Mask Use and Ventilation Improvements to Reduce COVID-19 Incidence in Elementary Schools

Georgia, November 16-December 11, 2020

Jenna Gettings, DVM; Michaila Czarnik, MPH; Elana Morris, MPH; Elizabeth Haller, MEd; Angela M. Thompson-Paul, PhD; Catherine Rasberry, PhD; Tatiana M. Lanzieri, MD; Jennifer Smith-Grant, MSPH; Tiffiany Michelle Aholou, PhD; Ebony Thomas, MPH; Cherie Drenzek, DVM; Duncan MacKellar, DrPH

Disclosures

Morbidity and Mortality Weekly Report. 2021;70(21):779-784.

Abstract and Introduction

Introduction

To meet the educational, physical, social, and emotional needs of children, many U.S. schools opened for in-person learning during fall 2020 by implementing strategies to prevent transmission of SARS-CoV-2, the virus that causes COVID-19.^[1,2] To date, there have been no U.S. studies comparing COVID-19 incidence in schools that varied in implementing recommended prevention strategies, including mask requirements and ventilation improvements*.^[2] Using data from Georgia kindergarten through grade 5 (K–5) schools that opened for in-person learning during fall 2020, CDC and the Georgia Department of Public Health (GDPH) assessed the impact of school-level prevention strategies on incidence of COVID-19 among students and staff members before the availability of COVID-19 vaccines.^† Among 169 K–5 schools that participated in a survey on prevention strategies and reported COVID-19 cases during November 16–December 11, 2020, COVID-19 incidence was 3.08 cases among students and staff members per 500 enrolled students.^§ Adjusting for county-level incidence, COVID-19 incidence was 37% lower in schools that required teachers and staff members to use masks, and 39% lower in schools that improved ventilation, compared with schools that did not use these prevention strategies. Ventilation strategies associated with lower school incidence included methods to dilute airborne particles alone by opening windows, opening doors, or using fans (35% lower incidence), or in combination with methods to filter airborne particles with high-efficiency particulate absorbing (HEPA) filtration with or without purification with ultraviolet germicidal irradiation (UVGI) (48% lower incidence). Multiple strategies should be implemented to prevent transmission of SARS-CoV-2 in schools;^[2] mask requirements for teachers and staff members and improved ventilation are important strategies that elementary schools could implement as part of a multicomponent approach to provide safer, in-person learning environments. Universal and correct mask use is still recommended by CDC for adults and children in schools regardless of vaccination status.^[2]

Beginning in fall 2020, many Georgia schools opened for in-person learning. At that time, GDPH required all Georgia schools to submit weekly data on the aggregate number of COVID-19 cases among students and staff members.^¶ School-associated cases were self-reported by parents and guardians of students, or staff members, or those reported by local public health officials. On November 16, 2020, the Georgia Department of Education and local health districts emailed an online survey on behalf of CDC and GDPH to all Georgia public K–5 school district superintendents (1,321 schools) and private school leaders (140 schools) to assess school and student characteristics and COVID-19 prevention strategies implemented at the time of the survey. Weekly reminders were sent for 3 additional weeks. Surveys were completed by principals (67.0%), nurses (12.0%), assistant principals (4.7%), or other school representatives (16.4%). School characteristics assessed included school type,** urban-rural classification,^†† and instructional model.^§§ Student characteristics assessed included racial/ethnic distribution^¶¶ and percentages of students who received in-person instruction. Prevention strategies assessed included mask requirements for teachers, staff members, and students; ventilation improvements***; physical distancing of desks (≥6 ft apart); barriers on student desks; class size (number of students in a classroom); cohort size (small groups of students who stay together throughout the day during in-person learning); and number and locations of available handwashing stations. Survey data were collected by CDC and stored in REDCap (version 9.7; Vanderbilt University).

Reported COVID-19 cases submitted to GDPH and online survey data collected during November 16–December 11, 2020, were linked by school to examine associations between prevention strategies and COVID-19 incidence, defined as number of cases among students and staff members per 500 enrolled students during the study period. Rate ratios (RRs) and 95% confidence intervals (CIs) were estimated with negative binomial regression models, adjusted for county-level 7-day incidence (cases per 100,000 population) on December 1, 2020.^††† Rate ratios with 95% CIs excluding 1.0 were considered statistically significant. Analyses were conducted in R (version 4.0.2; The R Foundation). This activity was reviewed by CDC and was conducted consistent with applicable federal law and CDC policy.^§§§

Representatives from 169 (11.6%) of 1,461 schools in 51 (32.1%) of 159 Georgia counties (median = two schools per county) completed the survey and also had available COVID-19 case data (Figure).^¶¶¶ Schools reporting 100% virtual learning were excluded. Among the 169 schools, 162 (95.9%) were public, representing 47 (26.0%) of 181 public school districts in Georgia (median = two schools per district). Schools had a median of 532 enrolled students (attending virtually and in-person), 91.1% were publicly funded, 71.0% were located in metropolitan areas, and 82.2% used hybrid learning (Table 1). Median class size was 19.0 students (interquartile range [IQR] = 15.0–21.0); median cohort size was 20.0 students (IQR = 15.0–21.0). Among all schools, the proportion of students receiving at least some in-person instruction ranged from 8.5% to 100% (median = 84.7%); 3.0%–100% (median = 64.0%) were eligible for free or reduced-cost meal plans, and approximately one half of students were White (median = 55.1%), followed by Black (median = 17.0%), Hispanic (median = 9.0%), multiracial (median = 4.5%), and Asian (median = 1.0%).****

(Enlarge Image)

Figure.

County-level COVID-19 incidence* on December 1, 2020, among counties with one or more participating elementary schools^† and counties without participating schools — Georgia, November 16−December 11, 2020
Abbreviations: GDPH = Georgia Department of Public Health; K–5 = kindergarten through grade 5.
*County incidence was calculated as the 7-day cumulative sum of COVID-19 cases reported to GDPH divided by the county population multiplied by 100,000 on December 1, 2020. Population estimates for 2019 were provided by the Annual Estimates of the Resident Population for Counties in Georgia from April 1, 2010, to July 1, 2019.
^†GDPH and Georgia Department of Education contacted all public Georgia K−5 superintendents (1,321 schools) and private school leaders (140 schools). Representatives from 169 schools with available case data completed the survey (11.6% of schools contacted).

Prevention strategies implemented at participating schools included requiring masks for teachers and staff members (65.1%) or students (51.5%), flexible medical leave for teachers (81.7%), improved ventilation (51.5%), spacing all desks ≥6 ft apart (18.9%), and using barriers on all desks (22.5%). Schools reported a median of 9.0 (IQR = 8.0–9.0) locations with handwashing stations (Table 1).

During the 26 days from November 16 through December 11, 2020, participating schools reported a median of two COVID-19 cases (range = 0–15); COVID-19 incidence for all schools combined was 3.08 cases among students and staff members per 500 enrolled students. Community incidence in counties with participating schools during the same period was 1,055 per 100,000 persons of all ages, or approximately 5.28 per 500 population.^†††† Mask requirements for teachers and staff members (RR = 0.63) and improved ventilation (RR = 0.61) were associated with lower incidence (Table 1). Among 123 schools that reported on ventilation improvements, dilution methods (opening doors, opening windows, or using fans) alone (RR = 0.65), or in combination with filtration (installation of HEPA filters) with or without purification (installation of UVGI) (RR = 0.52) were associated with lower COVID-19 incidence (Table 2).

1 2

Next Section

Abstract and Introduction
Discussion
References
Sidebar

*Ventilation strategies include dilution methods (opening doors, opening windows, and using fans to improve circulation from open windows); filtration methods (installation of high-efficiency particulate absorbing [HEPA] filters); and purification methods (installation of ultraviolet germicidal irradiation [UVGI] units, installed in upper room areas and shielded from persons or installed in the heating, ventilation, and air conditioning [HVAC] system). https://www.cdc.gov/coronavirus/2019-ncov/community/schools-childcare/ventilation.html
^† https://www.cdc.gov/coronavirus/2019-ncov/community/schools-childcare/operation-strategy.html
^§This denominator closely represents the size of elementary schools included in this study (median = 532).
^¶COVID-19 cases among staff members and students are defined as laboratory-confirmed reverse transcription–polymerase chain reaction or rapid antigen positive test results self-reported to the school by staff members and parents or guardians of students or by local public health officials. Schools report aggregate counts of cases among students and staff members weekly to GDPH and are required to report even if they have no cases.
**Public school; public charter, magnet, or alternative school; private, parochial, or independent school.
^††Based on the 2013 National Center for Health Statistics classification. Metro counties include large metro (county population ≥1,000,000), medium metro (250,000–999,999), and small metro (<250,000); nonmetro counties include micropolitan (10,000–49,999) and noncore (nonmetropolitan counties that did not qualify as micropolitan).
^§§For schools that are 100% in-person, students attend in-person for the full school week; for hybrid models, a combination of in-person and remote learning occurs on an alternating schedule.
^¶¶White, African American or Black, Hispanic, Asian, American Indian or Alaska Native, Other Pacific Islander, and Multiracial.
***Schools reported "Yes" or "No" to the question, "Are steps being taken to improve air quality and increase the ventilation in the school?" Schools that responded "Yes" were asked to select from the following options: opening doors, opening windows, using fans to increase effectiveness of open windows, installation of HEPA filtration systems in high-risk areas, or installation of UVGI in high-risk areas. Multiple choices were allowed.
^†††County incidence was calculated as the 7-day cumulative sum of COVID-19 cases reported to GDPH on December 1, 2020, divided by the county population multiplied by 100,000. Population estimates for 2019 were provided by the Annual Estimates of the Resident Population for Counties in Georgia from April 1, 2010, to July 1, 2019. Data were obtained from the U.S. Census Bureau on October 1, 2020.
^§§§45 C.F.R. part 46, 21 C.F.R. part 56; 42 U.S.C. Sect. 241(d); 5 U.S.C. Sect. 552a; 44 U.S.C. Sect. 3501 et seq.
^¶¶¶"Available case data" refers to the weekly aggregate COVID-19 case reports provided by schools to GDPH. Not all schools that completed the survey reported during the study period. Sixty-one schools that completed the survey but did not provide GDPH any weekly COVID-19 reports during the study period were excluded.
****Median proportions of American Indian or Alaska Native and Native Hawaiian or Other Pacific Islander were <1%. Each school reported the proportion of students who identified within the different racial and ethnic groups. The cumulative proportions could not exceed 100%.
^††††Community incidence was calculated for the survey period November 16–December 11, 2020, to allow comparison to school-level incidence during the same period. County-level incidence used for RR estimation and the figure are 7-day cumulative cases per 100,000 population as reported on December 1, 2020.

Table 1. COVID-19 incidence* and rate ratios in 169 elementary schools, ^† by county COVID-19 incidence, school characteristics, and COVID-19 prevention strategies — Georgia, November 16–December 11, 2020
Characteristic	No. (%) of schools	No. of enrolled students	No. of cases^§	Cases per 500 students enrolled (95% CI)	RR^¶ (95% CI)
Total	169 (100)	91,893	566	3.08 (2.84–3.34)	—
County COVID-19 incidence**
0–150	25 (14.8)	12,358	52	2.10 (1.61–2.76)	Ref
151–200	54 (32.0)	32,399	169	2.61 (2.24–3.03)	1.21 (0.75–1.96)
201–250	45 (26.6)	24,482	106	2.16 (1.79–2.62)	1.00 (0.60–1.66)
251–300	21 (12.4)	11,556	122	5.28 (4.42–6.30)	2.55 (1.47–4.47)
>300	24 (14.2)	11,098	117	5.27 (4.40–6.31)	2.26 (1.32–3.88)
School type
Public	154 (91.1)	86,878	536	3.08 (2.84–3.36)	Ref
Public charter/Magnet/Alternative	8 (4.7)	4,645	27	2.91 (2.00–4.22)	0.97 (0.50–1.97)
Private/Parochial/Independent	7 (4.1)	370	3	4.05 (1.38–11.78)	1.46 (0.31–5.33)
Urban–rural setting^††
Metropolitan	120 (71.0)	65,501	386	2.95 (2.67–3.25)	Ref
Nonmetropolitan	49 (29.0)	26,392	180	3.41 (2.95–3.94)	1.14 (0.83–1.58)
Instructional model^§§
100% in-person	30 (17.8)	14,538	106	3.65 (3.02–4.41)	Ref
Hybrid	139 (82.2)	77,355	460	2.97 (2.71–3.26)	0.91 (0.60–1.36)
Mask requirements for teachers and staff members ^¶¶
Optional	57 (33.7)	29,881	264	4.42 (3.92–4.98)	Ref
Required	110 (65.1)	61,190	298	2.44 (2.17–2.73)	0.63 (0.47–0.85)
Mask requirements for students
Optional	82 (48.5)	42,761	326	3.81 (3.42–4.25)	Ref
Required	87 (51.5)	49,132	240	2.44 (2.15–2.77)	0.79 (0.50–1.08)
Flexible medical leave policies for teachers
Not offered	31 (18.3)	17,194	137	3.98 (3.37–4.71)	Ref
Offered	138 (81.7)	74,699	429	2.87 (2.61–3.16)	0.81 (0.56–1.17)
Ventilation improvements
No***	37 (21.9)	21,844	183	4.19 (3.63–4.84)	Ref
Yes	87 (51.5)	44,771	234	2.61 (2.30–2.97)	0.61 (0.43–0.87)
Don't know	45 (26.6)	25,278	149	2.95 (2.51–3.46)	0.63 (0.42–0.95)
Desks or tables separated ≥6 ft
Some/No classrooms	137 (81.1)	76,348	472	3.09 (2.83–3.38)	Ref
All classrooms	32 (18.9)	15,545	94	3.02 (2.47–3.70)	0.97 (0.66–1.45)
Desks or tables with barriers
Some/No classrooms	131 (77.5)	71,163	445	3.13 (2.85–3.43)	Ref
All classrooms	38 (22.5)	20,730	121	2.92 (2.44–3.48)	0.98 (0.69–1.41)
Students per classroom, median (IQR)	19 (15–21)	—	—	—	1.02 (0.98–1.06)
Cohort size,^††† median (IQR)	20 (15–21)	—	—	—	1.00 (1.00–1.00)
Handwashing stations, median (IQR)	9 (8–9)	—	—	—	0.88 (0.76–1.01)

Abbreviations: CI = confidence interval; IQR = interquartile range; GDPH = Georgia Department of Public Health; K–5 = kindergarten through grade 5; RR = rate ratio; Ref = referent.
*Case incidence in schools was calculated as the sum of cases reported to GDPH during November 16–December 11, 2020, divided by the number of students enrolled multiplied by 500.
^†GDPH and Georgia Department of Education contacted all public Georgia K–5 superintendents (1,321 schools) and private school leaders (140 schools); 169 schools with available case data completed the survey (response rate 11.6%).
^§Number includes both students and staff members with a case of COVID-19 during the study period.
^¶All RR estimates except for county COVID-19 incidence were adjusted for county-level 7-day case incidence per 100,000 population on December 1, 2020. RRs that exclude 1 are statistically significant.
**Per 100,000 population. County incidence was calculated as the 7-day cumulative sum of COVID-19 cases reported to GDPH on December 1, 2020, divided by the county population multiplied by 100,000. Population estimates for 2019 were provided by the Annual Estimates of the Resident Population for Counties in Georgia from April 1, 2010, to July 1, 2019.
^††Based on the 2013 National Center for Health Statistics classification. Metro counties include large metro (county population ≥1,000,000), medium metro (250,000–999,999), and small metro (<250,000); nonmetro counties include: micropolitan (10,000–49,999) and noncore (nonmetropolitan counties that did not qualify as micropolitan).
^§§For schools that are 100% in-person, students attend in-person for the full school week; for hybrid models, a combination of in-person and remote learning occurs on an alternating schedule.
^¶¶Two schools had discordant mask requirements for teachers and other staff members (i.e., one school required mask use among teachers, but not other staff members, and one school required mask use among other staff members, but not teachers). These were excluded from the calculation of the RR for mask requirements for teachers and staff members. All other schools either required masks for both teachers and staff members or allowed for optional mask use among both groups.
***Includes schools that reported "No" to improving ventilation and six schools that reported decreasing room occupancy as the only ventilation improvement.
^†††Small groups of students who stay together throughout the day during in-person learning.

Table 2. COVID-19 incidence* and rate ratios in 123 elementary schools, ^† by type of ventilation improvement as a COVID-19 prevention strategy — Georgia, November 16–December 11, 2020
Ventilation improvement	No. (%) of schools	No. of enrolled students	No. of cases^§	Cases per 500 students enrolled (95% CI)	RR^¶ (95% CI)
Total	123 (100)	66,499	417	3.13 (2.84–3.44)	—
None**	37 (30.1)	21,844	183	4.19 (3.63–4.84)	Ref
Dilution only^††	39 (31.7)	21,562	127	2.94 (2.48–3.50)	0.65 (0.43–0.98)
Filtration ± purification only^§§	16 (13.0)	9,133	45	2.46 (1.84–3.29)	0.69 (0.40–1.21)
Dilution and filtration ± purification^¶¶	31 (25.2)	13,960	62	2.22 (1.73–2.84)	0.52 (0.32–0.83)

Abbreviations: CI = confidence interval; GDPH = Georgia Department of Public Health; HEPA = high-efficiency particulate absorbing; RR = rate ratio; UVGI = ultraviolet germicidal irradiation; ± = with or without.
*Case incidence in schools was calculated as the sum of cases reported to GDPH during November 16–December 11, 2020, divided by the number of students enrolled multiplied by 500.
^†Excludes schools from the original 169 that reported "Don't know" to improving ventilation (n = 45) and one school that reported only using an air purification strategy.
^§Number includes both students and staff members with a case of COVID-19 during the study period.
^¶Adjusted for county-level 7-day case incidence per 100,000 population on December 1, 2020.
**Includes schools that reported "No" to improving ventilation and six schools that reported decreasing room occupancy as the only ventilation improvement.
^††Opening doors, opening windows, or using fans.
^§§Using HEPA filters with or without using UVGI and not opening doors, opening windows, or using fans.
^¶¶Opening doors, opening windows, or using fans, and using HEPA filters with or without using UVGI.