ืจ ื‘ืื•ื ื™ื‘ืจืกื™ื˜ืช ืงื•ืœื•ืžื‘ื™ื”.
11:52
He did this in his PhD for hundreds of pairs of drugs.
261
712986
4072
ื”ื•ื ืขืฉื” ืืช ื–ื” ื‘ื“ื•ืงื˜ื•ืจื˜ ืฉืœื• ืขื ืžืื•ืช ื–ื•ื’ื•ืช ืฉืœ ืชืจื•ืคื•ืช.
11:57
He found several very important interactions,
262
717082
2517
ื”ื•ื ืžืฆื ื›ืžื” ืื™ื ื˜ืจืืงืฆื™ื•ืช ืžืื“ ื—ืฉื•ื‘ื•ืช,
11:59
and so we replicated this
263
719623
1214
ื•ื›ืš ืฉื™ื›ืคืœื ื• ืืช ื–ื”
12:00
and we showed that this is a way that really works
264
720861
2574
ื•ื”ืจืื™ื ื• ืฉื–ื• ื“ืจืš ืฉื‘ืืžืช ืคื•ืขืœืช
12:03
for finding drug-drug interactions.
265
723459
2339
ื•ืžืฆืœื™ื—ื” ืœืืชืจ ืื™ื ื˜ืจืืงืฆื™ื•ืช ื‘ื™ืŸ-ืชืจื•ืคืชื™ื•ืช.
12:06
However, there's a couple of things.
266
726282
1734
ืื‘ืœ ื™ืฉ ื›ืžื” ื“ื‘ืจื™ื.
12:08
We don't just use pairs of drugs at a time.
267
728040
3046
ืื ื—ื ื• ืœื ืœื•ืงื—ื™ื ื‘ื‘ืช ืื—ืช ืจืง ื–ื•ื’ื•ืช ืฉืœ ืชืจื•ืคื•ืช.
12:11
As I said before, there are patients on three, five, seven, nine drugs.
268
731110
4469
ื›ืคื™ ืฉืืžืจืชื™ ืงื•ื“ื, ื™ืฉื ื ืžื˜ื•ืคืœื™ื ืฉืœื•ืงื—ื™ื ืฉืœื•ืฉ, ื—ืžืฉ, ืฉื‘ืข, ืชืฉืข ืชืจื•ืคื•ืช.
12:15
Have they been studied with respect to their nine-way interaction?
269
735981
3642
ื”ืื ื”ืŸ ื‘ืืžืช ื ื—ืงืจื• ื‘ื”ืงืฉืจ ืœืื™ื ื˜ืจืืงืฆื™ื” ื”ืชืฉืข-ื›ื™ื•ื•ื ื™ืช ืฉืœื”ืŸ?
12:19
Yes, we can do pair-wise, A and B, A and C, A and D,
270
739647
4208
ื ื›ื•ืŸ, ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื‘ื—ื•ืŸ ื–ื•ื’ื•ืช: ื ื•-ื‘, ื ื•-ื’, ื ื•-ื“,
12:23
but what about A, B, C, D, E, F, G all together,
271
743879
4286
ืื‘ืœ ืžื” ืขื ื, ื‘, ื’, ื“, ื”, ื•, ื•-ื– ื›ื•ืœื ื‘ื™ื—ื“,
12:28
being taken by the same patient,
272
748189
1762
ื‘ื”ื™ืœืงื—ืŸ ืขืœ ื™ื“ื™ ืื•ืชื• ื”ืžื˜ื•ืคืœ,
12:29
perhaps interacting with each other
273
749975
2118
ืฉืื•ืœื™ ื™ื•ืฆืจื•ืช ืื™ื ื˜ืจืืงืฆื™ื” ื‘ื™ื ื™ื”ืŸ
12:32
in ways that either makes them more effective or less effective
274
752117
3778
ื‘ื“ืจื›ื™ื ืฉืžื’ื‘ื™ืจื•ืช ืื• ืžืงื˜ื™ื ื•ืช ืืช ื™ืขื™ืœื•ืชืŸ,
12:35
or causes side effects that are unexpected?
275
755919
2332
ืื• ื’ื•ืจืžื•ืช ืœืชื•ืคืขื•ืช ืœื•ื•ืื™ ื‘ืœืชื™ ืฆืคื•ื™ื•ืช?
12:38
We really have no idea.
276
758275
1827
ื‘ืืžืช ืฉืื™ืŸ ืœื ื• ืžื•ืฉื’.
12:40
It's a blue sky, open field for us to use data
277
760126
3756
ื–ื”ื• ืื•ืคืง ืคืชื•ื—, ืชื—ื•ื ืœื-ื™ื“ื•ืข ืขื‘ื•ืจื ื• ื‘ืืฉืจ ืœืฉื™ืžื•ืฉ ื‘ื ืชื•ื ื™ื
12:43
to try to understand the interaction of drugs.
278
763906
2502
ื›ื“ื™ ืœื ืกื•ืช ื•ืœื”ื‘ื™ืŸ ืื™ื ื˜ืจืืงืฆื™ื•ืช ื‘ื™ืŸ ืชืจื•ืคื•ืช.
12:46
Two more lessons:
279
766848
1370
ืฉื ื™ ืœืงื—ื™ื ื ื•ืกืคื™ื:
12:48
I want you to think about the power that we were able to generate
280
768242
4199
ืื ื™ ืจื•ืฆื” ืฉืชื—ืฉื‘ื• ืขืœ ื”ื›ื•ื— ืฉื”ืฆืœื—ื ื• ืœื™ืฆื•ืจ
12:52
with the data from people who had volunteered their adverse reactions
281
772465
4711
ืขื ื”ื ืชื•ื ื™ื ืžืื ืฉื™ื ืฉื ื™ื“ื‘ื• ืืช ื”ืชื•ืคืขื•ืช ื”ืฉืœื™ืœื™ื•ืช ืฉืœื”ื
12:57
through their pharmacists, through themselves, through their doctors,
282
777200
3269
ื“ืจืš ื”ืจื•ืงื—ื™ื ืฉืœื”ื, ื‘ืขืฆืžื, ื“ืจืš ื”ืจื•ืคืื™ื ืฉืœื”ื,
13:00
the people who allowed the databases at Stanford, Harvard, Vanderbilt,
283
780493
3667
ื”ืื ืฉื™ื ืฉืื™ืคืฉืจื• ืœืžืื’ืจื™ ื”ืžื™ื“ืข ื‘ืกื˜ืื ืคื•ืจื“, ื”ืจื•ื•ืจื“, ื•ื•ื ื“ืจื‘ื™ืœื˜,
13:04
to be used for research.
284
784184
1427
ืœื”ื™ื•ืช ืžื ื•ืฆืœื™ื ืœืฆืจื›ื™ ืžื—ืงืจ.
13:05
People are worried about data.
285
785929
1445
ืื ืฉื™ื ืžื•ื“ืื’ื™ื ื‘ืงืฉืจ ืœืžื™ื“ืข.
13:07
They're worried about their privacy and security -- they should be.
286
787398
3187
ื”ื ื“ื•ืื’ื™ื ืœืคืจื˜ื™ื•ืช ื•ืœื‘ื˜ื—ื•ืŸ ืฉืœื”ื-- ื•ื‘ืฆื“ืง.
ืื ื—ื ื• ืฆืจื™ื›ื™ื ืžืขืจื›ื•ืช ืžืื•ื‘ื˜ื—ื•ืช.
13:10
We need secure systems.
287
790609
1151
13:11
But we can't have a system that closes that data off,
288
791784
3406
ืื‘ืœ ืื ื—ื ื• ืœื ื™ื›ื•ืœื™ื ืฉืชื”ื™ื” ืžืขืจื›ืช ืฉื—ื•ืกืžืช ืืช ื”ื ืชื•ื ื™ื,
13:15
because it is too rich of a source
289
795214
2752
ื›ื™ ื–ื” ืžืงื•ืจ ื™ื•ืชืจ ืžื“ื™ ืขืฉื™ืจ
13:17
of inspiration, innovation and discovery
290
797990
3971
ืฉืœ ื”ืฉืจืื”, ื—ื“ืฉื ื•ืช ื•ื’ื™ืœื•ื™
13:21
for new things in medicine.
291
801985
1578
ืขื‘ื•ืจ ื“ื‘ืจื™ื ื—ื“ืฉื™ื ื‘ืจืคื•ืื”.
13:24
And the final thing I want to say is,
292
804494
1794
ื•ื”ื“ื‘ืจ ื”ืื—ืจื•ืŸ ืฉืื ื™ ืจื•ืฆื” ืœื•ืžืจ ื”ื•ื,
13:26
in this case we found two drugs and it was a little bit of a sad story.
293
806312
3357
ื‘ืžืงืจื” ื”ื–ื” ืžืฆืื ื• ืฉืชื™ ืชืจื•ืคื•ืช ื•ื–ื” ื”ื™ื” ืกื™ืคื•ืจ ืงืฆืช ืขืฆื•ื‘.
13:29
The two drugs actually caused problems.
294
809693
1921
ืฉืชื™ ื”ืชืจื•ืคื•ืช ื‘ืืžืช ื™ืฆืจื• ื‘ืขื™ื•ืช.
13:31
They increased glucose.
295
811638
1475
ื”ื ื”ื’ื‘ื™ืจื• ืจืžื•ืช ื’ืœื•ืงื•ื–.
13:33
They could throw somebody into diabetes
296
813137
2446
ื”ื ื™ื›ืœื• ืœื’ืจื•ื ืœืžื™ืฉื”ื• ืกื›ืจืช
13:35
who would otherwise not be in diabetes,
297
815607
2294
ืฉื‘ื ืกื™ื‘ื•ืช ืื—ืจื•ืช ืœื ื”ื™ืชื” ืœื•,
13:37
and so you would want to use the two drugs very carefully together,
298
817925
3175
ื›ืš ืฉื›ื“ืื™ ืœื”ืฉืชืžืฉ ื‘ืฉืชื™ ื”ืชืจื•ืคื•ืช ื”ืืœื” ืžืื“ ื‘ื–ื”ื™ืจื•ืช ื‘ื™ื—ื“,
ืื•ืœื™ ืœื ืœืงื—ืช ืื•ืชืŸ ื™ื—ื“,
13:41
perhaps not together,
299
821124
1151
13:42
make different choices when you're prescribing.
300
822299
2340
ืื•ืœื™ ืœืจืฉื•ื ืชืจื•ืคื•ืช ืื—ืจื•ืช.
13:44
But there was another possibility.
301
824663
1846
ืื‘ืœ ื”ื™ืชื” ืืคืฉืจื•ืช ืื—ืจืช.
13:46
We could have found two drugs or three drugs
302
826533
2344
ื™ื›ื•ืœื ื• ืœืžืฆื•ื ืฉืชื™ื™ื ืื• ืฉืœื•ืฉ ืชืจื•ืคื•ืช
13:48
that were interacting in a beneficial way.
303
828901
2261
ืฉืคื•ืขืœื•ืช ื‘ืื™ื ื˜ืจืืงืฆื™ื” ื•ื™ื•ืฆืจื•ืช ืืคืงื˜ ื—ื™ื•ื‘ื™.
13:51
We could have found new effects of drugs
304
831616
2712
ื”ื™ื™ื ื• ื™ื›ื•ืœื™ื ืœื’ืœื•ืช ืืคืงื˜ื™ื ื—ื“ืฉื™ื ืฉืœ ืชืจื•ืคื•ืช
13:54
that neither of them has alone,
305
834352
2160
ืฉืœื ืงื™ื™ืžื™ื ืœืืฃ ืื—ืช ืžื”ืŸ ื‘ื ืคืจื“,
13:56
but together, instead of causing a side effect,
306
836536
2493
ืืš ื‘ื™ื—ื“, ื‘ืžืงื•ื ืœื’ืจื•ื ืœืชื•ืคืขื•ืช ืœื•ื•ืื™,
13:59
they could be a new and novel treatment
307
839053
2425
ื”ืŸ ื™ื›ื•ืœื•ืช ืœื”ื•ื•ืช ื˜ื™ืคื•ืœ ื—ื“ืฉ ื•ื—ื“ืฉื ื™
14:01
for diseases that don't have treatments
308
841502
1882
ืœืžื—ืœื•ืช ืœื”ืŸ ืื™ืŸ ืœื ื• ื˜ื™ืคื•ืœ ืื• ืฉื”ื˜ื™ืคื•ืœ ื‘ื”ืŸ ืื™ื ื ื• ื™ืขื™ืœ.
14:03
or where the treatments are not effective.
309
843408
2007
ืื ื ื™ืงื— ื˜ื™ืคื•ืœ ืชืจื•ืคืชื™ ื›ื™ื•ื, ืืช ื›ืœ ืคืจื™ืฆื•ืช ื”ื“ืจืš ื”ื—ืฉื•ื‘ื•ืช --
14:05
If we think about drug treatment today,
310
845439
2395
14:07
all the major breakthroughs --
311
847858
1752
14:09
for HIV, for tuberculosis, for depression, for diabetes --
312
849634
4297
ืœืื™ื™ื“ืก, ืœืฉื—ืคืช, ืœื“ื™ื›ืื•ืŸ, ืœืกื›ืจืช --
14:13
it's always a cocktail of drugs.
313
853955
2830
ื–ื” ืชืžื™ื“ ืงื•ืงื˜ื™ื™ืœ ืฉืœ ืชืจื•ืคื•ืช.
14:16
And so the upside here,
314
856809
1730
ื›ืš ืฉื”ืฆื“ ื”ื—ื™ื•ื‘ื™ ื›ืืŸ,
14:18
and the subject for a different TED Talk on a different day,
315
858563
2849
ื•ื”ื ื•ืฉื ืœื”ืจืฆืืช TED ืื—ืจืช, ื‘ื™ื•ื ืื—ืจ,
14:21
is how can we use the same data sources
316
861436
2593
ื”ื•ื ืื™ืš ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื”ืฉืชืžืฉ ื‘ืื•ืชื ืžืงื•ืจื•ืช ื”ืžื™ื“ืข
14:24
to find good effects of drugs in combination
317
864053
3563
ื›ื“ื™ ืœืžืฆื•ื ืืคืงื˜ื™ื ื—ื™ื•ื‘ื™ื™ื ืฉืœ ืฉื™ืœื•ื‘ื™ ืชืจื•ืคื•ืช
14:27
that will provide us new treatments,
318
867640
2175
ืฉื™ื™ืฆืจื• ืœื ื• ื˜ื™ืคื•ืœื™ื ื—ื“ืฉื™ื,
14:29
new insights into how drugs work
319
869839
1852
ืชื•ื‘ื ื•ืช ื—ื“ืฉื•ืช ื‘ืืฉืจ ืœื“ืจื›ื™ ื”ืคืขื•ืœื” ืฉืœ ืชืจื•ืคื•ืช,
14:31
and enable us to take care of our patients even better?
320
871715
3786
ื•ื™ืืคืฉืจื• ืœื ื• ืœื˜ืคืœ ื‘ื—ื•ืœื™ื ืฉืœื ื• ืืคื™ืœื• ื™ื•ืชืจ ื˜ื•ื‘?
14:35
Thank you very much.
321
875525
1166
ืชื•ื“ื” ืจื‘ื”.
14:36
(Applause)
322
876715
3499
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
ืขืœ ืืชืจ ื–ื”

ืืชืจ ื–ื” ื™ืฆื™ื’ ื‘ืคื ื™ื›ื ืกืจื˜ื•ื ื™ YouTube ื”ืžื•ืขื™ืœื™ื ืœืœื™ืžื•ื“ ืื ื’ืœื™ืช. ืชื•ื›ืœื• ืœืจืื•ืช ืฉื™ืขื•ืจื™ ืื ื’ืœื™ืช ื”ืžื•ืขื‘ืจื™ื ืขืœ ื™ื“ื™ ืžื•ืจื™ื ืžื”ืฉื•ืจื” ื”ืจืืฉื•ื ื” ืžืจื—ื‘ื™ ื”ืขื•ืœื. ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ื”ืžื•ืฆื’ื•ืช ื‘ื›ืœ ื“ืฃ ื•ื™ื“ืื• ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ ืžืฉื. ื”ื›ืชื•ื‘ื™ื•ืช ื’ื•ืœืœื•ืช ื‘ืกื ื›ืจื•ืŸ ืขื ื”ืคืขืœืช ื”ื•ื•ื™ื“ืื•. ืื ื™ืฉ ืœืš ื”ืขืจื•ืช ืื• ื‘ืงืฉื•ืช, ืื ื ืฆื•ืจ ืื™ืชื ื• ืงืฉืจ ื‘ืืžืฆืขื•ืช ื˜ื•ืคืก ื™ืฆื™ืจืช ืงืฉืจ ื–ื”.

https://forms.gle/WvT1wiN1qDtmnspy7