Perhaps the BEST B1G Forum anywhere, here at College Football Fan Site, CFB51!!!

the media is never wrong

it's the "experts" they use for the source

the sources should be uncovered and held accountable - never happens, unless it's a politician

I don;t think antibody testing is very widespread yet, and at least one version of it seems to be deemed unreliable by the medical/scientific community.

Ultimately, a reliable and widely available antibody test will be the real key to helping open up and make people feel safer-- that is, of course, IFF some reasonable amount of immunity is conferred.

The issue that I've seen with antibody testing is that the false positive rate is too high for small exposure numbers.

In the Santa Clara study, they used a test that they claimed had only a 0.5% false positive rate, by combining the results of several studies regarding that test, but several of those studies showed much higher false positive rates (some at 1.5% and higher).

Even so, they tested 3330 people and got 50 positive results (1.5%). Fully a third of those positive results could have been false positives using their own estimation of the false positive rate. Their range for potential false positive rate was 0.3-0.8%, which means that if the false positive rate was at the top of their scale, they could have more than half of their positives be false. And if the false positive rate was as high as some of the other studies of that test suggest (1.5%)? Literally they could have potentially tested zero people who had been exposed/infected by COVID-19 and found 50 false positives.

Then there were other issues. It wasn't a random sampling. They basically advertised the test on Facebook, so you can't be certain that what they got was anywhere near a random sample of the population--it could be people who thought they might have had it and wanted to check, or it could be people who had higher risk tolerance to going outside (since they went for antibody testing during a stay-at-home order), etc.

Finally, when they tried to account for false positives/negatives, they said the likely real infection rate was 1.2% in their sample. So they then weighted that sample for the demographics of the county based on where the positive results came from and said the likely infection rate in the county was 2.8% (95CI 1.3-4.7%). Literally they said that accounting for test performance the likely prevalence rate was 1.2%, but that was below the lowest 95% confidence interval for the lower bound of prevalence due to weighting. 

They should have just thrown out all their data and walked away... Or expanded their test to a much larger sample size. Or done something, ANYTHING, other than publish something that suggests the true mortality rate was between 0.12-0.20%, which is quite frankly BS.

So per your bolded part: I don't think there's any problem with the test itself. The issue is that you KNOW there's a false positive rate. If you're measuring a sample and the actual positive rate from the test is close to that false positive rate, you don't have enough prevalence to draw strong conclusions. If your false positive rate is 1.5%, and your actual positive rate from testing is near 25% (as in NYC), then you know that 1.5% false positive is only a small part of your potential measurement error. If your false positive rate is 1.5%, and your actual test result is 2.5% positive, you know that somewhere near 60% of your positive results are suspect and you can't really say much about the result.

bwarb,

I'm talking about the "expert" that predicted surges in deaths in Georgia and Texas on reopening

no papers written, just someone's opinion

that someone may not be an expert at all, simply a source that draws attention

I'm an expert in floodplain and stormwater management (and land surveying).

As I've pointed out... A lot of times the "media" is someone who went to J-school and doesn't have any real expertise in the actual subject matter they're writing about. So what happens?

• Scientist--the only "expert" in the following chain does research. This is likely to be a very dry research paper. As with most things, the scientist is probably studying some small part of a phenomenon. Let's say it's the reaction (in a petri dish) of cells being attacked by SARS-CoV-2 in the presence of an antibody (such as the recent Sorrento Therapeutics stuff). The results show a strong inhibition to infection. 
• The scientist writes a REALLY boring paper on it. It's full of Greek symbols and all sorts of science words that are meaningless to anyone but other scientists. There's a lot of math. In the end it's seen that there is a p-value and it's below 0.05, so he publishes in a peer-reviewed journal.
• The PR team for the journal wants to get people to subscribe to the journal. So they put out a press release about the specific antibody that inhibits infection of the deadly coronavirus. The PR team starts pitching this to journalists to get them to write about it. (I know a little about this as WAY back in the day my ex worked in a PR firm representing tech in Silicon Valley. It was a constant pitch cycle to try to get tech writers to pick up on the press releases from the companies she represented and write about them).
• A journalist sees the press release. The journalist doesn't read the research paper. The journalist only reads the press release, which is by definition a hype document.
• The journalist makes mistakes. Suddenly they don't report that the inhibition of infection was only proven in a petri dish, and not in the actual human immune system. They include quotes from the company owner that sponsored the research, they include a whole lot of speculation about how this "cure" will change everything. And of course, they bury in paragraph 8 that it's still months away from even Phase I human trials, much less availability.
• The editor, who wants people to spread their story, then takes the actual article and writes a clickbait headline sure to lead to a lot of shares on Twitter and Facebook. "Discovery of antibody proven to show 100% blockage of COVID-19 infection"
• The public laps it up and can't wait for the miracle cure.

Then the treatment fails badly in Phase I trials, or doesn't even make it there because something else is found that's wrong, and everyone sweeps it under the rug.

But instead of blaming the media for hyping something WAY beyond what the actual expert proved--which was correct even if not applicable to human treatment (which he never claimed)--we say that we can't trust experts.

How about you blame the media for reporting it wrong? They're usually too stupid to understand what the experts have even said, much less report accurately on its meaning.

Based on this chart, the breakdown is that the media folks, who most likely can't decipher the preliminary scientific paper, should ignore the PR team?

That seems like the solution, but with the optics of "Media ignores good news from science company"

I think the answer is not that it was reported wrong, per say, not only because the media is mostly parroting what the researcher's company is saying but because it's not so much wrong as warped by level after level of over exuberance. A company sees a gain from pretending it's a bigger deal. Some media see it as a gain, others might not, but it doesn't matter because once a few folks put it out there, "the media" has reported it. And readers/news consumers run it through all sorts of prisms, mostly ending with complaints. 

So in one case "the experts are wrong" but in others, when this stuff is reported by some, it's "why is the MSM keeping down this good news."

Basically, we need everyone to take a damn step back and realize the world is often not as super interesting as we think (sometimes it is, but not here)

Oh, so probably some op-ed writer or some talking head on TV? Someone who is probably part of the media more than an actual expert on anything?

Most experts I know tend to speak about their predictions in very controlled ways. "Well, if the governor opens the economy, and the businesses open to normal capacity, and people return to their previous behavior, then we are likely to see a spike in cases."

What happens is the media sees that and says "Expert Tom Smith predicts spike in cases due to reopening."

And then when--as in Georgia--many restaurants stay closed, the ones that are open have few patrons, and generally the bulk of society practices social distancing in a way completely unlike they did before March, the expert is wrong because people only looked at his conclusion without recognizing the conditions--which never happened--he predicated that conclusion on?

Ed Zachery!


the experts might not know jack squat, but the media is the asshat here

They are spouses or better half’s of people I work with in banking throughout Florida ( but very little in Dade County.) Some are Doctors, some nurses, a couple administration and a few in Finance.

here where I live we specialize in banking doctors from the bigger local hospitals.

what they are saying is they geared up for the big onslaught that never came and in fact they are all slow now with some reducing their hours

The exception was Palm Beach county- they were busy ( not at all overwhelmed) when the virus first wave hit.  Many ( anecdotal) of the patients they were seeing were from New York/ NE USA. 

one thing that was done here in Florida, and at least for now appears to have been effective- is they really focused hard in nursing Homes and assisted living facilities. They went in with inspection teams to examine conditions supplies and cleanliness levels and said some pretty strict rules.

On this side, at least in the area where we live, most people are still wearing masks and observing the 6' suggestion. How are things over there?