Analysis of mask mandate effectiveness

In our recent reports on COVID-19 masks and school reopenings, we intentionally avoided policy recommendations. We wanted to provide a neutral presentation of the available scientific evidence and opinion. However, the Provo City Council asked me to provide an informal assessment of the effectiveness of mask mandates. They are considering an ordinance in preparation of the return of college students. 

Before getting into the report, I wanted to say how impressed I have been at the decency and concern of our community and the city council. Last night, there was a 4-hour council meeting on this subject, with abundant public comment. Nearly every caller was concise and respectful. The council was deliberative and thoughtful. We need honest and respectful communication and collaboration to get through this, and the interactions over the past few days have made me hopeful for the future.

A rapid review of mask mandates

From our conversation, it seemed there were three specific questions, which I try to answer below:

1.  How effective are mask ordinances (how much increase in compliance and decrease in spread/death can be expected)?
2. How much does executive-legislative conflict influence effectiveness of mandates?
3. What specific behaviors and environments should the ordinance target to be the most effective while limiting unnecessary inconveniences and restrictions?

1. Effectiveness of mask mandates:

The response to mandates varies with local conditions, with estimates ranging from a 10 to 40% increase, after controlling for other factors^1–6. This aligns with the less reliable self-reported data indicating ~90% of individuals living in areas with mandates are wearing masks in public settings versus 65 to 80% of individuals overall^7,8. The range of estimates of how much mandates decrease transmission and subsequent deaths is broader, ranging from 2 to 55%^1,5,9,10. The reasons for this large range are that local conditions are very diverse, mandated actions are different, compliance is variable, and the current data are limited.

My personal assessment based on available information is that the initial decrease in transmission would likely be in the high single digits (6-9%), with the decrease in serious complications and death perhaps twice that (10-20%). The larger decrease in deaths and serious complications is due to the growing evidence that viral load (how much of the virus you are exposed to initially) strongly correlates with seriousness of symptoms^10,11. I point out that my estimates are qualitative and that we will not know the actual effect size of action or inaction until after this crisis is past.

Because there is so much uncertainty, I think it’s useful to consider the “least effective” scenario at the low end of the estimates (~2% decrease over the first few weeks). When you calculate how that decrease compounds with each generation of the virus, that seemingly modest level of decrease could have a very substantial effect on the duration of the outbreak in our community and the total number of people infected and killed^12–15. Given the higher level of trust and community cohesiveness in our area relative to the national scene, I anticipate that the benefits here would be larger than this scenario.

2. Effects of conflicted government:

In response to the second question, governmental division seems to be the rule not the exception with the response to COVID-19 in the U.S.⁶. At the community, state, and federal level, there are abundant examples of the legislature limiting the executive branch and the other way around^3,16–21. The bright side of this unfortunate situation is that the effect size of mandates estimated by the studies cited in the last section probably are applicable to a “divided” or “conflicted” mandate, such as the one under consideration by the council. Though there are not many quantitative estimates, it seems clear that governmental division decreases but does not negate the effectiveness of mandates^3,6.

One pattern that I observed while reading accounts of different mandates was that opposition and controversy seemed higher before the mandate was passed and much quieter after it was in place. This was the opposite of what I expected, but it seemed consistent across conservative and liberal areas (Texas, Georgia, Michigan, Oregon, etc.)^6,18. In fact, most of the media accounts of conflict surrounding masks seemed to be associated with businesses trying to enforce mask wearing in areas where their municipality or state governments hadn’t acted. I don’t have quantitative data to back this up, but it seemed like a consistent trend: lots of opposition and conflict before the mandate, but acceptance and relative solidarity afterward, especially as infection numbers drop. This is supported by actual data from Germany, where mandatory masking was perceived as more fair and effective than voluntary masking, resulting in increased compliance even among skeptical citizens after mandated use²².

3. Specific actions to maximize benefit and minimize costs

Nationwide, 34 states now mandate face coverings in public, and many counties and cities in states without mandates have ordinances of their own (map of areas with mandates below)¹⁸. The details of the mandates vary widely, but the general requirements are surprisingly similar. Most require masks in indoor public spaces, public transportation, and outdoor spaces when unable to maintain physical distancing. There is a useful nontechnical summary of all 50 states’ current policies here, and a more detailed table of specific types of mandates here.

Most of the mandates target areas and behaviors that are known to be high risk or unavoidable. Some researchers have used the “three Cs” to describe risk factors: closed spaces with poor ventilation, crowded places, and close-contact settings²³. The commonsense principle is that risk increases with proximity, duration of contact, and lack of ventilation. While it is best to avoid these conditions altogether, ensuring mask wearing in these situations should be a priority.

There is some new evidence from contact tracing studies and cell-phone mobility data that is directly pertinent to questions of how surgical or general a mandate needs to be to have an effect^24–27. One of the most rigorous studies to date found that large gatherings are increasingly driving the outbreak in the United States²⁶. Based on nearly 8,000 cases with traced contacts, they found that 2% of the infected individuals (mainly people under 40) accounted for 20% of the new cases across age groups. Cities and counties with more events have substantially higher spread²⁶. It also appears that work-related transmission (especially healthcare, transportation, retail, and service industry) accounted for almost half of all new cases before the shutdown²⁸, meaning that work is likely to be a hot spot of transmission as economic activity returns. Targeting these areas (large gatherings, workplaces, and any indoor spaces) would likely have the biggest impact.

However, there is evidence that individual “surgical” actions will be less effective than a general mandate^3,22. For example, in one of the largest analyses of mandate effectiveness, the 15 states in the study that issued public mandates for all citizens saw substantial and rapid decreases in disease, while the 20 states that issued mandates only for the workplace environment saw no change⁵. This emphasizes how masking is not a perfect protection or silver bullet. It works best when used widely in combination with other measures. However, it has the large advantage of having very few social and economic side effects compared to shutdowns, distancing, and other more invasive interventions.

One of the things emphasized by the public health and policy research was that clear explanations for why mandates are given increases trust and compliance¹³. People that do not understand or believe that masks work are understandably less likely to wear them effectively^7,29, though they still show an increase in masking and other protective measures following mandates^7,22. It can be helpful to share information about the many secondary benefits of masking, which have now been consistently observed: greater economic activity, lower risk in the same situations, and complementary benefits from the “reminder factor” (there is greater awareness, greater social distancing, better hygiene, and fewer large gatherings)^2,22,30. Additionally, new research indicates that cloth masks actually do decrease risk to the wearer¹¹, adding another reason to comply, even if only out of self-interest.

 Citations:

1. Chernozhukov, V., Kasahara, H. & Schrimpf, P. Causal Impact of Masks, Policies, Behavior on Early Covid-19 Pandemic in the U.S. medRxiv 2020.05.27.20115139 (2020) doi:10.1101/2020.05.27.20115139.
2. Friedrichsen, J. & Balleyer, A. The Behavioral Effect of Facial Masks During the COVID-19 Pandemic. (2020) doi:10.1257/rct.5735-1.0.
3. Gostin, L. O., Cohen, I. G. & Koplan, J. P. Universal Masking in the United States: The Role of Mandates, Health Education, and the CDC. JAMA (2020) doi:10.1001/jama.2020.15271.
4. Li, T., Liu, Y., Li, M., Qian, X. & Dai, S. Y. Mask or no mask for COVID-19: A public health and market study. PLOS ONE 15, e0237691 (2020).
5. Lyu, W. & Wehby, G. L. Community Use Of Face Masks And COVID-19: Evidence From A Natural Experiment Of State Mandates In The US. Health Affairs 39, 1419–1425 (2020).
6. Milosh, M., Painter, M., Van Dijcke, D. & Wright, A. L. Unmasking Partisanship: How Polarization Influences Public Responses to Collective Risk. https://papers.ssrn.com/abstract=3664779 (2020) doi:10.2139/ssrn.3664779.
7. Knotek, E. S. et al. Consumers and COVID-19: Survey Results on Mask-Wearing Behaviors and Beliefs. EC (FRB Cleveland) 1–7 (2020) doi:10.26509/frbc-ec-202020.
8. More Democrats than Republicans say they wore masks in stores all or most of the time. Pew Research Center https://www.pewresearch.org/wp-content/uploads/2020/06/ft_2020.06.23_masks_02.png (2020).
9. Leffler, C. T. et al. Association of country-wide coronavirus mortality with demographics, testing, lockdowns, and public wearing of masks. Update August 4, 2020. http://medrxiv.org/lookup/doi/10.1101/2020.05.22.20109231 (2020) doi:10.1101/2020.05.22.20109231.
10. Leffler, C. T., Ing, E., McKeown, C. A., Pratt, D. & Grzybowski, A. Final Country-wide Mortality from the Novel Coronavirus (COVID-19) Pandemic and Notes Regarding Mask Usage by the Public. In Review 14 (2020).
11. Gandhi, M. Masks Do More than Protect Others during COVID-19: Reducing the Inoculum of SARS-CoV-2. 11 https://ucsf.app.box.com/s/blvolkp5z0mydzd82rjks4wyleagt036 (2020).
12. Ali, S. T. et al. Serial interval of SARS-CoV-2 was shortened over time by nonpharmaceutical interventions. Science (2020) doi:10.1126/science.abc9004.
13. Fisher, D. A. & Carson, G. Back to basics: the outbreak response pillars. The Lancet 0, (2020).
14. Goldman Sachs Research. Face Masks and GDP. https://www.goldmansachs.com/insights/pages/face-masks-and-gdp.html (2020).
15. Kai, D., Goldstein, G.-P., Morgunov, A., Nangalia, V. & Rotkirch, A. Universal Masking is Urgent in the COVID-19 Pandemic: SEIR and Agent Based Models, Empirical Validation, Policy Recommendations. arXiv:2004.13553 [physics, q-bio] (2020).
16. NW, 1615 L. St, Suite 800Washington & Inquiries, D. 20036USA202-419-4300 | M.-857-8562 | F.-419-4372 | M. Republicans, Democrats Move Even Further Apart in Coronavirus Concerns. Pew Research Center - U.S. Politics & Policy https://www.pewresearch.org/politics/2020/06/25/republicans-democrats-move-even-further-apart-in-coronavirus-concerns/ (2020).
17. Parshley, L. America’s slow, painful shift to enforcing mask mandates. Vox https://www.vox.com/2020/8/4/21354325/coronavirus-face-mask-mandate-enforcement (2020).
18. Wright, A. L., Chawla, G., Chen, L. & Farmer, A. Tracking Mask Mandates During the Covid-19 Pandemic. SSRN Journal (2020) doi:10.2139/ssrn.3667149.
19. Alemany, J. Analysis | Power Up: Mask mandates are coming to more states. They were controversial in 1918, too. Washington Post.
20. McKelvey, T. Why are Americans so angry about masks? BBC News (2020).
21. Aratani, L. How did face masks become a political issue in America? The Guardian (2020).
22. Betsch, C. et al. Social and behavioral consequences of mask policies during the COVID-19 pandemic. https://osf.io/gn6c9 (2020) doi:10.31234/osf.io/gn6c9.
23. Stein-Zamir, C. et al. A large COVID-19 outbreak in a high school 10 days after schools’ reopening, Israel, May 2020. Eurosurveillance 25, 2001352 (2020).
24. Abbott, B. W. et al. Making sense of the research on COVID-19 and school reopenings. 22 http://rgdoi.net/10.13140/RG.2.2.24052.17285 (2020).
25. Fontanet, A. et al. Cluster of COVID-19 in northern France: A retrospective closed cohort study. medRxiv 2020.04.18.20071134 (2020) doi:10.1101/2020.04.18.20071134.
26. Lau, M. S. Y. et al. Characterizing superspreading events and age-specific infectiousness of SARS-CoV-2 transmission in Georgia, USA. PNAS (2020) doi:10.1073/pnas.2011802117.
27. Park, Y. J. et al. Early Release - Contact Tracing during Coronavirus Disease Outbreak, South Korea, 2020. Emerging Infectious Diseases Volume 26, (2020).
28. Lan, F.-Y., Wei, C.-F., Hsu, Y.-T., Christiani, D. C. & Kales, S. N. Work-related COVID-19 transmission in six Asian countries/areas: A follow-up study. PLOS ONE 15, e0233588 (2020).
29. Brzezinski, A., Kecht, V., Van Dijcke, D. & Wright, A. L. Belief in Science Influences Physical Distancing in Response to COVID-19 Lockdown Policies. SSRN Journal (2020) doi:10.2139/ssrn.3587990.
30. Mantzari, E., Rubin, G. J. & Marteau, T. M. Is risk compensation threatening public health in the covid-19 pandemic? BMJ 370, (2020).