WEBVTT - Virus Hunting With the 'Pirate CDC' 0:00:05.080 --> 0:00:10.440 Welcome to Prognosis. I'm Laura Carlson. It's day seventy six 0:00:10.720 --> 0:00:15.560 since coronavirus was declared a global pandemic. Our main story. 0:00:16.720 --> 0:00:20.000 The US seems to have finally gotten its testing operation 0:00:20.040 --> 0:00:23.880 on track, but it's still hard to find data on 0:00:24.040 --> 0:00:28.360 just how many tests are being done, so a volunteer 0:00:28.400 --> 0:00:32.600 group of journalists took up the challenge. Now the group 0:00:32.640 --> 0:00:36.720 produces reports that are so reliable even the White House 0:00:36.720 --> 0:00:48.839 depends on that. But first, here's what happened today. The 0:00:48.960 --> 0:00:52.320 US drug giant Mark revealed its plans to develop a 0:00:52.360 --> 0:00:57.160 pill that will treat the coronavirus. The company is also 0:00:57.240 --> 0:01:01.480 working on two vaccines. Work bought the rights to develop 0:01:01.480 --> 0:01:06.160 a promising anti viral drug that was discovered at Emory University. 0:01:07.640 --> 0:01:12.280 Over the past century, Mark has pioneered inoculations for diseases 0:01:12.400 --> 0:01:17.640 from ebola to diphtheria. The company pledged that if they 0:01:17.640 --> 0:01:20.880 are able to make a coronavirus vaccine, they will work 0:01:20.920 --> 0:01:24.240 to make it accessible to anyone in the world who 0:01:24.319 --> 0:01:29.560 needs it. WU Want, the epicenter of China's coronavirus outbreak, 0:01:29.840 --> 0:01:33.720 said it tested nearly seven million people in twelve days. 0:01:34.920 --> 0:01:38.280 That's after a handful of infections prompted fears of a 0:01:38.319 --> 0:01:43.039 second wave and spurred a campaign to test the entire population. 0:01:44.560 --> 0:01:49.360 The six point eight million nucleic acid tests uncovered two 0:01:49.480 --> 0:01:55.040 hundred and six asymptomatic cases, according to Bloomberg calculations based 0:01:55.080 --> 0:01:59.760 on daily numbers released by the local Health Commission. And Finally, 0:02:00.760 --> 0:02:03.600 researchers may have found a small piece of the puzzle 0:02:03.680 --> 0:02:09.320 about whether people with COVID nineteen later become immune. An 0:02:09.360 --> 0:02:12.400 early draft of a study of hospital staff in northeastern 0:02:12.440 --> 0:02:16.080 France found that almost all doctors and nurses who got 0:02:16.120 --> 0:02:21.840 mild forms of COVID nineteen produced antibodies that could prevent reinfection. 0:02:23.280 --> 0:02:28.120 The research by Institute Pasteur and University Hospitals in Strasbourg 0:02:28.600 --> 0:02:33.519 addresses a crucial question regarding the new coronavirus, whether people 0:02:33.600 --> 0:02:37.640 who had COVID nineteen, and especially those who didn't get 0:02:37.680 --> 0:02:44.000 severely ill, develop antibodies against the disease. As recently as April, 0:02:44.919 --> 0:02:48.320 the World Health Organization said that there's no evidence yet 0:02:48.680 --> 0:02:52.680 that people who have recovered and have antibodies are protected 0:02:52.840 --> 0:03:04.480 from a second infection, And now our main story at 0:03:04.480 --> 0:03:07.640 the start of the coronavirus pandemic. The scathed number of 0:03:07.720 --> 0:03:11.760 virus tests being conducted in the US was a major fiasco. 0:03:13.080 --> 0:03:17.000 Now though most supply problems have been solved and many 0:03:17.080 --> 0:03:20.919 more tests are being done throughout the country, but it's 0:03:20.919 --> 0:03:23.720 still hard to get a sense of just how many. 0:03:25.320 --> 0:03:28.240 Knowing who is being tested is essential for getting an 0:03:28.280 --> 0:03:31.960 accurate picture of the spread of the virus, but the 0:03:32.080 --> 0:03:38.080 government hasn't readily provided this data. Instead, experts, media outlets, 0:03:38.160 --> 0:03:42.440 and even the Trump administration have turned to its surprising source, 0:03:43.440 --> 0:03:46.760 a volunteer effort by a team of journalists called the 0:03:46.880 --> 0:03:52.200 COVID Tracking Project. Bloomberg News reporter and a Court spoke 0:03:52.240 --> 0:03:55.800 with the project's co founder, Alexis Magical, about why he 0:03:55.880 --> 0:03:59.000 decided to chase these numbers and what they're telling us 0:03:59.240 --> 0:04:06.960 about who with the virus is affecting. Alexis Magical started 0:04:07.000 --> 0:04:10.080 paying close attention to the data on COVID nineteen in 0:04:10.160 --> 0:04:14.200 early March Alexis as a reporter for The Atlantic. He 0:04:14.240 --> 0:04:17.359 found that a month after the first confirmed case of 0:04:17.440 --> 0:04:20.760 COVID nineteen in the US, the country's numbers on the 0:04:20.800 --> 0:04:24.760 infections were way too low. A friend of his from college, 0:04:24.880 --> 0:04:28.919 Jeff Hammerbacker, was also watching the subject closely, so they 0:04:29.000 --> 0:04:32.720 joined together to create the COVID Tracking Project. It's a 0:04:32.839 --> 0:04:36.320 volunteer led effort based out of The Atlantic magazine that 0:04:36.440 --> 0:04:41.680 gathers state level data about testing cases, deaths, and hospitalizations. 0:04:42.279 --> 0:04:46.080 The project also tracts race and ethnicity data when possible. 0:04:47.120 --> 0:04:49.880 That's time consuming work. In the early days, it was 0:04:49.960 --> 0:04:54.240 just Alexis Robinson mayor his colleague at The Atlantic, Jeff 0:04:54.320 --> 0:04:58.320 and some other volunteers like co founder Aaron Cassane. They 0:04:58.360 --> 0:05:01.640 were all working around the clock. I spoke with Alexis 0:05:01.720 --> 0:05:05.800 about the project. The data collection effort, you know, is 0:05:05.839 --> 0:05:09.279 not that difficult to do once, right, you know, you 0:05:09.320 --> 0:05:11.960 go to all these state department state health department websites, 0:05:12.279 --> 0:05:15.120 you write down a bunch of information and a spreadsheet. Um, 0:05:15.160 --> 0:05:17.880 you know, doing that one time. You know, anyone can 0:05:17.920 --> 0:05:20.160 imagine doing that. You know, lots of reporters have done 0:05:20.160 --> 0:05:22.800 this with different kinds of COVID nineteen data, and you can. 0:05:22.839 --> 0:05:25.640 You can do it once. But now let's say you 0:05:25.680 --> 0:05:28.839 need to do it three times a day for a month. 0:05:29.440 --> 0:05:31.280 You can't have the same person doing it like that. 0:05:31.600 --> 0:05:33.479 No one person is going to survive that, No small 0:05:33.480 --> 0:05:36.440 team is going to survive that the relentlessness of it, 0:05:36.560 --> 0:05:39.280 you know, is just too tough. So now you've got 0:05:39.279 --> 0:05:42.640 to create rules for how you're going to code certain 0:05:42.640 --> 0:05:45.120 types of information. The states are all reporting things and 0:05:45.760 --> 0:05:48.840 very very different ways, and insofar as we can standardize 0:05:49.000 --> 0:05:59.960 those things, we try. You might imagine technology could be very, 0:06:00.040 --> 0:06:04.520 very useful in tracking COVID nineteen data, but Alexis says 0:06:04.680 --> 0:06:10.120 it's actually a highly manual process. Volunteers pull the numbers, 0:06:10.240 --> 0:06:15.080 and more volunteers double check their work. This requires systems 0:06:15.120 --> 0:06:20.240 to ensure data quality and train new volunteers. Basically a 0:06:20.320 --> 0:06:25.679 whole organization complete with Slack and Google Docs. The data 0:06:25.720 --> 0:06:29.240 is collected from public health authorities via official reports, as 0:06:29.279 --> 0:06:34.200 well as news conferences, Twitter, and other sources. Jordan's Gasparay, 0:06:34.640 --> 0:06:37.280 who produced the segment, is one of the many journalists 0:06:37.320 --> 0:06:41.239 from various outlets who contributed to this massive data collection effort. 0:06:41.720 --> 0:06:45.680 Alexis says that in an ideal world, the Centers for 0:06:45.720 --> 0:06:49.479 Disease Control and Prevention would be gathering and standardizing this 0:06:49.600 --> 0:06:54.520 crucial data, not an army of volunteers. In fact, he 0:06:54.640 --> 0:06:58.560 likes to call their group the pirates c DC. You know, 0:06:58.680 --> 0:07:01.479 these systems for gathering data at a you know, in 0:07:01.480 --> 0:07:03.280 a country as large as the United States with a 0:07:03.279 --> 0:07:07.320 healthcare system is fragmented as it is, are just remarkably difficult. 0:07:07.440 --> 0:07:09.640 And a lot of it is voluntary reporting, which means, 0:07:09.960 --> 0:07:12.640 you know, even if the federal government says to states like, hey, 0:07:12.640 --> 0:07:14.680 can you please report all of your data to us, 0:07:15.120 --> 0:07:17.480 the hospitals in that state just may not do that. Um, 0:07:17.520 --> 0:07:19.200 they should do that, but they may not do that, 0:07:19.280 --> 0:07:22.200 you know. And the same goes for like race and 0:07:22.200 --> 0:07:24.760 ethnicity data and things like that. People just don't fill 0:07:24.800 --> 0:07:26.840 it out. They don't fill out when they're ordering a test, 0:07:26.920 --> 0:07:29.520 they just don't fill out race and ethnicity on the form. 0:07:29.640 --> 0:07:34.440 And so the kind of data holes that begin, you know, 0:07:34.520 --> 0:07:37.239 with small actions, you know, down at the point of care, 0:07:37.880 --> 0:07:40.760 blow up to you know, on the national scale, to 0:07:40.840 --> 0:07:43.360 be these massive data holes that don't allow us to 0:07:43.360 --> 0:07:45.960 to you know, truly understand what's going on with the 0:07:46.000 --> 0:07:49.120 outbreak or or how to respond to it. Even the 0:07:49.160 --> 0:07:53.160 Trump administration has cited numbers from the COVID tracking project, 0:07:53.640 --> 0:07:56.840 As my colleague Kristen V. Brown and I have reported, 0:07:57.840 --> 0:08:01.520 the White House released a nationwide testing strategy document in 0:08:01.640 --> 0:08:05.680 late April that credits data to Alexis and his colleagues project. 0:08:06.600 --> 0:08:10.240 The Trump administration didn't return our request for comments about 0:08:10.280 --> 0:08:13.560 this or answer our questions about where it's data is 0:08:13.600 --> 0:08:18.360 coming from, but the administration has previously released numbers on 0:08:18.560 --> 0:08:23.280 testing that roughly match up to the COVID tracking projects reporting. 0:08:24.480 --> 0:08:27.680 I asked Alexis if the federal government is using the 0:08:27.720 --> 0:08:30.920 project's data, and whether the White House has its own 0:08:30.960 --> 0:08:36.200 source of COVID nineteen information. What I choose to believe 0:08:36.559 --> 0:08:40.640 is that the federal government has data um and that 0:08:40.920 --> 0:08:43.600 some of it is way better than ours, even but 0:08:43.720 --> 0:08:47.120 that perhaps our data has its own kinds of utility, 0:08:47.200 --> 0:08:51.720 like stretching back in time and standardizing in certain ways 0:08:51.720 --> 0:08:55.120 and sort of allowing the federal government to understand how 0:08:55.160 --> 0:08:57.320 the states are reporting this relative to how they are 0:08:57.360 --> 0:09:00.079 seeing data come in. I don't know. I mean that 0:09:00.280 --> 0:09:02.120 the truth is I don't know the answer to these questions. 0:09:02.160 --> 0:09:05.200 But but you know, just having been around this testing 0:09:05.280 --> 0:09:07.640 data and talk to enough people who are in diagnostic 0:09:07.679 --> 0:09:10.319 labs and other kinds of reporting, I think there's a 0:09:10.400 --> 0:09:14.160 zero percent chance the federal government has nothing. But the 0:09:14.200 --> 0:09:16.880 fact that there's nothing public means they probably don't have 0:09:16.960 --> 0:09:21.119 everything for whatever reason. All of this matters because widespread 0:09:21.160 --> 0:09:25.040 testing is understood to be key to measuring the number 0:09:25.120 --> 0:09:29.560 of COVID nineteen cases and reopening the economy as safely 0:09:29.760 --> 0:09:35.920 as possible. Without mass testing, experts can't be confident in 0:09:35.960 --> 0:09:39.800 the number of cases that have been reported. And while 0:09:39.800 --> 0:09:42.920 the White House has long insisted that the US has 0:09:43.000 --> 0:09:47.240 excelled at testing, getting a complete, up to date picture 0:09:47.520 --> 0:09:54.560 of how it's all going has often been difficult. We 0:09:54.600 --> 0:09:56.560 don't have a good idea of how many people are sick. 0:09:57.400 --> 0:10:02.640 That's why. But part of what's happening is that you know, 0:10:02.880 --> 0:10:05.400 New York, New Jersey, and Connecticut. You know, the New 0:10:05.480 --> 0:10:09.320 York metro area had a blazing outbreak that caught the 0:10:09.520 --> 0:10:14.240 entire nation, entire world by surprise because of the botched 0:10:14.679 --> 0:10:16.959 early testing efforts in the United States, they were only 0:10:16.960 --> 0:10:18.880 testing very sick people until you know, one out of 0:10:18.920 --> 0:10:22.920 every two people was coming back positive. Now what's happening 0:10:23.040 --> 0:10:28.160 is the outbreak is growing outside of those um three places. 0:10:28.640 --> 0:10:31.960 But we but we have better eyes on it because 0:10:31.960 --> 0:10:34.120 there's more testing capacity now than there was in the 0:10:34.120 --> 0:10:37.600 beginning of March. Testing grew to become one of the 0:10:37.600 --> 0:10:40.560 biggest crises in the US in the first months of 0:10:40.640 --> 0:10:46.319 COVID nineteen, and even though the country's testing abilities have improved, 0:10:46.640 --> 0:10:53.080 many problems have lingered, including around capacity. I would like 0:10:53.120 --> 0:10:55.600 to say that we had far seeing vision and we 0:10:55.640 --> 0:10:57.520 of course knew that this would be the crucial issue 0:10:57.520 --> 0:10:59.959 of this entire pandemic, But it's not true. I mean, 0:11:00.040 --> 0:11:03.240 and for me, it's not true. My reporting partner, Rob 0:11:03.280 --> 0:11:06.400 Meyer at The Atlantic, he immediately identified. He was like, 0:11:06.440 --> 0:11:08.400 this is the most important number in America right now, 0:11:08.520 --> 0:11:10.800 is how many people have been tested? UM. He did 0:11:10.800 --> 0:11:13.319 say that at the time. Uh, And so I will say, 0:11:13.360 --> 0:11:15.720 perhaps Rob knew that this would be this big. But 0:11:15.800 --> 0:11:20.520 to be honest, in the weeks moving through March, when 0:11:20.720 --> 0:11:22.480 there was all this talk about the scale up and 0:11:22.520 --> 0:11:24.760 blah blah blah blah blah, I did not think that 0:11:24.800 --> 0:11:27.000 it would that testing would continue to be a number 0:11:27.080 --> 0:11:30.199 that was so deeply important. I really didn't. I thought 0:11:30.240 --> 0:11:33.679 that the project would probably wrap up UM sometime in April, 0:11:34.320 --> 0:11:36.440 either because the CDC came or it just didn't matter 0:11:36.480 --> 0:11:40.800 anymore because testing availability was so enormous. And then basically 0:11:40.800 --> 0:11:44.120 every report that came out in April basically said like, actually, 0:11:44.120 --> 0:11:46.520 we need millions of tests to safely reopen this country. 0:11:46.640 --> 0:11:49.280 And like when every expert is to saying with something 0:11:49.360 --> 0:11:52.560 like that, you go like, well, damn, I guess this 0:11:52.720 --> 0:11:59.840 really is it. The US can now do about four 0:12:00.080 --> 0:12:03.760 hundred thousand tests a day, less than half the nine 0:12:03.880 --> 0:12:08.679 hundred thousand one group of experts says is needed. That's 0:12:08.679 --> 0:12:13.000 the case even as states pushed towards reopening their economies, 0:12:13.760 --> 0:12:18.640 sending people back to work in school and eating at restaurants. 0:12:19.840 --> 0:12:22.199 There's where we are with testing right now in the US. 0:12:22.400 --> 0:12:25.880 We tested less than two thousand people total by March six. 0:12:26.120 --> 0:12:28.520 You know, we entered April doing you know, just about 0:12:28.559 --> 0:12:31.800 a hundred thousand tests a day. We left April doing 0:12:31.880 --> 0:12:35.320 you know, about two hundred thousand tests a day. But 0:12:35.440 --> 0:12:38.760 most of April was a big long plateau actually, and 0:12:38.800 --> 0:12:40.320 it was only really at the end of the month 0:12:40.520 --> 0:12:42.960 um that testing started to pick up. You know, most 0:12:43.000 --> 0:12:46.640 testing now is done by lab corps and Quest and 0:12:46.760 --> 0:12:50.400 other you know, big sort of commercial laboratories. Alexis says 0:12:50.480 --> 0:12:54.480 he's a believer that once the American innovation system gets 0:12:54.520 --> 0:13:00.920 cranking along, testing numbers will eventually take off. Over the 0:13:01.000 --> 0:13:05.040 next few months. He says we'll get there, maybe even 0:13:05.160 --> 0:13:10.920 stand out on a global level. But Alexis also says 0:13:11.080 --> 0:13:15.440 it's possible that by trying to bring on testing while 0:13:15.480 --> 0:13:20.520 also reopening, there will be massive outbreaks and major damage 0:13:20.800 --> 0:13:23.720 done along the way. That's kind of how it looks 0:13:23.760 --> 0:13:25.400 to me. It looks like a race between sort of 0:13:25.400 --> 0:13:31.400 our innovative capacities, you know, and are kind of reactionary 0:13:31.440 --> 0:13:33.840 impulses to just try and get everything to go back 0:13:33.840 --> 0:13:36.120 to normal even though it's clearly not going to happen. 0:13:48.920 --> 0:13:51.840 That was in the court and that's our show today. 0:13:52.600 --> 0:13:56.080 For coverage of the outbreak from one buros around the world, 0:13:56.520 --> 0:14:01.560 visit Bloomberg dot com, slash coronavirus and if you like 0:14:01.679 --> 0:14:04.480 the show, please leave us a review at a rating 0:14:04.800 --> 0:14:08.440 on Apple Podcasts or Spotify. It's the best way to 0:14:08.440 --> 0:14:13.880 help more listeners find our global reporting. The Prognosis Daily 0:14:14.000 --> 0:14:18.160 edition is produced by Topher Foreheads, Jordan gas Pure, Magnus 0:14:18.160 --> 0:14:23.080 Hendrickson and me Laura Carlson. Today's main story was reported 0:14:23.080 --> 0:14:28.240 by Emma Court. Original music by Leo Sidrin. Our editors 0:14:28.320 --> 0:14:33.560 are Francesco Levi and Rick Shine. Francesco Levi is Bloomberg's 0:14:33.760 --> 0:15:00.280 head of podcasts, Thanks for listening. One