Adoption of Strategies to Mitigate Transmission of COVID-19 During a Statewide Primary Election — Delaware, September 2020

Delaware, September 2020

Eva Leidman, MSPH; Noemi B. Hall, PhD; Amy E. Kirby, PhD; Amanda G. Garcia-Williams, PhD; Jose Aponte; Jonathan S. Yoder, MPH; Rick Hong, MD; Anthony Albence, MA; Fátima Coronado, MD; Greta M. Massetti, PhD


Morbidity and Mortality Weekly Report. 2020;69(33):1571-1575. 

In This Article

Abstract and Introduction


Elections occurring during the coronavirus disease 2019 (COVID-19) pandemic have been affected by notable changes in the methods of voting, the number and type of polling locations, and in-person voting procedures.[1] To mitigate transmission of COVID-19 at polling locations, jurisdictions have adopted changes to protocols and procedures, informed by CDC's interim guidance, developed in collaboration with the Election Assistance Commission.[2] The driving principle for this guidance is that voting practices with lower infection risk will be those which reduce the number of voters who congregate indoors in polling locations by offering a variety of methods for voting and longer voting periods. The guidance for in-person voting includes considerations for election officials, poll workers, and voters to maintain healthy environments and operations. To assess knowledge and adoption of mitigation strategies, CDC collaborated with the Delaware Department of Health and Social Services and the Delaware State Election Commission on a survey of poll workers who served during the statewide primary election on September 15, 2020. Among 522 eligible poll workers, 93% correctly answered all three survey questions about COVID-19 transmission. Respondents noted that most voters and poll workers wore masks. However, masks were not always worn correctly (i.e., covering both the nose and mouth). Responses suggest that mitigation measures recommended for both poll workers and voters were widely adopted and feasible, but also highlighted gaps in infection prevention control efforts. Strengthening of measures intended to minimize the risk of poll workers acquiring COVID-19 from ill voters, such as additional training and necessary personal protective equipment (PPE), as well as support for alternative voting options for ill voters, are needed. Adherence to mitigation measures is important not only to protect voters but also to protect poll workers, many of whom are older adults, and thus at higher risk for severe COVID-19–associated illness. Enhanced attention to reducing congregation in polling locations, correct mask use, and providing safe voting options for ill voters are critical considerations to minimize risk to voters and poll workers. Evidence from the Delaware election supports the feasibility and acceptability of implementing current CDC guidance for election officials, poll workers, and voters for mitigating COVID-19 transmission at polling locations.[2]

Among the 2,498 poll workers who served at one of the 434 polling locations operational during Delaware's primary election, 1,595 (64%) with valid e-mail addresses* were invited by their county elections office to complete a self-administered survey during September 23–26. Poll workers with e-mail addresses were eligible to participate if they worked on election day (September 15, 2020), were aged ≥18 years, and provided written consent to participate. Overall, 568 (36%) persons responded to the survey, among whom 522 (92%) were eligible to participate. Survey questions focused on direct observation of supply availability and polling location setup, training received, knowledge and attitudes about transmission and personal protection, and mitigation measures practiced by themselves, other poll workers, and voters. The survey was administered as a web-based Epi Info questionnaire. Data were analyzed using R statistical software (version 3.5.0; The R Foundation) and SAS (version 9.4; SAS Institute). Differences in proportions were assessed using chi-squared tests, with p-values <0.05 considered statistically significant. Industry and occupation were coded using CDC's National Institute for Occupational Safety and Health Industry and Occupation Computerized Coding System.§ This activity was reviewed by the Delaware Department of Health and Social Services and CDC and was conducted consistent with applicable federal law and CDC policy.

The median age of respondents was 59 years (interquartile range = 52–69 years); 42% were aged >65 years (Table 1). The majority (57%) of respondents were male, 48% were non-Hispanic White, 42% were retired, nearly one third (32%) reported having one underlying medical condition associated with increased COVID-19 severity, and approximately one quarter (27%) reported having two or more such conditions.[3]

Physical modifications to polling locations were reported by respondents, including spacing of voting booths ≥6 feet apart (88%), modifying polling location layout such that voters moved through the space in one direction (80%), and use of visual cues to remind voters to stay ≥6 feet apart (87%) (Table 2). Use of physical barriers, such as plexiglass shields, at registration desks and between voting booths was reported by 5% and 7% of respondents, respectively. Separate doors for entry and exit were reported by 45% of respondents. In response to questions about supplies to support safe hygiene behaviors, 94% of respondents reported that hand sanitizer was available for poll workers, 82% reported that hand sanitizer was available for voters, and 93% reported that cleaning supplies were available; however, 14% reported that their polling location ran out of hand sanitizer or cleaning supplies on election day. Availability of masks for poll workers at polling stations was reported by 88% of respondents and for voters by 70%.

Receipt of training specific to COVID-19 mitigation was reported by 80% of respondents (Table 3). The training content most commonly reported by respondents included guidance on hand hygiene, mask use, and procedures for poll workers with symptoms. Among those respondents who received training, only 30% reported receiving training specific to assisting voters with symptoms consistent with COVID-19 or with known COVID-19 infection. Despite differences in training duration and content, 93% of respondents correctly answered all three survey questions about COVID-19 transmission,** and 94% agreed or strongly agreed that they knew how to keep themselves safe from COVID-19.

Personal prevention practices were reported to have been widely adopted by poll workers and voters. Nearly all respondents (99%) reported that masks were worn by most (i.e., 80%–100%) other poll workers. A similarly high proportion of respondents (98%) reported that masks were worn by most voters. A larger percentage of respondents (73%) reported very rarely or never observing incorrect mask use (i.e., not worn over both the nose and mouth) by other poll workers compared to 54% of respondents reporting very rarely or never observing incorrect mask use by voters. In addition, a larger percentage of respondents reported frequently or very frequently observing hand sanitizer use among poll workers (78%) than reported observing hand sanitizer use among voters (43%). As well, 91% of respondents reported frequently or very frequently having observed fellow poll workers cleaning high touch surfaces and equipment. Nearly all (91%) respondents reported frequently or very frequently observing voters maintaining ≥6 feet of distance from one another.

Nearly three quarters (72%) of respondents reported contact (within 6 feet) with >100 persons and 27% reported close contact (within 6 feet for ≥15 minutes) with >100 persons on election day. Only 19 (4%) of 522 respondents reported knowingly having had contact with a person identified as being ill (with or without a known COVID-19 diagnosis); 15 of those persons reported having worn a mask during contact with the ill voter, but none reported wearing all PPE (respiratory protection, face shields, gowns, and gloves) recommended in interim guidance.[2]

As a proxy for total voters per polling location, experiences of respondents reporting contact with >100 persons were compared with those of respondents reporting fewer contacts for all analyses of mitigation strategies, training, knowledge and attitudes, and exposures. Among respondents indicating polling location worked (128), at least 99 unique sites (23% of all operational polling locations) were represented. Availability of separate doors for voter entry and exit was reported by 37% of respondents having contact with ≤100 persons, compared with 48% of those having >100 contacts (p = 0.02). Compared with respondents having contact with >100 persons, those having contact with ≤100 persons were more likely to report very rarely or never observing voters wearing masks incorrectly (63% versus 49%, p = 0.01). No other statistically significant differences were observed.

*E-mail addresses were considered valid if nonmissing, without obvious typographic errors, and did not return an automated error message.
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.
**Respondents were asked three true or false questions regarding whether 1) SARS-CoV-2 can spread through respiratory droplets, 2) SARS-CoV-2 can spread when in close contact with an infected person, and 3) SARS-CoV-2 can spread by touching a contaminated surface before touching one's face, eyes, or mouth.