Anyone have any ideas (or cites) for estimating the actual number (a/o %) of contagious people walking around in a given region (that is, those not yet isolating themselves, but out and about)? They would have built up a sufficient viral load to be infectious, and not yet be symptomatic, or symptomatic but not yet removed from circulation, or “asymptomatic”? This seems like it would be a pretty useful number (and for those in the US, estimates like it may part of the county-by-county plan announced by White House today).
Without a representative sample across the population there really isn’t anyway to know to any degree of statistical certainty.
FWIW, I made a model of distribution a couple of weeks ago when cases started showing up in Canada, California, and New York based on the publicly available data from Johns Hopkins, and from the data at the time I found the R[SUB]0[/SUB] value to be at least 4 and maybe higher than 6, and the distribution not explainable without asymptomatic or mildly symptomatic carriers. I have a strong if anecdotal suspicion that there are a vast number of asymptomatic carriers and many others who attributed symptoms to mild cold, allergies, or fatigue.
We won’t know the real numbers or even have a good estimate until an antibody test (showing who has had the infection and developed immune response) is made widely available and data is compiled.
Stranger
I wonder if the current number of deaths can be a useful number to drive an estimate? There is a clear disparity between deaths and and confirmed infections across jurisdictions. One assumes that this is largely a result differences in, or simple lack of, testing. But deaths now should be a good proxy for infections roughly two weeks ago. Whilst it is hard to know infection numbers without widespread testing, it is pretty clear when someone is dead. Whether we have enough good parameters to both work out the historical infection numbers and then forward estimate I don’t know.
My gut feeling on this that the situation is far from good.
Would a seriously high R0 value like the ones you said, along with extremely mild cases, point to perhaps a quicker end to this? I remember seeing that herd immunity will start to help when somewhere around 50% (I think it was 20% to 60%) have been infected.
Just on an academic level, if huge percentage of the population already had this or will have it soon, will that help us?
No, it doesn’t change the mortality rates for untreated critical cases, and if we see the same kind of glut of cases coming to a head in a short period, as we saw in Italy and as we are now seeing in Spain (and will soon see in the UK, Sweden, and Russia, among other nations), the ability to treat new cases will drop precipitously to almost zero, resulting in huge mortality. We are on the same trend that Italy was showing two weeks ago, so basically we can expect a similar per capita number of deaths they are experiencing right now in mid-April. And Italy has arguably done a better job of isolation once they realized the extent of the problem than the United States is right now in most states. If Trump has his way of “reopening the country by Easter” all it will do is assure maximum fatalities in the people who have not yet been exposed.
If you want analysis based on the best available data, I give you Dr. John Campbell — Thursday, 26 March. “Let’s just say that it’s less than encouraging.” The dryly understated wit of the British in the face of existential horror.
Stranger
With widespread concealed carry of firearms, we must assume that anyone we encounter in USA is armed and dangerous, even if nude. With COVID we must assume that anyone nearby is contagious except maybe the dead. Are morgue workers contracting COVID?
As mentioned, reliably estimating a contagious proportion is unlikely without ubiquitous testing and re-testing. Typhoid Mary unknowingly spread her contagion; are similar Covid Charlies innocently spewing disease? We also read of fucktards caught trying to spread the virus. How many willful carriers haven’t been caught?
It seems to be going something like Dengue, 80% of all cases show no symptoms. or mild nonspecific ones. However, the 20% who do get symptoms, get it really badly. Its usually not fatal though, (numbers I have seen indicate something like 0.1% death rate but the “down” period for a sufferer is weeks. It usually quickly overwhelms healthcare systems.
The saying about it is that “it doesn’t kill you, only makes you wish it would”. Maybe COVID will become like that.
(I know one person who just recovered. He suffered a gunshot wound in the Army a few years ago. He said he felt much worse with COVID :eek: )
I’ll watch the video a bit later since I don’t have 30 minutes right now, but maybe that answers my question. I’m more asking about how long we’ll be on the upside of this.
If we double every 4 days and need to get to 170M infected and are at 80k right now, that implies 40 to 44 days to get to that number. That puts us nearing a downturn towards the end of April, not an upswing as the original paper predicted. Your numbers seem to indicate a peak sooner. So the question isn’t about the mortality rate, it is about the timing of the wave.
To the OP:
Cited on these boards by several before is this Science article which estimated that in China at the beginning of the epidemic, before control measures, and then after.
First the before.
They further felt that those cases were contagious for on average 3 1/2 days each.
The number of undocumented went down after restrictions went in place.
Unclear if the virus or the level of presumptive identification has changed since.
To cmosdes’ question and Stranger’s reply -
Yes cmosdes your hypothetical would change the curve dramatically, as would how we handle children in the models. The initial portion of the curve would be the same in all cases that do not include any control efforts, and yes, “the mortality rates for untreated critical cases” would still be as high, but the flattening out portion of the curve where herd immunity begins to play a role would be reached much faster and at a lower peak of sick people. Which control methods would be needed to achieve the avoidance of overwhelming the healthcare systems surge capacity would possibly be different depending on different inputs.
I don’t want to pick on anyone in particular because there is a lot of misinformation that is being spread around even by reputable media outlets but a few things should be made clear:
[ul]
[li]We cannot rely upon past experience with outbreaks in living memory as guidance for what to do in this pandemic. The SARS-CoV-2 has a combination of mortality and transmissibilty in a way not seen since the poliomyelitis epidemics in the 1940s and 1950s or the 1918 Spanish Flu pandemic. The normal means by which to prevent community transmission–hand washing, not sharing food or utensils, et cetera–which work well enough to contain normal influenza outbreaks or lethal but mildly contagious epidemics such as Ebola are good measures to follow but are not adequate here where there is clearly untraceable community spread via aerosol vectors. [/li][li]“Flattening the curve” is intended to prevent a glut of cases from occurring all at once, and while that is most beneficial to the at-risk populations, it requires everybody to try to prevent transmission. Given how contagious this virus is it will do very little to protect the people who will suffer critical illness if they contract the disease regardless of the current treatment, but it gives the people who can be saved with with treatment a fighting chance, and also allows time to develop and test more effective treatments for at-risk individuals and deploy widespread testing to find out who is infectious and who is immunized.[/li][li]“Herd immunity” is a real thing but from a practical standpoint it applies mostly to endemic disease like influenza where ensuring that there aren’t enough people able to spread the pathogen for it to get a serious foothold. If half of all people get a flu shot and follow good hygiene practices (and are allowed to stay home with guaranteed sick leave instead of being forced to decide between working or going hungry) then the virus just can’t get to enough people to spread effectively. SARS-CoV-2 is uniquely suited to rapid, pervasive, and nearly undetectable (by signs and symptoms) spread through the population so the normal degree of herd immunity to prevent epidemic outbreak (typically estimated at 30%-50% of the population) does not apply here. [/li][li]Relaxing isolation measures as soon as the curve starts to flatten will just ensure that the infection and mortality rates start to ramp up again. To be effective, isolation has to last through the entirely latency period (so 2-3 weeks) that the virus may be resident without apparent effects. People who are at risk should double that time at a minimum and remain vigilant until effective treatments or vaccination is available. [/li][li]To make any realistic estimate of how many are immunized, we need mass antibody testing of a large sample of the population (at least 5% of the population across demographics for statistical reliability to an 80% or higher confidence level). This–not “reopening businesses by Easter”–should be the focus of the government in doing the most effective actions to return to a something-like-normal commerce and social environment. [/li][li]“Keep distance to preserve our existence,” should be the mantra you are saying to everyone who is dismissive or obstructive in social distancing and isolation measures. Handwashing, spraying down surfaces with disinfectant, and other hygiene measures are all good but with evidence of wide scale community transmission and apparent aerosol transmission, keeping distance is the only certain way of preventing further transmission. I’m sorry if this ruins your social life, and it is truly terrible for those who make a living in customer-facing service and support industries where it is a choice between getting a paycheck and protecting yourself, but to the maximum extent possible you need to isolate, and if you are in a public-facing job, isolate yourself from household members who can distance themselves even if they are not personally in an at-risk demographic, because you never know if they may have some unknown underlying condition or could convey the virus to someone else who is at-risk.[/li][/ul]
I’ve been sharing this far and wide, but for those who aren’t familiar: Dr. John Campell, a teaching physician in the UK who is providing non-sensationalist, fact-based analysis of how the COVID-19 outbreak is progressing in various countries with comparison between those who took prompt distancing and isolation measures and those which haven’t. I know, he’s dry, he does a lot of math, and he doesn’t have a cool beat in the background, but if you want the absolutely unvarnished, not-apocalyptic-but-not-sugarcoated assessment of how the pandemic is affecting people around the world, this is what you should be following.
Stranger
Thanks for the article.
So for an initial stab, just to get some idea:
Consider two (time) windows, corresponding to the two main groups (“bad” and “mild” cases). The windows are in days from (say) the day of infection, between first being contagious, and being taken out of circulation:
(a) for the conscientious overtly symptomatic people, between becoming contagious and removing themselves from circulation
(b) for the mildly/asymptomatic people, between becoming contagious to becoming non-contagious.
If I understand it correctly, the article suggests window (b) is approx from days 3 to days 6 from the time of infection. For simplicity’s sake suppose (a) is as well (by approx day 7 they feel bad enough to start isolating).
So if we could get (or estimate) the number of newly infected people between 3 days ago to 6 days ago in a given region, and add those values up–that would be a rough est for how many are contagious and circulating.
(This question came up regarding trips to, say, the grocery store, where there can be hundreds of people gathered: how many (on average) might be carriers?)
Stranger - you seem to know more than the experts -
-
Who do NOT believe that aerosols are a significant factor in transmission (mostly droplets).
-
Who understand that a curve can be flattened by multiple means. Keeping those who use the resources the most if they get sick from getting sick, is a means of flattening the curve. The issue is that with the current sets of assumptions it would not flatten it enough, and we don’t know enough to be able to say that those assumption are definitely not true. Theoretically if in Italy those uninfected over 70 were isolated early on and kept from catching the virus, Italy’s curve would have been significantly flattened. Just not realistic to achieve the goal with that alone.
-
Who understand that herd immunity is NOT something that only applies to endemic diseases and is a critical part of modeling the dynamics of every epidemic or pandemic. Local herd immunity is what broke the spread of Zika in Brazil, for example. Herd immunity is what happens under a flatten the curve scenario as well, just slowly and controlled enough to avoid overwhelming the healthcare system at its peak.
-
Who understand that good confidence intervals do not require testing 5% of the population with decent random sampling techniques. In fact [the WHO protocol for such studies](file:///Users/donseidman/Downloads/covid-19-seroepidemiological-investigation-protocol-v3%20(1).pdf) (see section 4.1) discusses the 95% confidence intervals with just 100, 200, and 300 samples at different levels of true seroprevalence.
There is indeed a whole mess of misinformation going on. Claims that COVID-19 is primarily aerosol spread (other than only aerosol in very specific high risk circumstances), claims that herd immunity does not apply to it, claims that studies need 5% of a population to be valid … lots of misinformation.
Given the information we don’t yet know about COVID-19, and its current rate of rise, the only prudent course is to maintain “the hammer” right now. No question that arbitrary dates should not be how decisions of what to modify in the approach when are made, but additional information as it is learned, and the response of disease in our population to the interventions.
Nevertheless to THE HYPOTHETICAL that cmosdes posed - yes, experts (like Mark Lipsitch) believe that for this virus 50% of a population immune/resolved would be sufficient to achieve herd immunity and stop spread (go to 9:20 on, especially at the 10:55 mark). Lipsitch takes pain to point out that case fatality rate is not the same as infection fatality rate and that we simply do not know what the latter is. Note he offered other social distancing option than full hammer approach as possibly effective.
IF, theoretically but with some reason to believe, children functioned the same as those resolved, then the 21 to 27% (depending on if you want to define child as under 15 or under 19 yo) of NYC that are children, get you halfway there. IF, also theoretically but with some reason to believe, there are 8 asymptomatic to symptomatic ones, then herd immunity and flattening of the growth would occur when symptomatic (not confirmed) infections reached about 270,000. Which would be by three weeks later would have 2700 total deaths before the daily death rate started to slow down. Which would be in a week, and would correspond with ICU admissions slowing down within the next few days. Which of course might occur due to social distancing beginning two weeks ago as well.
If hypothetically children are actually equally as contagious as adults, and there are no asymptomatic cases … well then herd immunity is a long long way off. Any flattening is exclusively because of the hammer approach in that case.
Reality is likely somewhere between the two. But none of us know.
A hundred or so New Yorkers in each decadal age cohort tested for serologies according to the WHO protocol would be able to at least give a broad answer to which of the two are closer to the truth, at least on the asymptomatic infection side.
Settimo, I’m not sure how to do the math you propose, but from a practical POV you are best off considering any individual outside your immediate household circle as if they were contagious and limit your time within their 6 foot circle and the frequency of those contacts as best you can. Along with washing your hands, etc.
Sorry, never entered the Lisitch link.
So, this thread seems to be going down a familiar path of being more about personally disproving me than a factual discussion about the issue at hand, so I am going to bow out of the thread but I do want to address a few of the points of DSeid’s post in as even and rational a manner as possible.
I have not said and do not believe that I “know more than the experts”. I do enough to know that the official numbers about the replication rate–based on data reported from China, and on infection reporting rates in various countries all of which lack comprehensive testing for even symptomatic cases–is not consistent with what we are seeing in terms of how quickly and widely the epidemic has spread, and how rapidly cases are appearing en masse. This isn’t an issue of health officials deliberately misleading the public but simply a lack of objective data upon which to base a firm estimate.
The range of transmission rates that I estimated–which was based upon some speculative modeling that tried to reproduce the trends seen in the first weeks of reported cases in the US, and thus, is neither precise or firmly grounded enough to take as anything but a caution–was based upon the very sparse data available at the time and while I haven’t bothered to update it because the data we have now is even more unreliable in reporting total infections due a criminal lack of testing it is clear that both that there are a lot of asymptomatic and marginally symptomatic cases which have at least been qualitatively verified by testing, and that the mode of transmission cannot simply be by intimate contact or fomite transmission that would support this rate of expansion, and droplet transmission would require carriers to exhibit symptoms (coughing, sneezing) in order to distribute droplets so widely.
That we are now seeing significant numbers of medical personnel who are becoming infected despite wearing all recommended PPE which would indicate that the normal measures to prevent fomite and droplet transmission are not adequate. This is also indicated in the transmission on the Diamond Princess where the contagion continued to spread even though passengers were confined to quarters and the crew were asymptomatic, suggesting that either the crew or the HVAC system were facilitating spread despite the isolation and hygiene measures in place. There is evidence that closely related SARS-CovV(-1) virus responsible for the 2002-2004 outbreak could be transmitted by aerosol modes, so this is unsurprising.
The Zika virus is primarily spread by mosquitos with a secondary mechanism of intimate contact (unprotected sex), and of course from mother to fetus. I haven’t been able to find a reliable estimate for the replication rate of Zika but because it resides in an insect host it is considered endemic, and the massive concern about it during the 2016 Rio Olympics was less that there would be an unconfined outbreak that would spread to nations in temperate areas than that it would be unknowingly contracted by attendees and athletes and lead to a series of miscarriages and congenital brain abnormalities such as microcephaly in the fetuses of pregnant women.
Herd immunity prevents the uncontrolled spread of disease but as long as there are reservoirs–and we don’t really know endemic the SARS-CoV-2 virus is going to be in the population once this epidemic wave is over, or whether additional epidemics may recur–and the virus is extremely contagious (which SARS-CoV-2 gives every indication of being) this does not protect at-risk individuals from the potential for being infected. This is why despite recommending that the vulnerable (very young, elderly, and immunocompromised) are recommended to get the annual influenza vaccine the are still warned to take measures to prevent infection and to self-isolate during regional flu outbreaks. We don’t actually know how long or how well acquired immunity or hypothetical immunization will limit the virus in the future (although there are some indications that the virus does not mutate fast, which if true is hopeful), but regardless, those people who are in at-risk demographics are going to have to maintain isolation for an extended period of time until widespread antibody testing is available and there is a clear indication of how many people could still carry and transmit the virus.
Small sample sizes are statistically representative of a homogeneous population, but the United States is not one big bag of interchangeable marbles that we can estimate from a small sample size. We see now that not only do affected cities have different rates of COVID-19 incidence but the demographic distributions (age, sex, wealth) are distinct as well, with the only real commonality that young children without underlying conditions have almost no incidence and the mortality starts to ramp up dramatically after age 60. Every public health official I have seen talk about dealing with the issue of limiting and ending the epidemice has specifically called out the need for wide scale testing to get a handle on how big the problem is and how many people may be immune.
I haven’t been able to view the video you linked to in the subsequent post because the website keeps hanging and reloading so I don’t know when the presentation was made, what data it was based upon, or how the conclusions were arrived at but if this is Dr. Mark Lipsitch of Harvard School of Public Health, I would certainly acknowledge his expertise in this area as a primary authority. In this interview on 02 March he indicates that the ultimate global infection rate may be 40%–70%, so at least of that time he clearly didn’t seem to feel that 50% would be the limiting threshold after which herd immunity would effectively limit further transmission.
As for assuming that if all children are immune that conveys half of the supposed 50% threshold for herd immunity, that just isn’t the way herd immunity works. That distribution has to be spread across the entire demographic such that there is no large scale transmission between anyone. If all children are immune but, say, middle-aged stockbrokers are generally not, then you will still see an epidemic of middle-aged stockbrokers transmitting and contracting the disease. This is, again, why wide scale testing is crucial to be able to track the amount of infection and immunity so we can know when lockdown and isolation measures can be lifted without great risk of another epidemic wave.
Despite all of the bad news we are seeing now in New York and New Orleans (and soon in Los Angles, Chicago, and essentially every large and mid-sized city in the United States), and in other countries around the globe that have not acted promptly to limit contagion and prepare for the onslaught of COVID-19 cases, there is some good news. The case fatality rate is about 1%-2% overall for people exhibiting serious symptoms of COVID-19, so while millions of people around the world are going to die (many of whom could be saved if medical care was available), this is not an existential threat that is going to disrupt food production and logistics in developed nations. It is a major catastrophe for developing nations and those with very dense and impoverished populations where quarantine is not possible and medical services are limited to non-existent. While front line medical personnel have limited and inadequate PPE and are getting hit with serious infections of their own, those who do survive will be able to treat patient using less effective PPE, and if immunity can be demonstrated and verified with testing, people who have been unemployed may be able to be trained in basic patient care and non-care duties to free up nurses to focus on more critical care duties. And there are indications that the virus does not mutate rapidly which may mean that an eventual vaccine and immunization campaign make eradicate the virus or at least render it sub-endemic such that outbreaks are rare and sporadic rather than periodic or seasonal.
I’m going to bow out of this thread now so the discussion can continue without the toxicity of just trying to prove that I’m wrong in my estimates (which is entirely possible given the poor data and speculative method) or that I’m not an expert (absolutely true, and an impression I never intended to give) and therefore what I say must be wrong. I would ask that others use evidence and fact-based reasoning with the acknowledgement that we are short on data and even the most vaunted experts lack the complete picture to make exact predictions on where this will go and how long it will last, which I think all of those experts have made clear in their qualified language and discussion of the unknowns.
Stranger
Re: aerosols, what I’ve read to date indicates that the virus can spread through aerosols, but that this is most likely to happen to healthcare workers during intubation. It’s not how it typically spreads in public, though.
What does seem to true, though, is that ‘air droplets’ can be exhaled simply by breathing, so if you’re sitting in close proximity to someone, or if you’re in a large meeting room and walking around meeting people for an hour, you’re in danger of contracting the disease even if nobody shows classic symptoms like coughing or sneezing.
It seems to me that the spread of the disease and the actual percentage rates of infection/mortality are really only estimates (but damn good ones) at this time, because of the horrendous lack of testing. The true epidemology of COVID-19 cannot be known until contacts can be tested and the routes of infection adequately traced.
~VOW
It’ll be impossible to get a true count of the total numbers of symptomatic and asymptomatic people actually infected until much, much later. It’s not really that important to know the number of asymptomatic people to understand the lethality. What doctors are concerned with is what are someone’s chances once they start presenting with signs of obvious infection. From what I’ve read, those numbers are estimated to be anywhere from 1-4% generally.
Stranger - if this is a familiar path you’ve experienced in threads then perhaps you should consider the possibility regarding the common denominator?
Assuming that Stranger meant their “bowing out” there is, in my mind, still a need to address things stated by the poster as facts that are at best speculations that go against the current experts’ assessments, which is of course fine to do, so long as you identify them as such.
The experts rather strong belief is that, outside of certain procedures which produce aerosols, transmission is mostly droplet transmission. Again fine to argue why one thinks the experts are mistaken, if one does think that, but one is doing a disservice to readers to state as if it is fact that transmission is by aerosols when expert opinion is very much NOT.
Again I defer to the WHO who believe that smallish studies can give reasonable confidence intervals on seroprevalence and that a study of 5% of Hubei (4.5 million people) is not needed to get a sense of how many have already had the disease and are presumed immune. By reported cases only 0.1% of Hubei had COVID-19. If a mere random 300 are tested and 20% are positive, then one might be unsure if the true seropositivity rate is 26 or 14%, but can state with high confidence that the true rate is at least one to 2 orders of magnitude greater than the reported case number.
The item that very much need correction though is on how herd immunity works. Models that predict herd immunity do NOT require homogenous mixing of the population. In point of fact most standard models if anything overweight the impact of children in spreading the disease because they are the Typhoid Marys of many infections: they have many more close, inclusive of physical contacts with others, both intra- and inter-generationally, than other age cohort individuals, and they literally spew out their germs over all surfaces they are near. Not specific to respiratory infections, but my old ID prof made it clear that if you are in a household with a child, especially one prechool aged or younger, just accept the fact that no matter how much clean there IS a microscopic layer of feces on every surface you touch. They are a very, likely the most important quarter to have not contagious, if such is conclusively proven to be the case. It is why, based on influenza dynamics, school closures are a standard part of the pandemic response kit.
It is well past the point for widespread RT-PCR testing, let alone contact tracing and testing, to be useful. We have reached the point where significantly symptomatic COVID-19 is more often a clinical diagnosis (especially given, in our region at least, the low rate of other viruses with similar presentations testing positive, with low influenza and RSV rates), and those numbers are not being collected. They are however quite large, many times larger than the number of confirmed cases. The number of minimally symptomatic to asymptomatic per each of those who would fit a clinical diagnosis is even more of a guess, even if it is an educated one. Population level seropositivity testing is going to be the only way to get any handle on that number now. (IMHO.)
Totally agree with those precautions. And re the following, I wouldn’t want anyone to make important health decisions based on a random person’s calcs on the internet. If anything, reasonable estimates along these lines should give people good reason to follow the precautions. With that disclaimer in mind, here’s one method (this gets slightly mathy):
From post #11, an estimate is needed on the number of new infections between 3 days ago and 6 days ago (inclusive)–this approximates the # of infected people in circulation (not isolating). If today is day zero, yesterday is day minus 1, etc. then we want those infected on days -6 to -3.
The dashboard site(s) (which I think are global) give regional daily numbers of confirmed cases (all the way to district/province/county, in some cases).
The lag: the dashboard sites give tested/confirmed cases. The lag is the amount of time from getting infected to getting the test result onto the dashboard. The article mentioned above suggests between 6 and 10 days from infectiousness to positive test report, so add 3 for approx start of infection. I’m going to conservatively estimate an average lag of 10 days from infection to reporting–ie todays dashboard value estimates total people infected as of approx day -10. The approach is to accumulate a few days worth of dashboard totals, and extrapolate.
To give an example, suppose I recorded the dashboard site amounts for my county/district/whatnot over the past 6 days (incl today), to give daily totals {48,71,92,110,172,221}. These would be approx cumulative # of people infected by days -15 to -10 (respectively).
Plot it: it looks exponential, so use linear regression on the logarithms of the values (site) (if the plot looks more linear, just apply linear regression to raw values (no logs)). The logs (base “e”) are approx {3.9, 4.3, 4.5, 4.7, 5.1, 5.4}. Linear regression gives the equation Y=0.3X+3.9 (X=days from day -15). To get back to normal scaling, take exp: exp(0.3X+3.9)=49*1.35^X. This amounts to estimates for what the dashboard will say (in future) for days (relative to today) -9 through 0:
{49,67,90,122,164,221,299,403,545,735,992,1339,1808,2441,3294,4447} (day -15 to day 0)
The other variable is the efficiency in testing–how many with the disease actually get tested. I’ve seen ratios actual/tested anywhere from 6 to 11 (not sure if this includes the asymptomatic cases; the “11” was from a NYT article I can’t relocate); to be conservative, use 8.
To estimate the new infections that occurred in the window, take the difference between days -7 and -3 (trailing difference–we want new infections on days -6 thru -3 inclusive; could also do between -6 and -2, or average trailing and leading):
8*(1808-545)=10104
If the county has pop 500K, then that’s an estimated 2% of the population who are out and contagious. (Yikes.)
The combination of the lag between infection and test results, the exponential growth, and the testing undercount has a big effect. I would not have thought such small numbers in confirmed disease totals could translate into that high of a percentage in circulation.
In South Korea, where they managed quite well, largely by testing LOTS, they found 30% of those who tested positive, had no symptoms.
That’s a scary number indeed.