Throughout the SARS-CoV-2 pandemic, the New York Times, regarded as America’s “newspaper of record”, has been the standard bearer for propagandistic reporting serving to manufacture consent for extreme and harmful “lockdown” policies by contributing to the sense of fear and mass panic among the public. To that end, the Times has consistently reported about viral transmission in an alarmist manner, delivering fearmongering messages that grossly misrepresent the science.
In part one of this series, we saw how the Times on March 17 characterized the science has having proven that the virus is airborne transmissible. Since the World Health Organization (WHO) maintained that this mode of transmission remained theoretical, the Times characterized it as having stubbornly refused to acknowledge the science. Yet, by examining the Times’ own cited sources, we could see that the WHO was right.
In part two, we saw how the Times on March 31 similarly characterized the science as having firmly established that a fifth or more of community spread is driven by people who have no symptoms. Yet, not only did the Times fail to produce any studies to support that claim, but it went so far as to deceptively cite a study as proof even though its authors explicitly stated that there remained no clear evidence of asymptomatic transmission.
In part three, we saw how the Times on June 5 falsely characterized the WHO as having belatedly and reluctantly endorsed government policies recommending—or, as is the case in numerous states, mandating—the universal use of masks in the community setting. In fact, the WHO had rightly observed that such policies are not evidence-based and instead reasonably advised that masks be worn by members of the public in circumstances in which there is community spread and prolonged close contact with others is unavoidable.
In sum, the narrative pushed by the Times is that community spread is largely driven by asymptomatic individuals including through virus-laden aerosols, as distinguished from larger respiratory droplets that scientists believe to be the primary mode of transmission. This narrative instills fear in people that they could become infected, for example, by momentarily passing an unmasked person in a supermarket aisle, by entering an airspace someone else had occupied several minutes earlier, or even by just being in the same building with despite maintaining a considerable distance from an infected person.
While the March 17 article included the caveat that, even if airborne, a person would still need to be within a few feet of an infected person for there to be a significant risk of transmission, this acknowledgment came only later in the article, with the headline and first half of the piece serving to instill fear. In the March 31 article, the Times further acknowledged that a passing close encounter would not be enough and that the close contact would need to occur for a prolonged duration for the risk of infection to be high. Hence, the WHO’s mask guidance made sense even if airborne transmission is assumed to play a significant role in community spread.
Yet, because information from the WHO contradicted the story that the Times preferred to convey to the public, the author of all three of those articles, Apoorva Mandavilli, set out to sustain the fearmongering narrative by further mischaracterizing both the WHO’s position and the science in a June 9 piece titled “In the W.H.O.’s Coronavirus Stumbles, Some Scientists See a Pattern”.
- How the New York Times Characterized Asymptomatic Transmission as Common
- The New York Times’ Self-Contradictions
- The WHO Did Not “Walk Back” Its Statement on Asymptomatic Transmission
- Studies Had Not Estimated the Proportion of Asymptomatic Spread at 40 Percent
- The Distinction Between Asymptomatic and Presymptomatic Transmission is Not Semantic
- The Times’ Cited Estimate of Presymptomatic Transmission Had Important Limitations
- The WHO Had Rightly Noted the Limited Evidence on Mask Effectiveness
- The WHO Had Rightly Noted Airborne Transmission Remained Theoretical
How the New York Times Characterized Asymptomatic Transmission as Common
Below the headline, the article’s description elaborated that “The agency’s advice sometimes lags behind rapidly evolving research into the coronavirus, experts contend.”
Underneath that, the Times featured a photo of a WHO official with the caption, “Initially, Dr. Maria Van Kerkhove, the technical lead of the W.H.O’s coronavirus response, downplayed the role of asymptomatic spread in the pandemic.”
The article then states that, during a press briefing on June 8, Dr. Van Kerkhove “asserted that transmission of the coronavirus by people without symptoms is ‘very rare.’ Following concerted pushback from researchers, officials on Tuesday walked back the claim, saying it was a ‘misunderstanding.’” (Emphasis added.)
This was “not the first time the W.H.O.’s assessment has seemed to lag behind scientific opinion”, Mandavilli continued. The WHO had “delayed endorsing masks for the general public” until June 5, “claiming there was too little evidence that they prevented transmission of the virus.”
The WHO had also “said repeatedly that small airborne droplets, or aerosols, are not a significant factor in the pandemic’s spread, although a growing body of evidence suggests that they may be.”
The “scientific disagreements have wide policy implications” since many countries, including the US, had “adopted lockdown strategies” based on the assumption that community spread is largely driven by asymptomatic individuals through the airborne route, which places people at risk when “congregating in poorly ventilated spaces”.
The Times article continued (all emphasis added):
The W.H.O.’s thinking on asymptomatic transmission does not appear to have changed much since February, when the W.H.O. China Joint Mission reported that “the proportion of truly asymptomatic infections is unclear, but appears to be relatively rare and does not appear to be a major driver of transmission.”
Studies later estimated this number could be as high as 40 percent; the current best estimate from the Centers for Disease Control and Prevention [CDC] is 35 percent.
The research prompted many countries, including the United States, to endorse use of masks by everyone.
But on Monday, Dr. Maria Van Kerkhove, the W.H.O.’s technical lead for coronavirus response, said that “it still seems to be rare that an asymptomatic person actually transmits onward to a secondary individual.”
Her statement provoked an immediate backlash from scientists, who noted that study after study had shown transmission of the virus from people before they ever felt symptoms.
The reaction, the Times continued, prompted Dr. Van Kerkhove to “clarify” the WHO’s position by saying that her comment “had been based on only two or three studies.”
The Times then quoted Lawrence Gostin, director of the WHO Collaborating Center on National and Global Health Law, criticizing the WHO for having “come out with things that are clearly contradicted by the scientific establishment”, which “significantly reduces their credibility.”
Next, the Times reported:
A key point of confusion is the difference between people who are “pre-symptomatic” and will go on to develop symptoms, and those who are “asymptomatic” and never feel sick. Dr. Van Kerkhove suggested that her comments were about people who are truly asymptomatic.
A widely cited paper published in April suggested that people are most infectious up to two days before the onset of symptoms, and estimated that 44 percent of new infections were a result of transmission from people who were not yet showing symptoms.
To characterize the WHO’s distinction between asymptomatic and presymptomatic individuals as practically insignificant, the Times then quoted Dr. Ashish Jha, director of the Harvard Global Health Institute, saying, “OK, technically fine. . . . But for all intents and purposes, they are asymptomatic—they are without symptoms.”
The WHO also continued to maintain “that large respiratory droplets expelled by sneezing or coughing are the main route of transmission and to downplay a possible role for aerosols, smaller particles that may linger in the air.”
“But evidence is piling up”, the Times said, “that aerosols may be an important route.”
The article then quoted Linsey Marr, a professor of engineering at Virginia Tech, saying that “coughing and talking, even breathing in some cases,” produce infectious aerosols.
The Times also quoted Dr. Benedetta Allegranzi, the WHO’s technical lead on the coronavirus, saying that, to date, there had been “no demonstration of transmission by this type of aerosol route”.
Then the Times cast doubt on that statement by saying, “But the W.H.O. defines airborne transmission too narrowly, some scientists said. Airborne transmission also includes the possibility that the virus is aloft for shorter distances, then inhaled.”
According to the Times, experts criticized that the WHO “should be more cautious in concluding that transmission by air or people without symptoms is not significant.”
“We don’t know,” the article quoted Dr. Don Milton, a professor of environmental health at the University of Maryland, as saying. “But they also don’t know.”
The article concluded,
Some experts said that when the W.H.O. uses the phrase “there is no evidence” to indicate uncertainty, it is in fact conveying certainty about the absence of a phenomenon.
Dr. Van Kerkhove conceded that point.
“That’s a fair statement,” she said. “There’s a lot of research that needs to be done to really understand this, and we are open to the fact that there is new research every day.”
The New York Times’ Self-Contradictions
The first observation to be made about this New York Times article is how it contradicts Mandavilli’s earlier reporting.
In her March 17 article, she had characterized studies as having proven the airborne transmissibility of SARS-CoV-2, with the question remaining of how long infectious aerosols remain in the air.
In fact, as could be seen simply by examining her own sources, the question remained whether viable virus could be detected in aerosols produced by coughing, sneezing, speaking, or breathing. A key fact she omitted from her earlier article was that the detection of viral RNA in air samples is not necessarily indicative of the presence of infectious virus.
Yet, now, in her June 9 piece, Mandavilli was acknowledging that the airborne transmissibility of SARS-CoV-2 remained theoretical, with evidence “piling up” that the virus “may be” airborne.
Yet, now, she was acknowledging that scientists “don’t know” the extent to which asymptomatic people were responsible for the spread of the virus.
Mandavilli also once again acknowledged the difference between transmission from infected individuals who never develop symptoms and transmission occurring prior to symptom onset. Yet, to sustain the preferred narrative, Mandavilli sought to trivialize the distinction and persisted in deliberately conflating the two terms to misrepresent both Dr. Van Kerkhove’s statement and the science.
The WHO Did Not “Walk Back” Its Statement on Asymptomatic Transmission
As Mandavilli tells it, Dr. Van Kerkhove initially “downplayed the role of asymptomatic spread” by saying it was “very rare”, but this sparked outrage by scientists because many studies had shown that asymptomatic spread was common, and so Dr. Van Kerkhove subsequently “walked back” the statement by saying it was a “misunderstanding”.
But that is false.
Contrary to the article’s headline, the real pattern here is the Times’ insistence on concocting stories and lying to its readers in ways that contribute to the sense of fear and mass panic necessary for consent to be manufactured for extreme lockdown measures.
It is true that there was a misunderstanding about Dr. Van Kerkhove’s statement. However, this was not due to miscommunication on her part but rather to the fact that her statement was taken out of context by the news media—including by Mandavilli in this very New York Times article.
In fact, Mandavilli persisted in taking Dr. Van Kerkhove’s statement out of context despite the WHO official’s effort to correct the misreporting of her statement that had occurred.
The nature of the misreporting is alluded to further into Mandavilli’s article where the distinction between asymptomatic and presymptomatic transmission is acknowledged. According to Mandavilli, it was only later that Dr. Van Kerkhove clarified that she was speaking about transmission from “truly asymptomatic” individuals.
But that is a deception.
During the June 8 virtual press conference, Dr. Van Kerkhove was asked whether asymptomatic transmission might play a bigger role than the WHO had previously indicated by saying that there had been no documented cases of this occurring.
In her answer, Dr. Van Kerkhove began by explicitly distinguishing between cases “reported as not having symptoms” and those who “are in their pre-symptomatic phase which means it’s a few days before they actually develop severe symptoms.”
She then explained that, when WHO officials had sought to determine whether reported cases “were truly asymptomatic”, they discovered that many were not without symptoms but had “really mild disease”.
There were cases in which people had tested positive for the virus yet remained “truly asymptomatic”, she explained. But a number of reports from countries doing detailed contract tracing and following such cases were “not finding secondary transmission onward.”
Noting that she was drawing partly on studies “not published in the literature”, she described the data as indicating that transmission from truly asymptomatic individuals appeared to be “very rare”.
“We are constantly looking at this data”, Dr. Van Kerkhove continued, “and we’re trying to get more information from countries to truly answer this question. It still appears to be rare that an asymptomatic individual actually transmits onward.”
After emphasizing the importance of stopping transmission from people with symptoms, she reiterated, “But from the data we have it still seems to be rare that an asymptomatic actually transmits onward to a secondary individual.”
Hence, contrary to the false impression Mandavilli leaves Times readers with, when Dr. Van Kerkhove later responded to the outcry by emphasizing that she was referring to individuals who were truly asymptomatic, she was not walking back but reemphasizing what she’d said in the first place.
Studies Had Not Estimated the Proportion of Asymptomatic Spread at 40 Percent
Mandavilli took the deception further by quoting from the report of the WHO-China joint mission, which had been published in February, and then deliberately conflating symptomatic and presymptomatic transmission despite the fact the WHO had made clear that it distinguished between the two.
According to the Times, in the context of the WHO’s “thinking on asymptomatic transmission”, the WHO report had said that “truly asymptomatic” individuals did “not appear to be a major driver of transmission”, yet studies later estimated “this number” to be as high as 40 percent, and the CDC’s “best estimate” placed it at 35 percent.
But what did Mandavilli mean by “this number”? The statement quoted from the WHO report referred to “the proportion of truly asymptomatic infections”, so by “this number”, did she mean the proportion of truly asymptomatic infections or truly asymptomatic transmission?
Confusingly, as will be shown momentarily, she meant both, which was in keeping with how she had fallaciously equated the two in her March 31 piece. What is clear, though, is that she was not being honest with Times readers.
In its February report, the WHO had once again clearly distinguished between asymptomatic and presymptomatic transmission. In the sentence prior to the one quoted by Mandavilli, the WHO had stated that “Asymptomatic infection has been reported, but the majority of the relatively rare cases who are asymptomatic on the date of identification/report went on to develop disease.”
Hence, it was obvious that when the WHO used the phrase “truly symptomatic” in the sentence that followed, it meant individuals who did not go on to develop symptoms.
Yet to falsely characterize that statement as having been contradicted by later studies, Mandavilli presented the figure of 40 percent, for which she cited her March 31 article, titled “Infected but Feeling Fine: The Unwitting Coronavirus Spreaders”. In that article, she had reported that “A team in Hong Kong suggests that from 20 to 40 percent of transmissions in China occurred before symptoms appeared.” (Emphasis added.)
Consequently, by “this number”, Mandavilli meant the proportion of presymptomatic transmission yet falsely characterized the figure as contradicting the WHO’s statement about asymptomatic transmission. That it was her intent to deceive is a difficult conclusion to escape given the fact that she clearly understood that the WHO was distinguishing yet persisted in deliberately conflating the two anyway to try to discredit the WHO’s repeated statements that asymptomatic transmission appeared to be rare, which in turn contradicted her earlier reporting.
At the same time, by “this number”, Mandavilli also meant the proportion of asymptomatic infections, thus illustrating once again how she fallaciously equated this proportion with the proportion of asymptomatic transmission. This is evident in the fact that she cited the figure of 35 percent from the CDC.
Although the Times did not provide a link, the CDC’s “best estimate” had been provided on May 20 in a documented titled “COVID-19 Pandemic Planning Scenarios”. The figure of 35 percent did not refer to an estimated proportion of asymptomatic transmission but rather asymptomatic infections.
The CDC had also, like the WHO, explicitly distinguished between asymptomatic and presymptomatic transmission and estimated that 40 percent of transmission occurs “prior to symptom onset”; the CDC did not provide an estimate of the proportion of asymptomatic transmission.
Consequently, neither of the two figures that Mandavilli presented as though contradicting the WHO’s “thinking on asymptomatic transmission” were in fact contrary to the WHO’s statement that it appeared to occur only rarely.
The way the Times deceptively portrayed it, Dr. Van Kerkhove’s statement that asymptomatic transmission appeared to be very rare was based on just two or three studies when “study after study” had found otherwise. The implication is that numerous studies had contradicted her statement by demonstrating asymptomatic transmission to be common.
The truth is that the WHO was relying on a small number of studies because so few studies had documented asymptomatic transmission, and the Times’ claim that numerous studies had contradicted the WHO’s statements about asymptomatic transmission is false.
The Distinction Between Asymptomatic and Presymptomatic Transmission is Not Semantic
Presumably done in part to justify her own deliberate and deceptive conflation of the two terms, Mandavilli also went on to characterize the distinction between “asymptomatic” and “presymptomatic” transmission as having no practical significance, including by citing experts who criticized the WHO for doing so.
Yet, the Times could just as well have sought out experts who agreed with the WHO that it’s important to distinguish between the two, such as officials from the CDC, which, as already noted, did the same.
For another example, expert reactions to Dr. Van Kerkhove’s statements were published by the Science Media Centre the same day as the Times article. Among them was the observation from Professor Babak Javid of the Tsinghua University School of Medicine in Beijing, China, that Dr. Van Kerkhove had made “the important distinction” between asymptomatic and presymptomatic individuals and that it “may well be true” that asymptomatic people “rarely transmit”, as suggested by detailed contact tracing in Taiwan and Germany.
Other experts acknowledged the uncertainty and the lack of evidence to the contrary. “How much of a role asymptomatic transmission plays in the total number of new infections”, said Keith Neal, Emeritus Professor of the Epidemiology of Infectious Diseases at the University of Nottingham said, “remains unclear”. However, what was clear was that, as the WHO had pointed out, “symptomatic people are responsible for most of the new infections of COVID-19.”
With respect to the Times’ mischaracterization of both the WHO and the scientific evidence, Dr. Van Kerkhove’s sin was evidently not that her statement was contradicted by evidence that the WHO had stubbornly refused to acknowledge. Rather, the WHO official’s error was having made a truthful statement that did not conform to the propaganda narrative preferred by advocates of extreme lockdown measures.
This, too, was indicated in one of the reactions published by the Science Media Centre. As observed by Liam Smeeth, a Professor of Clinical Epidemiology at the London School of Hygiene and Tropical Medicine, the belief that “asymptomatic (people who never get symptoms) and pre-symptomatic people are an important source of infection to others” was “the main basis for steps such as self-isolation and lockdown”.
That is, Dr. Van Kerkhove’s statement contradicted the assumption underlying lockdown measures that not only presymptomatic but also asymptomatic transmission is common.
As discussed in part two, in her March 31 article, Mandavilli had similarly characterized the distinction as “unimportant” and a “semantic debate”. Yet, as the WHO points out on its website, it’s important for developing public health strategies to control transmission. For example, since data indicated that people “might be the most infectious at or around the time they develop symptoms”, the WHO recommends considering people to be “contacts” if they had contact with an infected person from two days before the infected person developed symptoms.
The distinction is also important because, as the WHO notes in a scientific brief published on July 9, studies also suggest “that individuals without symptoms are less likely to transmit the virus than those who develop symptoms.” (Emphasis added.)
This has practical relevance for the general public as well as policymakers and epidemiologists. For example, children are at low risk of severe COVID-19, the disease caused by SARS-CoV-2, so it makes no sense to keep schools closed due to the risk to children, which is recognized to be less than that from seasonal influenza.
Another rationale for keeping schools closed is that children might transmit the virus to older adults with underlying medical conditions that have been identified as risk factors for severe COVID-19. However, this reasoning, too, is questionable, given the “profound” harms caused by school closures when weighed against the data indicating that children are more likely to be asymptomatic and therefore less likely to spread the virus to others.
In short, the distinction is by no means insignificant or semantic but is an important one with real practical relevance for scientists, policymakers, and members of the public alike.
The Times’ Cited Estimate of Presymptomatic Transmission Had Important Limitations
While Mandavilli didn’t mention it, as noted, the CDC’s planning scenarios document had estimated the proportion of transmission from presymptomatic individuals at 40 percent but did not provide any information about how that number was estimated.
As noted in part two, when Mandavilli had relayed the estimate that 20 percent to 40 percent of transmission occurs before people develop symptoms in her March 31 article, she cited the lead researcher of a team out of Hong Kong, Dr. Benjamin Cowling, but without producing any such study or otherwise providing any information about how it was estimated.
Subsequently, on April 15, a study by Dr. Cowling and his team was published in Nature Medicine. Titled “Temporal dynamics in viral shedding and transmissibility of COVID-19”, this was the study Mandavilli was referring to in her June 9 article as a “widely cited paper” that suggested “people are most infectious up to two days before the onset of symptoms” and estimated the proportion of presymptomatic transmission at 44 percent.
What the Times still declined to inform readers was how that estimate was arrived at. Importantly, the estimate did not reflect documented instances of presymptomatic transmission but was a modeling study that made assumptions about the proportion based on other data.
The figure of 44 percent was calculated from the estimated “incubation period”, which is the time from infection until the development of symptoms, and the estimated “serial interval”, which is the time from symptom onset in a person who transmits the virus until symptom onset in the person to whom the virus was transmitted.
“If the observed mean serial interval is shorter than the observed mean incubation period,” the study authors noted, “this indicates that a significant portion of transmission may have occurred before infected persons have developed symptoms.”
For the mean incubation period, they cited a prior study estimating it at 5.2 days.
For the mean serial interval, they obtained information on 77 transmission pairs from a wide variety of sources, including reports of public health authorities in China and Japan, news articles, and Wikipedia, to arrive at an estimate of 5.8 days.
Additionally, they examined data on collected throat swabs from 94 patients for up to 32 days after symptom onset and tested for viral RNA in patients using reverse transcription polymerase chain reaction (RT-PCR) assays, the results of which indicated “high viral loads after symptom onset, which then gradually decreased towards the detection limit at about day 21.” (Emphasis added.)
Citing Dr. Cowling and his team, Mandavilli had written in her March 31 article that infected individuals “are most contagious about one to three days before they begin to show symptoms”; and in her June 9 piece, citing their Nature Medicine study, she wrote that “people are most infectious up to two days before the onset of symptoms”.
What she did not tell Times readers was that the researchers “did not have data on viral shedding before symptom onset”.
Instead, based on viral load data and the estimated mean incubation period and serial interval, they inferred that patients could transmit the virus starting from “2.3 days before symptom onset” and peaking at “0.7 days before symptom onset”.
They also assumed that viral RNA detected by RT-PCR assay was equal to “infectiousness”.
Once again, Times readers were consequently also not informed that the detection of viral RNA by RT-PCR assay is not necessarily indicative of the presence of infectious virus. As noted in another study published in the same journal, Nature Medicine, on June 18, “measurable virus shedding does not equate with viral infectivity, and further evaluation is needed to determine the respiratory SARS-CoV-2 viral load that is correlated with culturable virus.”
RT-PCR assays are a proxy measure of viral load. They work by cyclically amplifying any viral RNA present in a sample, with the “threshold cycle” value being inversely proportional to the amount of RNA present. If fewer replication cycles are required to reach the threshold of a positive result, it indicates a higher amount of viral RNA, whereas if more cycles are required, it indicates a lower “viral load”.
Counterintuitively, the study authors observed no association between viral load and disease severity. However, an equally high viral load in a person with mild or no symptoms does not necessarily mean that they are as contagious as a person with more severe disease.
For example, although studies have found similar viral loads between asymptomatic and symptomatic individuals, they also show that asymptomatic cases are less likely to transmit the virus. As the authors of a systematic review and meta-analysis published on the preprint server medRxiv on June 4 concluded, “asymptomatic spread is unlikely to be a major driver of clusters or community transmission of infection”, while the extent of presymptomatic or paucisymptomatic (mildly symptomatic) transmission “remains unknown”.
One likely reason for this is that a lack of symptoms means that there is no opportunity for droplet transmission to occur through coughs or sneezes. As Mandavilli acknowledged in her March 31 piece, even if it can be spread through aerosols from speaking or breathing, the virus “is still most likely to be expelled with a cough or a sneeze”.
Another possible reason is that an effective immune response could result in RT-PCR assays detecting nonviable virus fragments that pose no risk of infection to others if expelled by the patient, whether by droplets or aerosols.
In a study published on July 27 in Clinical & Translational Immunology, for example, researchers examined shedding patterns in patients and concluded, “Less active T-cell responses at the initial phase of infection were associated with prolonged viral RNA shedding, suggesting that early immune responses are beneficial to control viral load and prevent viral RNA shedding.”
They suggested that “more effective T-cell activation at the early phase of infection promotes virus clearance and subsequently shortens viral RNA shedding duration in COVID-19.”
Furthermore, “detection by RT-PCR is only a surrogate marker for infectivity. Positive RT-PCR test may represent non-viable viruses or remnant nucleic acid products.” (Emphasis added.)
“Little is known about the infectiveness of asymptomatic patients”, noted the authors of a study published in JAMA Internal Medicine on August 6. As had prior studies, they found that viral loads “in asymptomatic patients were similar to those in symptomatic patients”, which offered “biological plausibility” to reports of asymptomatic transmission.
However, they also observed that, counterintuitively, threshold cycle values for envelope genes of the virus indicated that viral loads of asymptomatic patients tended to decrease more slowly than those of presymptomatic or symptomatic patients. Their explanation for this was that “The genetic material of dead viral particles remaining within epithelial cells can be detected as epithelial cells are desquamated.” In this context, they emphasized that “detection of viral RNA does not equate [to] infectious virus being present and transmissible.”
Returning to the Nature Medicine study cited by the New York Times, since the estimated incubation period was less than the mean serial interval, it indicated that most transmission would occur after the onset of symptoms, while “viral shedding may begin 2 to 3 days before the appearance of the first symptoms”, with viral loads peaking on the day of symptom onset.
However, the researchers noted that most patients for which they obtained data on transmission pairs “were isolated after symptom onset”, which limited the potential for post-symptomatic transmission. “Places with active case finding”, they explained, “would tend to have a higher proportion of presymptomatic transmission, mainly due to quick quarantine of close contacts and isolation, thus reducing the probability of secondary spread later on in the course of illness.”
In other words, the transmission pairs for which they obtained data were more likely to have contributed to a shorter mean serial interval than if the cases had gone unidentified or otherwise had not been isolated. This ascertainment bias limits the generalizability of their estimate to populations where rigorous contact tracing isn’t done, including populations like that in the US where such rigorous tracing isn’t practically feasible.
This limitation was also observed in an analysis published on medRxiv on June 11. “Populations where symptomatic people are rapidly isolated”, its authors stated, “are likely [to] have relatively more pre-symptomatic transmission.”
Another limitation of the Nature Medicine study was that it relied on patients’ recollections of when their symptoms began, which introduced “recall bias” that “would probably have tended toward the direction of under-ascertainment, that is, delay in recognizing first symptoms.” This would result in overestimation of the incubation period, which would in turn mean that the proportion of presymptomatic transmission estimated was “artifactually inflated.”
Furthermore, they state that their estimate of the proportion of presymptomatic infection is relevant to “settings with substantial household clustering”.
In China, over 70 percent of SARS-CoV-2 transmission is estimated to have occurred in families, and where people with infection are required to isolate themselves in their homes, household contacts will be at risk of infection. Where compulsory home quarantine or “stay-at-home” lockdown measures are in place, it is unsurprising that high proportion of transmission occurs in households.
According to the corresponding author of the Nature Medicine study, Eric Lau, transmission “within a household with frequent and more intensive contact, especially during a lockdown results in shorter serial intervals.”
Lockdown measures are thus a confounding factor in estimating the proportion of presymptomatic transmission. Under lockdown, a higher proportion of transmission is likely to occur within households. The estimated mean serial interval is therefore likely to be shorter. The estimated proportion of presymptomatic transmission is therefore likely to be higher.
Consequently, to take the estimated 44 percent of presymptomatic transmission and generalize it to the community setting outside the household especially in the absence of lockdown measures is a fallacy of composition.
It is ironic that this estimated proportion of presymptomatic transmission has been so widely cited to justify lockdown measures when such measures have likely biased the data toward this very result, which could be to a great extent a statistical artifact from this and other biases inherent in the data.
Another limitation with respect to the estimated mean incubation period and serial interval is that the confidence intervals of each of these estimates overlap. For the mean incubation period of 5.2 days, turning to their source study, we can see that researchers determined that there was a 95 percent probability that the mean incubation period was between 4.1 and 7.0 days. For their own estimated mean serial interval of 5.8 days, the confidence interval was between 4.8 to 6.8 days.
Thus, the confidence interval for the serial interval fell entirely within that for the incubation period.
The problem of overlapping confidence intervals in this and other such modeling studies was discussed in a systematic review of studies on asymptomatic and presymptomatic transmission of SARS-CoV-2, which was published at medRxiv on June 17.
The first problem was that the serial interval was calculated “by compiling data from publicly available sources or from municipal datasets”, making it “difficult to control for quality and bias” as well as introducing variability in “standards of reporting cases or symptom onset.” Additionally, “all of the serial interval studies are confounded by their reliance on self-reported symptom start date”, which introduced the acknowledged recall bias.
“Furthermore,” the review authors noted, “in the datasets, the authors report the date of symptom onset rounded to the nearest day. This is especially problematic because the difference in serial interval and incubation period calculated in these studies often differed by less than a day. It is therefore not possible to ascertain if the difference between calculated serial interval and incubation period are true differences, or an artefact of rounding error.” (Emphasis added.)
The modelers had interpreted the data as supportive of the hypothesis that infectiousness began two to three days before the appearance of symptoms. However, since the calculated serial interval fell within the 95 percent confidence interval of the estimated incubation period, the data was also “compatible with the hypothesis that infectiousness appears to emerge at symptom onset.” (Emphasis added.)
The review authors also pointed out that “a positive RT-PCR test does not confirm that an individual is contagious.” While one study had shown that “success of virus isolation in culture was a function of viral load”, meaning that there was a correlation between higher threshold cycle values and the ability to isolate viable virus, researchers also observed that “no isolates were obtained after day 8, despite continuing high viral loads. This finding suggests persistent RNA detection represents non-viable virus that is not infectious. This finding demonstrates that while viral load can be predictive of transmissibility, it is not a perfect correlation.”
Findings that viral loads were highest around the time observations began were problematic since no data was collected on viral loads prior to symptom onset.
This problem was compounded by the fact that “patients often will not see a clinician immediately after symptom onset”. One study of patients in China had found that “an average of 2.5 days elapsed between symptom onset and first healthcare consultation.” Therefore, the possibility could not be ruled out “that viral load peaks after symptom onset.”
Additionally, there was a lack of specificity in studies about “how soon the first swab was taken after symptoms were reported; a margin error of a day might dramatically change the viral load in patients.”
While viral load data indicated the plausibility of presymptomatic transmission, “there is not enough information about the distribution of SARS-CoV-2 viral kinetics in presymptomatic stage to conclude when infectiousness begins.”
Indeed, while the authors of the Nature Medicine study stated that they utilized data on throat swabs collected from patients “from symptom onset”, this does not appear to be accurate.
All 94 of the patients had already been admitted to a hospital in Guangzhou, China, and 66 percent “were moderately ill” at the time of their admission. In Supplementary Table 1 of their study, they show further that only two patients had no symptoms at the time of admission, while the remaining 32 percent had mild symptoms. Furthermore, no information is presented on the timing after admission that the first samples were collected from patients. Therefore, it appears highly likely that the first samples were only taken a considerable time after symptom onset.
The authors of the systematic review also commented on the bias introduced by using data from household clusters. As they explained, “In household transmission cases, newly infected individuals will likely be exposed to a much higher dose of viral particulates than would occur in a more casual transmission case. Exposure to higher inoculum may result in a decreased incubation period for household transmission.”
Therefore, “the difference in inoculum between household transmission and community transmission may account for the difference between the calculated serial interval and incubation period. . . . It is not possible to differentiate an observation of shortened serial interval due to presymptomatic transmission from a decreased incubation period due to higher inoculum in household transmission.” (Emphasis added.)
In conclusion, the available evidence from modeling studies as well as case reports and viral dynamics studies “have been inadequate to ascertain the contribution of asymptomatic and presymptomatic transmission in the spread of SARS-CoV-2 infection.”
To summarize, the Times presented the estimated 44 percent presymptomatic transmission without informing readers that this number did not represent documented cases of such transmission but was rather a modeling study inferring this proportion from various datasets. Consequently, Times readers also remained uninformed about the serious limitations of the study’s methodology.
Times readers were not informed that the possibility that presymptomatic patients might be able to transmit the virus was inferred from data using RT-PCR tests done after symptom onset as a proxy measure for viral load on the fallacious assumption that detected viral RNA equals the presence of infectious virus and represents the contagiousness of the individual.
Times readers were also not informed that the proportion of 44 percent was inferred from an estimated mean incubation period and serial interval, each derived from different datasets that together likely biased the estimate toward artifactually finding a higher proportion of presymptomatic transmission than is likely to occur in the community setting in the US—especially in an absence of lockdown measures.
The WHO Had Rightly Noted the Limited Evidence on Mask Effectiveness
To support the characterization of the WHO as having made statements and policy recommendations about viral transmission that were contradicted by the science, Mandavilli’s June 9 article contained deceptions additional to those about asymptomatic and presymptomatic spread.
Her claim that the WHO did not endorse mask use by members of the general public until June 5 is false. This was a reiteration of the claim Mandavilli had made in her June 5 article, discussed in part three of this series, that the WHO had “opposed” the use of masks by members of the public until that date.
In fact, in its prior guidance on masks issued on January 29, the WHO had recommended medical masks to be worn by any members of the public who had symptoms.
The claim that the WHO had claimed that there was “too little evidence that they prevented transmission of the virus” is consequently also false.
Rather, the WHO’s recommendation for mask use was premised on the available evidence that masks are an effective means of “source control”, meaning that they help reduce droplet transmission from coughs and sneezes.
What the WHO’s earlier guidance had accurately said that there was “no evidence” to support was the idea that masks would protect healthy wearers in the community setting from becoming infected with SARS-CoV-2.
The WHO’s initial guidance document also suggested that the use of medical masks “when not indicated” could cause increased costs; create supply shortages for health care workers who truly needed them; and create “a false sense of security” among mask-wearing members of the general public, especially if improperly used.
Subsequently, evidence emerged indicating the possibility that infected individuals could transmit the virus before developing symptoms, and the WHO therefore updated its guidance on June 5. However, the suggestion that the WHO endorsed universal mask-wearing recommendations—if not mandates—is also false.
On the contrary, the WHO accurately observed in its updated guidance that “the widespread use of masks by healthy people in the community setting is not yet supported by high quality or direct evidence and there are potential benefits and harms to consider”, including the potential for wearers to increase their own or others’ risk of infection through improper use of the mask. (Emphasis added.)
Consequently, the WHO did not endorse universal mask use, but rather recommended that members of the public wear a mask in specific circumstances. Specifically, people were advised to wear a mask if there was evidence of community transmission in settings where prolonged close contact with others would be unavoidable.
The WHO Had Rightly Noted Airborne Transmission Remained Theoretical
As detailed in part one, in her March 17 article “How Long Will Coronavirus Live on Surfaces or in the Air Around You?”, Mandavilli falsely characterized studies as having proven that SARS-CoV-2 is airborne transmissible, meaning that it spreads via aerosolized particles as distinguished from larger respiratory droplets.
The New York Times was not alone in misrepresenting the science. Other media sources did the same, and the misinformation was spread on social media. On Facebook, the message was spread: “BREAKING NEWS! Covid-19 is confirmed as airborne and remains 8 hrs in air! So everyone is required to wear mask everywhere!!”
Consequently, the WHO responded on Facebook with a “Fact Check” post on March 28 rightly pointing out that that claim was “INCORRECT”.
However, the WHO post also misinformed Facebook users by expressing certainty that “COVID-19 is NOT airborne”.
Mandavilli did not cite that Facebook post or any other statements from the WHO expressing certainty that SARS-CoV-2 is not airborne. However, that WHO Facebook post illustrates that her June 9 article did contain a legitimate criticism of the WHO; namely, the WHO should not have expressed such certainty to the public.
However, in its scientific reports, the WHO did not draw this negative conclusion but rather emphasized the remaining uncertainty. As noted in part one, in a scientific brief published on March 29, the WHO rightly criticized news outlets for having reported airborne transmission as a proven fact and explained the limitations of the available evidence.
The WHO did not conclude that the lack of evidence was evidence of the absence of aerosol transmission but rather noted that further studies were needed to determine whether viable virus could be found in air samples and, if so, “what role it may play in transmission.”
The brief concluded with the statement, “WHO carefully monitors emerging evidence about this critical topic and will update this scientific brief as more information becomes available.”
So, while the criticism might be legitimate in the context of the WHO’s Facebook post, it is not legitimate in the context of the WHO’s formal reports to the public.
What’s even more important to observe about Mandavilli’s June 9 article, though, is how she falsely characterized the WHO as having “walked back” its statements about asymptomatic transmission while herself walking back from her earlier report characterizing airborne transmission as scientifically proven.
Yet, rather than acknowledging her own error, she sought to discredit the WHO as having “too narrowly” defined airborne transmission, which “also includes the possibility that the virus is aloft for shorter distances, then inhaled.”
This point goes to the public perception of what it means for a virus to be “airborne”, which conjures up the thought of a highly contagious virus, like measles, that a person can be at risk of exposure to by simply being in a room that an infected person had previously occupied, or by being in the same location as the infected person even if at a considerable distance.
Since the evidence indicated that close contact would be required for SARS-CoV-2 transmission to occur, droplet transmission alone was a sufficient explanation. If it were evident that the virus could be transmitted in circumstances that could not be explained by droplet transmission, it would be evidence of transmission via aerosols.
As Mandavilli acknowledged, even if airborne, prolonged close contact would still be required for transmission to occur. Specifically, she included the caveat that the risk potentially existed when people congregated “in poorly ventilated spaces”.
The criticism from scientists she was relaying was that aerosols, too, could play a role in short-range transmission. Indeed, there is ambiguity about the distinction since droplets and aerosols exist on a continuum.
While this is an important point, it was nevertheless a strawman argument because the WHO had acknowledged the possibility that aerosols could also play a role in transmission and stated that further research was required to determine this.
Similarly, to say that “there is no evidence” to indicate uncertainty does not in fact convey “certainty about the absence of a phenomenon”. This is a nonsensical argument. What it ipso facto indicates is uncertainty.
What is most fascinating about that criticism, though, is how it represented a tacit acknowledgment from the Times that its prior reporting on airborne transmission was misinformation.
Indeed, it was Mandavilli and the New York Times, not the WHO, who had done the walking back.
 Jeremy R. Hammond, “How the New York Times Lies about SARS-CoV-2 Transmission: Part 1”, JeremyRHammond.com, July 29, 2020, https://www.jeremyrhammond.com/2020/07/29/how-the-new-york-times-lies-about-sars-cov-2-transmission-part-1/.
 Jeremy R. Hammond, “How the New York Times Lies about SARS-CoV-2 Transmission: Part 2”, JeremyRHammond.com, August 3, 2020, https://www.jeremyrhammond.com/2020/08/03/how-the-new-york-times-lies-about-sars-cov-2-transmission-part-2/.
 Jeremy R. Hammond, “How the New York Times Lies about SARS-CoV-2 Transmission: Part 3”, JeremyRHammond.com, August 7, 2020, https://www.jeremyrhammond.com/2020/08/07/how-the-new-york-times-lies-about-sars-cov-2-transmission-part-3/.
 Hammond, “Part 1”.
 Hammond, “Part 2”.
 Hammond, “Part 3”.
 Apoorva Mandavilli, “In the W.H.O.’s Coronavirus Stumbles, Some Scientists See a Pattern”, New York Times, June 9, 2020, https://www.nytimes.com/2020/06/09/health/coronavirus-asymptomatic-world-health-organization.html.
 Hammond, “Part 1”.
 Hammond, “Part 2”.
 World Health Organization, “COVID-19: Virtual Press conference”, WHO.int, June 8, 2020, accessed July 16, 2020, https://www.who.int/docs/default-source/coronaviruse/transcripts/who-audio-emergencies-coronavirus-press-conference-08jun2020.pdf.
 Hammond, “Part 2”.
 World Health Organization, “Report of the WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19)”, WHO.int, February 2020, accessed July 16, 2020, https://www.who.int/docs/default-source/coronaviruse/who-china-joint-mission-on-covid-19-final-report.pdf.
 Apoorva Mandavilli, “Infected but Feeling Fine: The Unwitting Coronavirus Spreaders”, New York Times, March 31, 2020, https://www.nytimes.com/2020/03/31/health/coronavirus-asymptomatic-transmission.html.
 Centers for Disease Control and Prevention, “COVID-19 Pandemic Planning Scenarios”, CDC.gov, May 20, 2020, archived June 9, 2020, accessed August 16, 2020, http://web.archive.org/web/20200609002054/https://www.cdc.gov/coronavirus/2019-ncov/hcp/planning-scenarios.html.
 Science Media Centre, “expert reaction to comments from Dr Maria van Kerkhove at yesterday’s WHO press conference that asymptomatic transmission of SARS-CoV-2 is rare”, ScienceMediaCentre.org, June 9, 2020, https://www.sciencemediacentre.org/expert-reaction-to-comments-from-dr-maria-van-kerkhove-at-yesterdays-who-press-conference-that-asymptomatic-transmission-of-sars-cov-2-is-rare/.
 World Health Organization, “Q&A: How is COVID-19 transmitted?”, WHO.int, July 9, 2020, accessed July 16, 2020, https://www.who.int/news-room/q-a-detail/q-a-how-is-covid-19-transmitted. See also: Hammond, “Part 2”.
 World Health Organization, “Transmission of SARS-CoV-2: implications for infection prevention precautions”, WHO.int, July 9, 2020, accessed July 16, 2020, https://www.who.int/publications/i/item/modes-of-transmission-of-virus-causing-covid-19-implications-for-ipc-precaution-recommendations. See also: Hammond, “Part 2”.
 Lara S. Shekerdemian et al., “Characteristics and Outcomes of Children With Coronavirus Disease 2019 (COVID-19) Infection Admitted to US and Canadian Pediatric Intensive Care Unites”, JAMA, May 11, 2020, https://doi.org/10.1001/jamapediatrics.2020.1948. See also: Hammond, “Part 2”.
 For further discussion, see: Hammond, “Part 2”.
 Hammond, “Part 2”.
 Hammond, “Part 2”.
 Xi He et al.
 Xi He et al.
 Oyungerel Byambasuren et al., “Estimating the extent of asymptomatic COVID-19 and its potential for community transmission: systematic review and meta-analysis”, medRxiv, June 4, 2020, https://doi.org/10.1101/2020.05.10.20097543.
 Hammond, “Part 2”.
 Pei Hua Lee et al., “Associations of viral ribonucleic acid (RNA) shedding patterns with clinical illness and immune responses in Severe Acute Respiratory Syndrome Coronavirus 2 (SARS‐CoV‐2) infection”, Clinical & Translational Immunology, July 27, 2020, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7385430/.
 Seungjae Lee, “Clinical Course and Molecular Viral Shedding Among Asymptomatic and Symptomatic Patients With SARS-CoV-2 Infection in a Community Treatment Center in the Republic of Korea”, JAMA Internal Medicine, August 6, 2020, https://doi.org/10.1001/jamainternmed.2020.3862.
 Xi He et al.
 Yu Wang et al., “Reduction of secondary transmission of SARS-CoV-2 in households by face mask use, disinfection and social distancing: a cohort study in Beijing, China”, BMJ Global Health, May 28, 2020, http://dx.doi.org/10.1136/bmjgh-2020-002794.
 Eric Lau, email correspondence, June 22, 2020.
 Xi He et al.
 Xi He et al.
 Savvides and Siegel.
 Hammond, “Part 3”.
 World Health Organization, “Advice on the use of masks in the community, during home care and in health care settings in the context of the novel coronavirus (2019-nCoV) outbreak”, WHO.int, January 29, 2020, https://apps.who.int/iris/handle/10665/330987. See also: Hammond, “Part 3”.
 World Health Organization, “Advice on the use of masks in the context of COVID-19”, WHO.int, June 5, 2020, https://www.who.int/publications/i/item/advice-on-the-use-of-masks-in-the-community-during-home-care-and-in-healthcare-settings-in-the-context-of-the-novel-coronavirus-(2019-ncov)-outbreak. See also: Hammond, “Part 3”.
 Hammond, “Part 1”.
 A Google search turned up the following examples of this claim being spread on Facebook: JK News Network, “All are requested to take care of yourself, even wear masks at home also”, Facebook, March 29, 2020, https://www.facebook.com/JKNN1/posts/all-are-requested-to-take-care-of-yourself-even-wear-masks-at-home-also-breaking/3127152587318055/. Kashmir News, “All are requested to take care of yourself, even wear masks at home also”, Facebook, March 29, 2020, https://www.facebook.com/permalink.php?id=177736282407348&story_fbid=1422617374585893.
 World Health Organization, “FACT: COVID-19 is NOT airborne”, Facebook, March 28, 2020, https://www.facebook.com/WHO/posts/3019704278074935. Note that the WHO had posted this prior to the examples provided in the prior endnote so the posts I cited had not originated the claim, but are merely examples in which the claim was being spread on Facebook.
 World Health Organization, “Modes of transmission of virus causing COVID-19: implications for IPC precaution recommendations”, WHO.int, March 29, 2020, https://apps.who.int/iris/handle/10665/331616. See also: Hammond, “Part 1”.