Back in January or so, Bavaria at least mandated FFP2 (roughly equivalent to US N95) masks be worn in shops, buses and presumably similar environments. At least Berlin followed suit some months later with a similar measure.

Is there any epidemiological study that tries to isolate the effects of this specific measure (relative to mask mandates in general, using just surgical masks)?

I've done a bit of searching, all I can find are more news stories on the FFP2 mandating... but noting much on whether it had any specific effects.

I'm asking about this because the different times when masks (of any kind) were mandated in various parts Germany [earlier in the pandemic] was used as a way to set-up a comparison in a fairly cited study on mask-mandates effectiveness. So, I was hoping something similar could have been done for the effects of [mandating] FFP2/N95 in the general population (well, at least in those settings that they were mandated for).

2 Answers 2


Apparently no reports from that German experience, but recently researchers published a working paper with the results of "A randomized-trial of community-level mask promotion in rural Bangladesh during COVID-19 shows that the intervention tripled mask usage and reduced symptomatic SARS-CoV-2 infections ..." https://www.poverty-action.org/publication/impact-community-masking-covid-19-cluster-randomized-trial-bangladesh

It seems likely that this trial did not encounter much of the delta variant, as it seems that interventions were completed by April 2021 and delta seemed to hit Bangladesh in late June 2021.

Note - this is not yet reviewed/published, but it does seem to be a proper paper with methodological details, results, etc.

Here's the Stanford press release about the study.

  • Interesting find, which does in fact intend to cite the Bundgaard paper. But as you note, it's not peer reviewed and is not even on a pre-print server. Also the confidence intervals for the outcomes reach 1.0, so pretty weak evidence if even significant.
    – Ian Campbell
    Sep 19, 2021 at 23:34
  • @IanCampbell Although it's not yet peer reviewed, it does seem to be a "real" paper with the actual data and methodology. Among the authors are some with big-name affiliations (e.g. 4. Division of Infectious Diseases and Geographic Medicine, Stanford University, Stanford, CA, USA; 7. Department of Epidemiology and Public Health, Stanford University, Stanford, CA, USA) so I think it's worth taking seriously.
    – Armand
    Sep 19, 2021 at 23:40
  • 1
    I agree that the methodology seems sound, and the number of participants is impressive. I'm not sure too much weight should be placed on "big-name affiliations" though.
    – Ian Campbell
    Sep 19, 2021 at 23:48
  • Not sure how that Bangladesh study is relevant to N95/FPP2 tough. It seems it was regular surgical masks vs cloth masks vs no masks. Still an interesting study. Sep 20, 2021 at 0:08

JAMA has a randomized clinical trial in 2019: "N95 Respirators vs Medical Masks for Preventing Influenza Among Health Care Personnel". The finding is "there was no significant difference in the incidence of laboratory-confirmed influenza".

Regarding Bavaria's N95/FFP2 respirator mandate, there's a graph of Bavaria vs Germany daily new cases in this article: https://swprs.org/face-masks-evidence/#development-of-cases-after-mask-mandates

  • 1
    The linked webpage seems to have a strong anti-mask bias, and so presented data should be considered in that context. (Note, however, that I am not suggesting that there is strong evidence that mask mandates are effective either.)
    – Ian Campbell
    Sep 19, 2021 at 21:25
  • Not sure how relevant influenza is to COVID-19.
    – Armand
    Sep 19, 2021 at 23:30
  • 1
    the thing about studies like the influenza one, which was on healthcare workers (HCW)... using masks only at work is that there's plenty of opportunity to get influenza in the community, when they're not wearing masks. So that may or may not be indicative what happens in a mask-mandate situation. In the SARS outbreak, where the HCWs could really only get infected at work, there was a difference. pubmed.ncbi.nlm.nih.gov/15030692 Sep 20, 2021 at 0:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.