The Mission District mass testing effort that will run from today to Tuesday is ambitious in its scope. It aims to test some 2,000 to 6,000 individuals in a city that, to date, has only tested fewer than 13,000. You can register here.
Volunteers report back to us that some eligible Mission residents are reticent to be tested out of a sense of altruism. They feel healthy; let someone who needs this test take it.
A kindly impulse. But, in this case, the worst and most counterproductive impulse you can have. If you want to be altruistic — take this test.
Because, among all the reasons you should take the test if you’re eligible, one that has been underreported is: It’s not just about you.
At the massive outbreak at the MSC-South homeless shelter, the majority of the 100-plus positive cases were asymptomatic. The same goes for the deadliest outbreak in the city thus far, at the Central Gardens Convalescent Hospital.
So, clearly, asymptomatic carriers are out there — but that’s just who doesn’t get tested. All of the city’s testing has, to some extent, been reactive (much — one could argue too much — of the city’s COVID-19 strategy is reactive).
Thanks to a scandalous nationwide dearth of testing, we’ve focused on symptomatic people (in fact, only highly symptomatic people; plenty of sick folks have been denied) and people who’ve come into contact with symptomatic people following outbreaks or positive tests.
Obviously, it skews our picture of who’s sick and who isn’t and how prevalent COVID-19 is when testing is limited to people who are demonstrably unwell or think they are. The more universal testing is, the more we’ll know.
“We want everybody to take the test, regardless of symptoms,” says Dr. Gabriel Chamie of UCSF, who is helping to lead this four-day test-a-thon. If you’re worried about displacing someone needier — don’t. It’s paid for: “We have all of the kits we need for the testing.”
San Francisco has, thus far, been searching for its keys only under the streetlight. With the Mission District testing efforts, we’ll finally begin to see what’s out there in the dark.
But will we like what we see?

Is there an upside to San Francisco’s COVID-19 numbers all but certainly going up — likely way up — after this bevy of testing? Yes, obviously: You can’t fight a disease via wishful thinking any more than you can by injecting Clorox or beaming a UV light up your rear end.
This city will have a truer idea of what it’s up against; ignoring the engine light doesn’t keep you from breaking down. And all of the asymptomatic people who felt just fine will realize they’re carriers, take further precautions, and stop spreading the disease. That matters.
Yet perception matters too. And the borderline triumphalism this city embraced while other municipalities descended into charnel houses will take a blow with any negative-seeming news. That’s another reason why it’s never good to take victory laps before the race is near complete, nor stoke San Francisco’s formidable sense of exceptionalism.
It has taken a while, but this city is putting out some salient and usable data regarding the pandemic within the friendly confines of this city’s 47 square miles. With the thousands forthcoming tests in the Mission, the city will soon have scads more data.
And, on the whole, that’s for the good. But it could well distort and overpower the numbers we’ve got, and confuse the narrative.

Peter Khoury is a Mission District data scientist who’s been helping Mission Local distill the ocean of COVID-19 data washing over us. He’s incredibly enthusiastic about the forthcoming Mission testing. But he acknowledges they could still skew San Francisco’s COVID-19 data.
Take demographics: Already, it’s clear that Latinxs are disproportionately suffering from COVID-19. But with the lion’s share of new testing coming from the city’s Latinx stronghold, the picture it presents may be even uglier than reality.
Think about how, on maps, the Mercator Projection causes Antarctica — which is big — to appear monumentally vast. Khoury notes that heavy Mission data, combined with light overall city data, could potentially overstate the (already disproportionate) infection rate among Latinxs, while understating infection rates among other heavily afflicted communities, like black folks.
Khoury also worries that a vast influx of new data from a small district could lead to the mega-version of an effect we’re already seeing following outbreaks like those at MSC-South and Central Gardens.
The city has started releasing zip code-by-zip code data — and, to the surprise of nobody, poorer communities populated by people of color in which many people have no choice but to leave home to work public-facing jobs are heavier hit.
But take a look at 94115, the Fillmore/Western Addition. Statistically, it’s one of the harder-hit parts of the city. But is it really? Of the 77 cases recorded here in a recent tally, 65 were at Central Gardens. If you factor that out, this is one of the least affected parts of all San Francisco.
But you wouldn’t know that by looking at the data writ large. As Khoury puts it, this data indicates the entire neighborhood is reeling — when, really, it’s largely limited to one building.
That’s a different reality and it requires a different set of responses from the city, health officials, and everyone else.
The city is about to get much, much more data right at the time when it would be best to refine and fine-tune its data. This, obviously, presents a challenge.
But that’s a challenge for the near future. The challenge for today is to get tested if you’ve got the opportunity to do so.
We can’t mismanage the data we don’t collect.
***
We’re able to do this because of the generosity of our readers. Thank you. If you come here regularly and haven’t yet chipped in — there’s no time like now.
For all naysayers, this is a research project. The data is generates will be useful to determine how much general testing is necessary to determine how well isolation and distancing is working and how long it should be continued. Sure, it would be best for everyone to be instantly be tested and every infected, infectious person immediately isolated. But as usual, the need is not let the best be the enemy of the good enough. A month ago, limited testing ability meant even this small experiment wouldn’t have been possible.
It’s an upside down world these days, the Mission. Folks of course are interested in antibody test which this isn’t. So whatev’s
Both the swab and antibody tests are offered.
JE
It is great that people in this tiny geographic area can be tested, and that alone makes this a worthy effort.
I wish this well-intentioned effort would also be useful in gathering data — but it won’t be. Statistically significant findings require a very robust effort to assure random sampling and also require accurate tests. US tests are still unreliable, and in particular produce far too many false positives. Residents are more likely to test if they think they may have been exposed than if they believe that they haven’t. Residents won’t test if they know that they’ve already had Covid-19, and they are unlikely to take a test if they are fairy sure that they’ve had it. They may not be able to test if they are currently ill. Administering tests in a tiny geographic area creates an added risk that the area either does or does not contain a hot spot. For all these reasons, I’d expect the non-randomized results to be highly suspect, and that they will most likely show a far higher rate of asymptomatic carriers per 1000 residents than actually exists.
What we need to be seeing is a random sample of some representative size of people citywide who get tested every week.
Only by tracking a representative sample can we have any idea about trends in seroconversion, antibodies, reinfection and how subsequent infections, if possible, play out.
Why haven’t we seen capacity ramping up to provide surveillance along these lines?
I think that we’re finding out that the DPH public health operation was right on it at first, but when it comes time to deploy the lumbering non-profit jointed service provision operation, not so much.