COVID: do they work? maybe not in Denmark

 A recent Danish study questions the utility of wearing surgical masks in preventing SARS-CoV-2 transmission rate (seehttps://www.acpjournals.org/doi/10.7326/M20-6817 ) 

 

Details: 

-- 3030 people, randomly assigned to a recommendation to wear masks, were given a supply of 50 surgical masks (3-layer, disposable surgical mask with ear loops) and instructions for proper use. 2994 were assigned to be controls, in April and May 

   -- these facemasks were high quality surgical mask with filtration rate of 98% 

-- all were at least 18 years old and reported being outside the home for at least 3 hours a day and did not wear masks during their daily work

-- mean age 47, 64% female, 20% smokers, 40% wore eyeglasses daily 

-- occupation: shop employee 4%, cashier 4%, craftsperson 4%, office employee 12%, manager 5%, transportation employee 25%, service employee 4%, homecare/nursing home employee 9%, early childhood care staff 4%, salesperson 2% 

-- median of 4.5 hours per day outside of home 

-- all participants received materials and instructions for antibody testing on their receipt and at one month, as well as instructions for collecting oropharyngeal/ nasal swabs for PCR testing at one month and whenever symptoms occurred that were compatible with Covid  

-- 4860 to complete the study (81%) 

 

-- baseline mask use prior to the study in Denmark was <5%, outside of hospitals

   -- recommended public health measures included quarantining those with infection, social distancing, limiting the number of people seen, frequent hand hygiene and cleaning, and limiting visitors to hospitals and nursing homes. Cafés and restaurants were closed during the study until May 18 

-- those in the mask group were told to wear the mask when outside the home for the next month 

-- a non-blinded study (ie, the controls did not have a recommendation to wear anything on their mouths/noses) 

-- primary outcome: SARS-CoV-2 infection in the mask wearer at one month by antibody testing, PCR, or hospital diagnosis of Covid

-- secondary outcome: PCR positivity for other respiratory viruses 

-- 46% of participants reported that they wore their mask as recommended, 47% predominately as recommended, and 7% not as recommended 

 

Results: 

-- infection with SARS-CoV-2: 

   -- those recommended to wear a mask: 42 people (1.8%) 

   -- control group: 53 people (2.1%) 

   -- odds ratio: 0.82 (0.54-1.23), p=0.33 (not statistically significant) 

-- no difference in results by repeated multiple imputation for missing data 

-- no difference by per-protocol analysis that excluded the 7% to reported nonadherence 

-- post hoc (not pre-planned) analyses: 

   -- those who reported to have worn masks “exactly as instructed” had no difference compared to the control group (2.0% vs 2.1%) 

   -- excluding patients who did not provide antibody test results at baseline: no difference 

   -- and no subgroup was found where facemasks were effective at conventional levels of statistical significance (however data was not shown) 

-- for secondary outcomes: 

   -- 9 of the mask wearers were positive for one or more of the 11 respiratory viruses other than SARS-CoV-2, vs 11 in the control group (no difference) 

   -- positivity for respiratory virus including SARS-CoV-2 occurred in 9 wearing masks vs 16 controls, also not even close to statistically significant 

-- there was no difference in results comparing those who were studied during lockdown period in Denmark or those who were studied subsequently 

 

Commentary: 

-- observational evidence does support the efficacy of facemasks in healthcare settings 

    -- a systematic review and meta-analysis found a huge reduction in the risk of infection by 85%, with absolute risk reduction of 3.1% with facemask vs 17.4% without, though the benefit was stronger for N95s or similar respirators as opposed to disposable surgical masks  (see http://gmodestmedblogs.blogspot.com/2020/06/covid-physical-distancing-face-masks.html, or the original article https://www.thelancet.com/action/showPdf?pii=S0140-6736%2820%2931142-9 

        -- also, see the Kansas study from yesterday which reported that mandatory mask usage did decrease SARS-CoV-2 spread (see http://gmodestmedblogs.blogspot.com/2020/11/covid-do-masks-work-they-do-in-kansas.html )

    -- and several studies have suggested that mask effectiveness is greater in healthcare than community settings 

       -- which really is not so surprising. Healthcare settings tend to be more disciplined and focused on Covid than in the community, the quality of the masks used is higher (more N95s) and the proper usage is probably much higher even for surgical masks 

    --which actually supports the conclusion that masks do work well, just that there needs to be more education/reinforcement in the community setting

--although the Danish study was designed to detect a reduction in infection rate from 2% to 1%, the confidence intervals are compatible with a possible 46% reduction to 23% increase infection among mask wearers 

-- there was also no difference in people wearing eyeglasses vs not, suggesting that conjunctival transmission was not a major factor. results also differing from some studies done in other areas. why?? too low a prevalence of SARS-CoV-2 in the areas where the study was done?

--and, there was no difference in other viral infections by mask use, though mask use is considered an important mitigating factor in those with influenza-like syndromes (see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7253999/ and https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2662657/ ); however, another review did not find efficacy for influenza: https://pubmed.ncbi.nlm.nih.gov/32027586/ ).

-- seems to me that the pre-study use of masks in Denmark was so low (<5%) is concerning.  this might suggest that the Danish community had different perceptions of mask utility for respiratory infections. which might suggest that an effective mask-wearing intervention might be harder to effect without lots of education, perhaps a few checks on the actual content of what "mask-wearing" meant for people (covering their neck? mouth? nose?...). And perhaps a significant rrole of social pressures (some in the study reported that there was negative feedback when they wore masks: how did that affect their correct mask use???)

    -- fewer than 1/2 those in the mask group wore them "exactly as instructed", per their self-report. the fact the so few did wear them as instructed amplifies the concern that their self-reports may not be accurate. why did more than 1/2 not use them as reported? did those reporting correct usage do so to please those doing the study researchers? were they embarrassed that they really did not use them so "exactly as instructed"?

    -- and the lack of efficacy for other respiratory infections also raises concerns about actual mask usage and/or use of masks correctly

so, why were the results so different in Denmark than in the US studies??

-- seems to me that the pre-study use of masks in Denmark was so low (<5%), which is concerning.  this might suggest that the Danish community had different perceptions of mask utility for respiratory infections. which might suggest that an effective mask-wearing intervention might be harder to effect without lots of education, perhaps a few checks on the actual content of what "mask-wearing" meant for people (covering their neck? mouth? nose?...). And perhaps a significant role played by social pressures (some in the study reported that there was negative feedback when they wore masks: how did that affect their correct mask use???). And, this negative social role may be more prevalent in those areas in the US that are mask-aversive. my guess is that the minority of counties in Kansas who accepted the mandate to wear masks were also the ones where mask wearing was most tolerated and supported in the community

    -- fewer than 1/2 those in the mask group wore them "exactly as instructed", per their self-report. the fact the so few did wear them as instructed amplifies the concern that their self-reports may not be accurate. why did more than 1/2 not use them as reported? did those reporting correct usage do so to please those doing the study researchers? were they embarrassed that they really did not use them so "exactly as instructed"?

    -- and the lack of efficacy for other respiratory infections also raises concerns about actual mask usage and/or use of masks correctly

Limitations: 

--patient self-reports of mask use may be inaccurate in a society where so few use masks (though seem to value other self-protection against Covid), and there seems to be an underlying stigma against them. and, even those using them, did they do so correctly? 

-- patient self-reported the results of home testing (antibody/PCR). were they done correctly? were they reported correctly? 

-- this was a non-blinded study (of course), perhaps biasing the results. did those wearing masks act differently than those not wearing them?? did they do less distancing from others, thinking that the masks were protecting them??

-- only a one-month study (though this should uncover benefit, from our understanding of SARS-CoV-2 transmission)

-- limited numbers of participants for any effective subgroup analysis (and it seems that there were not enough data collected to see if there were subgroup differences)

-- primary outcome was antibodies against SARS virus (chosen because half the people infected are asymptomatic ), and the sensitivity of the point-of-care test was only 82.5%. would higher sensitivity testing with more accurate data have affected the results? especially in a study with so few endpoints reached?

 

So, this brings up a few issues: 

-- it is possible that surgical masks may in fact not protect the wearer from getting infected (though, the study does not assess whether wearing a facemask protects others, and there are better data showing benefit in this direction, e.g. see https://www.nature.com/articles/s41591-020-0843-2 ). and, surgical masks do take on added importance in preventing viral spread to others in the setting of a virus where 50% of the people are contagious but asymptomatic/presymptomatic  (ie, especially with SARS-CoV-2, generalized mask wearing might be an especially important public health benefit from individual spreading the infection) 

-- though these surgical facemasks by design were of high quality, it may be that those wearing these types of masks with ear loops may often not wear them correctly (as is evident on my frequent walks in my neighborhood where almost everyone wears masks, but often not covering all of the appropriate orifices) 

    -- maybe we should all be wearing N-95 masks, which have the advantage of being more likely to cover the nose and mouth by their design, have in general stronger data to support their effectiveness, and Covid 19 is such a miserable disease that community access to these masks may be totally appropriate 

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list