Background: Teachers experience some of the highest levels of occupational stress and lowest levels of well-being among all professions (Johnson et al., 2005; Bauer et al., 2006). The negative consequences of teachers' mental health are wide-ranging and may contribute to feelings of burnout, stress, and poor coping styles, ultimately impacting their ability to employ effective teaching practices (Hernan et al., 2018; Maslach et al., 2001; Braun et al., 2018). With the unprecedented disruption to the U.S. education system caused by the COVID-19 pandemic, teachers were forced to manage additional, unique disruptions and challenges, such as pivoting to remote teaching (Lupton-Smith et al., 2021). As districts continue to grapple with decisions regarding in-person and remote modalities, teachers have faced high uncertainty. Research is needed to understand teacher mental health during the pandemic, in order to understand groups in need of additional supports and potential interventions. Research Question: The aim of the current study was to deepen our understanding of the associations between the COVID-19 pandemic and teachers' mental health, among teachers across the United States. First, we examined differences in mental health during the pandemic between educators and other professions. Second, we compared mental health during the pandemic between in-person (IP) and remote teachers. Following findings by Johnson et al. (2005), we hypothesized that teachers would present higher distress, compared to other professions. We further hypothesized that IP teachers would show lower distress than remote teachers, as social interactions at school have been shown to contribute to teacher satisfaction (Veldman et al., 2016). Study Design and Procedure: We leveraged data from the COVID-19 Symptom Survey, a large online survey developed in collaboration between Carnegie Mellon University's Delphi Group and Facebook (Delphi Group, 2021). This cross-sectional daily survey invites a stratified random sample of Facebook users to take the survey and consists of questions related to health symptoms, preventive behaviors, mental health, and more (Delphi Group, 2021). We used data from all adult participants who responded to the survey from September 8 , 2020 until March 28 , 2021 (N = 5,186,378). Table 1 provides demographic information by job type across all employed respondents (N = 2,775,974) and by IP and remote for teachers only (N = 297,607). Measures: Three measures of mental health were examined: 1) depressive symptoms, 2) anxiety symptoms, and 3) feelings of isolation. All three items shared the following question stem: "In the past 7 days, how often have you ...". The three items were originally scored along a four-point Likert scale with responses ranging from 1 = none of the time to 4 = all of the time. We recoded each item into a dichotomous indicator, where 0 = none or some of the time, and 1 = most or all of the time. Sociodemographic characteristics (e.g., gender, age, education level, number of children, financial worry) were also collected in the survey and included in the models as covariates. We also controlled for county-level covariates: urbanicity, COVID-19 cases, and COVID-19 deaths, and included U.S. State and month as fixed effects. Analyses: For aim one, we assess potential disparities in mental health outcomes between education, healthcare, office, and other workers during the pandemic, in terms of their probability of reporting a negative mental health outcome. For aim two, inperson (IP) teachers were compared to remote teachers regarding that same probability. To quantify these associations, each of the three mental health indicators was separately considered an outcome, resulting in a total of six logistic regression models (three for each aim), with each of the outcomes modeled as a function of the individual- and county-level covariates listed above. Each model was weighted for non-response and coverage bias following the weighting scheme outlined in Barkay et al. (2020). All models were estimated using the survey package in R (Lumley, 2020), stratified by state. Results: Aim 1: All professions: Logistic regression results indicated educators, coded as 0 as the baseline group, were significantly more likely to report depression and anxiety symptoms than healthcare (OR = 0.91, p < 0.001; OR = 0.70, p < 0.001), office (OR = 0.99, p < 0.001; OR = 0.82, p < 0.001), and other (OR = 0.96, p < 0.001; OR = 0.76, p < 0.001) workers, although the point estimates were sometimes very close to 1 (see Table 2). Educators were significantly more likely to report feeling isolated than were healthcare (OR= 0.83, p < 0.001) and other workers (OR = 0.96, p < 0.001), but less so than office workers (OR = 1.05, p < 0.001). Aim 2: IP vs. remote teaching: Considering teachers exclusively, logistic regression results indicated remote teachers were significantly more likely to report depressive symptoms (OR = 0.89, p < 0.001), anxiety symptoms (OR = 0.92, p < 0.001), and feelings of isolation (OR = 0.65, p < 0.001) than IP teachers. See Table 2 for regression results. Conclusions: Teachers continue to navigate high uncertainty related to school reopening decisions due to the COVID-19 pandemic. Various guidelines have been proposed by organizations such as the U.S. Centers for Disease Control and Prevention (CDC; 2021) and the National Academies of Sciences, Engineering, and Medicine (2020) for safe and supportive learning environments as schools reopen, with reports often focusing on instructional modes of teaching or meeting students' social-emotional needs. Teachers play a crucial role in such efforts, but empirical studies using nationally representative data regarding teacher mental health throughout the pandemic have been scarce. Our study indicates teachers show a significantly higher prevalence of negative mental health outcomes compared to other professionals. Additionally, remote teachers show significantly higher levels of distress than IP teachers for all mental health items considered in the study. This line of research has important potential implications for intervention research and practice to address teachers' urgent mental health needs. Understanding the ways in which the mode of instruction during school reopening is related to teacher well-being can help inform interventions aimed to support and promote teachers' mental health in this and future public health emergencies, ultimately supporting effective teaching practices.