by Jon Sanders
Research Editor and Senior Fellow, Regulatory Studies, John Locke Foundation
Click here for Part 1 of this series.
To recap: The Cooper administration sent a list of 22 studies, all published this year, containing what they state is “overwhelming” “scientific evidence for the protective effect of face masks and respiratory virus infection in healthcare and community settings.”
Remember: Gov. Roy Cooper issued orders against the normal scope of the Emergency Management Act, and Cooper has forced face masks on North Carolinians, even in their own homes at times, at the threat of misdemeanor charges, civil fines, business closures, people being turned away at business doors, people being threatened with trespassing, people being urged by the governor to call the cops on their fellow citizens, and perhaps worse.
So what this document, which is not easy to find through searching the DHHS web site independently, should contain is no less than this: a slam-dunk case justifying Cooper’s extreme, open-ended emergency orders.
This series will examine and discuss each study provided here by the Cooper administration, in the order it is presented by the Cooper administration, and ask whether it supports Cooper’s extreme exercise of power, because that is the relevant standard.
Here is the second batch:
Identifying airborne transmission as the dominant route for the spread of COVID-19. Proceedings of the National Academy of Sciences (2020), 117 (26) 14857-4863.
Published June 11, 2020, this study is very important to the Cooper administration. It was one of the studies Cooper listed among the five “examples of studies studying the effectiveness of face coverings” that he included in his Nov. 23 executive order tightening his mask order against people. It was also among the three studies presented before the General Assembly by state health bureaucrat Mandy Cohen on June 17, a week before Cooper first leveled his mask order on people (discussed here).
What the study purported to find was that mandated face masks were “the determinant in shaping the pandemic trends worldwide,” arguing that “wearing of face masks in public corresponds to the most effective means to prevent interhuman transmission.”
I explained how Zhang et al. arrived at that conclusion:
Italy didn’t mandate face masks until April 6 (northern Italy, then the rest of Italy on May 4), and NYC didn’t mandate face masks until April 17. Zhang et al. projected a linear (“no mask”) assumption in case increases in Italy and NYC, and then compared their projection with actual cases. The differences between actual cases and their linear (“no mask”) assumption are where they derived their finding quoted by Cohen: “Our analysis indicates that face coverings reduced the number of infections by over 78,000 in Italy from April 6 to May 9 and by over 66,000 in NYC from April 17 to May 9.” …
A major problem was, as I pointed out, both Italy and NYC were well past their case peaks when masks were mandated. Cases were falling already as a consequence of being on the other side of the infections spike:
Italy and NYC’s impositions of their mask mandates well after their peak of daily new cases makes it difficult to ascertain how much actual effect their mask mandates had on cases. Expecting a linear progression of cases post-peak also seems inflationary to the number of cases (under a no-mask assumption) over time.
Of note here: The original link for this study provided by the Cooper administration is broken. Searching for a proper link to the study yielded a scientific controversy — including a letter signed by 45 epidemiologists signed on to a letter making a formal request for retraction of the paper. They cited numerous methodological errors and “verifiably false” statements.
This study is highly problematic. A study beset with so many methodological errors and “verifiably false” statements that 45 epidemiologists formally requested its retraction absolutely must not serve as a basis for an extreme emergency order.
Received May 1, 2020, this study is a retrospective cohort study of household transmission of COVID-19 by people with confirmed infections (“primary cases”). It relied on telephone interviews asking family members to self-reportmask-wearing, hygienic behaviors, physical distancing practices, close contact, living arrangements, room ventilation and cleaning, disinfectant use, separate eating utensils, etc. The primary cases had to have had fevers and a respiratory symptom, evidence of pneumonia, a normal to reduced white blood cell count, and a decreased lymphocyte count. They also must have either visited or lived in Wuhan, China, in the 14 days prior to symptoms, been in contact with someone who had, or been in a known cluster case.
Wang et al. conclude that:
This study reinforces the high risk of transmission in households but importantly shows that UFMU [universal face mask use] and hygiene measures can significantly reduce the risk of household transmission of COVID-19, independent of household size or crowding. This is the first study to show the effectiveness of precautionary mask use, social distancing and regular disinfection in the household, and can inform guidelines for prevention of household transmission.
Wang et al. found that in households with a proven case, “Face masks were 79% effective and disinfection was 77% effective, while close frequent contact in the household increased the risk of transmission 18 times, and diarrhoea in the index patient increased the risk by four times.” According to their introduction, “Community mask wearing, hand washing and social distancing are thought to be effective” — i.e., those communitywide effects were not being tested here — “but there is little evidence to inform or support community members on COVID-19 risk reduction within families.” They said their new findings “imply” support for masks and social distancing “inside the household with members at risk of getting infected.”
Nevertheless, Wang et al. realized that since “compliance of [universal mask use] would be poor in the home, there was difficulty and also no necessity for everyone to wear masks at home.” They narrowed their recommendations to those families with members “at risk of getting infected.”
This study focuses on mask wearing, social distancing, hand washing, and disinfection in the household setting. It cannot be used for Cooper’s extreme emergency orders affecting people statewide.
Also, this study relies on telephone interviews asking people to remember things after the fact. It will be affected by all kinds of biases and confounders that researchers cannot help, including recall bias, observer bias, information bias, etc. It’s not a randomized controlled trial (RCT), which is the gold standard. Studies like these are unlikely to provide evidence compelling enough for extreme emergency orders.
Even where Cooper’s mask order affects people inside their own homes, it is not supported by this study. Wang et al. recommend — not call for government orders for — mask-wearing at home only for “those families with members who were at risk of getting infected,” such as “ever having contact with a COVID-19 patient, medical workers caring for a COVID-19 patient or having a history of travelling to high risk areas).” Cooper’s order assumes everyone is a risk to others by virtue of simply breathing. Such an assumption is insufficient for extreme emergency orders.
Published in November 2020, this study is a retrospective case-control study following “contact investigations of 3 large COVID-19 clusters in nightclubs, boxing stadiums, and a state enterprise office in Thailand.” It relied on telephone interviews asking contacts to the best of their memories about mask-wearing and frequency during contact with the index patient, as well as hand-washing, social distancing, close contact, sharing the same dishes and cups, etc., and whether the index patient was symptomatic. (They acknowledge “several limitations” of their study, including the “common biases of retrospective case-control studies, including memory bias, observer bias, and information bias.”)
Doung-ngern et al. found lower odds of infection from “maintaining a distance of ≥1 m[eter] from a COVID-19 patient” and similar odds of infection from “duration of contact ≤ 15 minutes” than for “wearing a mask all the time during contact with a COVID-19 patient.” They found wearing a mask “sometimes during contact with the COVID-19 patient was not statistically significantly associated with lower risk of infection.” They also tested the odds of infection concerning type of masks worn, finding that medical masks were associated with lower odds of infection when worn during contact with a COVID-19 patient. They calculated slightly lower odds of infection from wearing nonmedical masks, but the upper range of their odds included a higher risk of infection.
Doung-ngern et al. stress “consistent” rather than intermittent mask use around COVID-19 patients and, as has other studies promoted by the Cooper administration, warned about incorrect mask wearing.
Setting is key for infection, Doung-ngern et al. found, especially in the home (see Wang et al. above): “The household secondary attack rate in our study (16.5%) is comparable with ranges reported previously (11%–23%) and relatively high compared with workplaces (4.9%) and other settings (1.4%).” Perhaps for that reason, Doung-ngern et al. find that “quarantine measures can be challenging and sometimes impractical.”
They conclude with could and might,
our findings provide evidence supporting consistent mask-wearing, handwashing, and adhering to social distancing recommendations to reduce SARS-CoV-2 transmission in public gatherings. Wearing nonmedical masks in public could help slow the spread of COVID-19. Complying with all measures could be highly effective; however, in places with a high population density, additional measures might be required.
They also say that “Clear and consistent public messaging” is “essential,” and “particularly for targeting those who wear masks intermittently or incorrectly.”
Probable effects (“could” help slow the spread, “could” be highly effective, etc.) are insufficient for extreme emergency orders. Observational studies as opposed to randomized controlled trials (RCTs) are also unlikely to provide evidence compelling enough for extreme emergency orders because of the many bias problems Doung-ngern et al. cite. They found small effects from concerning the nonmedical masks (i.e., the cloth to homemade “face coverings” that Cooper’s order mandates) with a range that suggested even a greater risk of infection in some instances. That evidence is certainly not compelling enough for extreme emergency orders.
Here again Doung-ngern et al. find the household is the key setting for infections, not workplaces and definitely not other settings. This finding would, if anything, caution against Cooper’s latest “Modified Stay-At-Home” order.
Furthermore, this study specifically cautions against some aspects of mask-wearing inherent in the extents of Cooper’s orders. They cause, for example, restaurant patrons, schoolchildren, people trying to speak clearly, and even governors and health secretaries taking turns at the same microphone for ≥15 minutes to resort to regularly donning and doffing masks. This study’s focus on intermittent use seems to miss the cautionary note about long-term use found in other research put forth by the Cooper administration. These aspects are contrary to Cooper’s extreme emergency order.
Factors Influencing Risk for COVID-19 Exposure Among Young Adults Aged 18–23 Years — Winnebago County, Wisconsin, March–July 2020. MMWR Morb Mortal Wkly Rep (2020); 69:1497–1502.
Posted October 9, 2020, this study relied on telephone interviews with 13 young adults, nine business owners, and eight community leaders in Winnebago County, Wisconsin, regarding COVID-19 infections among young adults (ages 18–23), since by July 18 they accounted for 32 percent of known infections there despite comprising only 12.5 percent of the population.
Wilson et al.’s interest was to “identify drivers of behavior that influence risk for exposure to COVID-19 among young adults.” Each survey participant was given “a gift card of $25” since Wilson et al. determined that amount would “balance motivating interviewees to participate without offering a coercive sum.” They tailored their questions to the different groups and stopped at 30 interviews because they had achieved “Thematic saturation” beyond which they determined they weren’t getting any new information.
The “drivers of behavior” they identified were “perceived low severity of disease outcome; perceived responsibility to others; peer pressure; and exposure to misinformation, conflicting messages, or opposing views regarding masks.” Wilson et al. start with the presumption that “Masks are an effective tool to prevent the spread of COVID-19,” so their focus is on “framing messaging to target those factors.”
Wilson et al. find that their report provides,
a framework for tailoring communication messages that are empathetic, that amplify personal responsibility and responsibility to protect others, and that focus on perceived pressure to not wear a mask, all of which might persuade young adults to adhere to public health guidelines (e.g., wearing masks) to prevent the spread of COVID-19.
Regardless of whether the themes of 30 phone conversations in Winnebago County, Wisconsin, are forcibly applicable to the entire state of North Carolina, nothing in this study pertains to extreme emergency orders. It is about “tailoring communication messages” to persuade people (young adults) to adopt wearing masks voluntarily. Tailoring messages is not the business of extreme emergency orders.
Cooper still has — and has always had — the option to treat North Carolinians as free-thinking adults and use his office and state health department “to recommend, to persuade, even to urge people to adopt practices such as wearing face coverings that they believe is healthy,” but he has chosen instead to issue extreme emergency orders.
Posted August 5, 2020, this study looked at “per-capita mortality” in “countries where mask use was either an accepted cultural norm or favored by government policies on a national level” as compared with “countries which did not advocate masks.” They looked at the early outbreak in 200 countries around the world, from the beginning of the outbreak through April 16 for infections and May 9 for deaths.
Leffler et al. found several factors associated with higher national per-capita mortality from COVID-19: “older age of the population, urbanization, obesity, and longer duration of the outbreak.” Other than international travel restrictions, other containment measures were not associated with lower per-capita mortality, including notably “internal lockdown and viral testing policies and levels.” They found that “societal norms and government policies supporting mask-wearing by the public were independently associated with lower per-capita mortality from COVID-19.”
They write that “Available scholarship and surveys do corroborate reports in the news media that mask wear was common in public in many Asian countries,” as well as “parts of the Middle East, Africa, Latin America and the Caribbean.” But they say, “Conversely, in Western countries which had no tradition of mask-wearing, and which only recommended (rather than mandated) mask-wearing by the public, such as the United States, the practice has been steadily increasing, but change has not been immediate.”
Nevertheless, the authors scored the United States as a “mask-wearing” country prior to most statewide mask mandates. The U.S. was one of 41 countries for which Leffler et al. had surveys of mask-wearing. They write that,
To determine the influence that actual mask-wear, as opposed to mask policies, might have on the model, we scored countries as mask-wearing if at least 50% of the public wore a mask, and non-mask wearing if less than 50% of the population did so.
Based on surveys, Canada, Finland, France, Germany, and Malawi were not considered mask-wearing countries at any time during the exposure period (ending April 16). In contrast, Italy was scored as mask-wearing beginning March 19, Spain and India beginning March 21, Saudi Arabia beginning April 1, Russia beginning April 4, Singapore beginning April 10, and the United States, Brazil and Mexico beginning April 12. (End notes omitted.)
Surveys cited by Leffler et al. found that mask-wearing had already reached 66–68 percent in the U.S. by late May into early June.
This study is very far-ranging and looks at 200 countries with different “societal norms” and government policies regarding mask-wearing. It is nevertheless an impressively extensive exercise in suggesting correlation is causation. Leffler et al. do not offer strong enough evidence for government coercion as opposed to recommendations (and public messaging) for their study to be used to justify an extreme emergency order.
This study, like some others used by the Cooper administration, prefers Asian mask-wearing “tradition” over the culture in the West and the U.S. Changing the culture is still no business of an extreme emergency order.
The U.S. in general was scored by the authors to be “mask-wearing” as early as April 12. North Carolina was well past 60 days into the outbreak before Cooper’s mask order. So the findings of the study seem moot here. A moot finding does not back an extreme emergency order.
Meanwhile, the subsequent “second wave” of the virus as fall arrived has upended early mask research, most done as spring gave way to summer, and this one looked at infections through April 16, when North Carolina was in the grips of Cooper’s initial lockdown order.
Furthermore, as Leffler et al. point out, a serious limitation in their study is that “the ultimate source of mortality data is often from governments which may not have the resources to provide a full accounting of their public health crises, or an interest in doing so.” Leffler et al. seem to regard the only way governments would err in death reporting would be to undercount, but it is also possible for governments to err in the other direction — indeed, the Cooper administration conflates death data so that researchers cannot disaggregate who died owing to COVID-19 or from fatal complications arising from it and those who died while coincidentally also having been diagnosed with COVID-19 (the difference between those who died from and who died with the virus).
The Cooper administration has, however, shown a lack of interest in reporting deaths other than COVID-19 deaths, causing North Carolina to lag the rest of the nation by over two months in reporting deaths data to the Centers for Disease Control and Prevention. There is a growing alarm among health experts, economists, policymakers, and others worldwide about the negative health impacts from government restrictions and lockdowns in response to COVID-19. If this is true in North Carolina and being exacerbated by Cooper’s orders, we’d find out far too late.
As for that, Leffler et al. also find that lockdowns are not associated with reduced mortality from COVID-19. Despite his interest in the mask portion of this study, Cooper seems doggedly uninterested with this finding — q.v., Cooper reiterated his threats for further lockdowns when announced his “Modified Stay-At-Home” (i.e., modified lockdown) order. Such cherry-picking cannot support extreme emergency orders.