key: cord-0902502-puwj8wj6
authors: Sugg, Margaret M.; Runkle, Jennifer D.; Andersen, Lauren; Weiser, Jaclyn; Michael, Kurt
title: Crisis response among essential workers and their children during the COVID-19 pandemic
date: 2021-10-19
journal: Prev Med
DOI: 10.1016/j.ypmed.2021.106852
sha: 1a4803cde7a6e2f865343af48ca797fd8c67a460
doc_id: 902502
cord_uid: puwj8wj6

Limited research has been conducted on the mental health concerns of frontline and essential workers and their children during the COVID-19 pandemic in the United States (U.S.). This study examined the association between working on the frontlines in the U.S. during the COVID-19 pandemic (March to July 2020) and personal crisis text concerns (e.g., self-harm, suicidal thoughts, anxiety/stress, and substance abuse) for frontline essential workers and the children of frontline workers. We used a novel data set from a crisis texting service, Crisis Text Line (CTL), that is widely used throughout the U.S.. Generalized Estimating Equations examined the individual association between eight specific crisis types (Depression, Stress/Anxiety, Self-Harm, Suicidal Thoughts, Substance Abuse, Isolation, Relationship Issues, and Abuse) and workers and their children during the early phase of the pandemic. Using CTL concerns as a proxy for the prevalence of mental health issues, we found that children of workers, specifically the youngest demographic (13 years old and under), females, and non-conforming youth had a higher risk of crisis events during the COVID-19 pandemic. Additionally, Hispanic children of workers reported higher rates of stress/anxiety, whereas African American children of workers had higher rates of abuse and depression. Frontline workers had a higher risk of suicidal thoughts, and the risk of crisis events was generally highest for the 25 to 44 age group and males. Increases in CTL usage among frontline workers were noted 14–28 days after spikes in local COVID-19 cases. The research to date has focused on the mental health of frontline essential workers, but our study highlights troubling trends in psychological stress among children of these workers. Supportive interventions and mental health resources are needed not only for frontline essential workers, but for their children too.

In the U.S., little research has been conducted on the mental health of frontline essential workers during the COVID-19 pandemic. Nearly a third of the essential workforce is comprised of healthcare workers (McNicholas & Poydock 2020), whereas others serve as non-health workers (e.g., agriculture, police, food production, or essential manufacturing). Prior research from China, Brazil, and Italy has shown that workers are at risk for developing many adverse mental health symptoms during the pandemic, including anxiety, depression, emotional distress, sleep problems, and substantially higher levels of perceived stress (Lai et 

To adjust for repeated text conversations for each CTL user, we fit logistic regression models using generalized estimating equations (GEE) to estimate adjusted odds ratios (aOR) and 95%

Confidence Intervals (CIs) for each crisis response outcome separately (e.g., stress/anxiety, depression). The covariance structure was selected as AR1 to account for the clustering of repeated texting conversations over time using an actor ID (a unique ID for each CTL user) and was determined using the smallest Quasi Information Criterion (QICu) (Hardin 2005) .

Covariates included in the model were age group, race/ethnicity, sexual orientation, and gender identity. GEE logistic regression models were estimated for the following groups: frontline essential workers compared to adults (non-frontline workers), children of frontline workers compared to children (non-frontline workers), frontline essential workers demographics, and children of frontline essential workers demographics.

We conducted a separate secondary analysis to examine the effects of the COVID-19 rates in the surrounding residential area on the mental health status of workers. Daily counts of COVID-19 cases for each U.S. county from March 23rd, 2020 (the first available date) through July 20th, 2020 were matched using CTL-participant area code data (Elfelt 2020) . As CTL crisis event data were available at the area-code-level (i.e., large irregular spatial boundary) rather than county-level, the daily maximum county-level COVID-19 cases within each area-code spatial unit were used as a proxy for the underlying COVID-19 cases in the community.

Although the area code is not an ideal spatial boundary, our analysis provides a first step in the understanding of how high COVID-19 burden may influence crisis events among workers and their children.

Poisson mixed-effect models were used to examine the effect of county-level COVID-19 rates on individual crisis response outcomes with area code as a random intercept. The association between daily COVID-19 rates and crisis response was investigated at the daily time scale, using a 7-day lag, 14-day lag, and 21-day lag. Results were examined at the national level and for New York (NY), which was an early emerger for the COVID-19 pandemic. For the NY analysis, we used a restricted temporal period with high-CTL usage (April 10 to June 20) among workers and their children (Supplemental Figure 1) . The significance level for all analyses was set at α = .05, and all tests were 2-tailed. All analyses were performed in R using the package geepack, tableone, and lme4 (Højsgaard 2006 , Yan & Fine 2005 , R Core Team 2020), and tables were created using sjPlot (Lüdecke 2020 ). This study was deemed exempt by the institutional IRB (#19-0270).

During the pandemic, a total of 3,045 workers and 4,021 children of workers engaged with the crisis service, resulting in 4,835 and 7,749 conversations, respectively. Active rescues (i.e., initiation of emergency services) were more common among workers themselves (0.7%) than children of frontline workers (0.2%) and non-workers (0.2%) ( Table 1) . Children of essential workers demonstrated higher and more frequent engagement with the CTL service than workers for all crisis concerns except bereavement (5.4% for workers, 5.0% for children), substance abuse (2.9% for workers, 1.5% for children), stress/anxiety (47.1% for workers, 39 .9% for children), and mentioning of COVID-19 (16.3% for workers, 9.6% for children) ( Table   2 ). Compared to other child texters, children of frontline workers were more likely to experience bereavement (5% for children of workers, 3.6% for children of non-workers), substance abuse (1.5% for children of workers, 1.3% for children of non-workers), bullying (3.05% for children of workers, 2.6% for children of non-workers), eating (4.5% for children of workers, 4.2% for children of non-workers), isolation (26.8% for children of workers, 25.2% for children of non-J o u r n a l P r e -p r o o f workers), abuse (8.0% for children of workers, 6.5% for children of non-workers), and relationship issues (36.1% for children of workers, 35.6% for children of non-workers). Frontline workers had a higher number of conversations associated with depression (36.5% for workers, 36 .4% for non-workers), substance abuse (2.9% for workers, 2.2% for non-workers), and bereavement (5.4% for workers, 4.7% for non-workers) ( Table 2) .

Analysis of Frontline/Essential Workers, their Children, and CTL users with no association with frontline/essential workers Figure 1 shows the adjusted odds ratios for a) adult frontline and essential workers compared to adults (non-frontline) and b) children of frontline essential workers compared to children (nonfrontline). Children of workers were 11% more likely to experience isolation and 23% more likely to experience abuse than other children using the service. Frontline workers were 1.15 times more likely to report suicidal thoughts compared to adult texters not engaged in frontline work.

Surprisingly, frontline workers also reported a significantly lower association with abuse and isolation than non-frontline adults texters. Table 3 depicts the relationship between crisis conversations for each outcome and the demographic characteristics of frontline essential workers. The odds of suicidal thoughts and self-harm were significantly lower for Hispanic workers than White workers. African Americans also had significantly lower odds of stress and anxiety than White workers. Notably, mixed-race workers had a 68% increase in the odds of relationship issues compared to White workers, although the sample size was small for this demographic (n=85).

Workers in the 25 to 44 age group were more likely to experience stress/anxiety and substance abuse and significantly less likely to experience abuse, depression, self-harm, and suicidal J o u r n a l P r e -p r o o f thoughts than the workers in the 14 to 24 age group. Workers over 45 had lower odds of relationship issues, depression, and suicidal thoughts (Table 3) .

Female workers had over 200% higher odds of self-harm and abuse and significantly lower odds of substance abuse and suicidal thoughts than their male counterparts. CTL-workers identifying as "no response, other for gender" had much higher odds of substance abuse and abuse and significantly lower odds of relationship issues. Most notably, texts for self-harm were higher among all gender categories compared to male workers, with the highest odds observed for non-binary workers (Table 3) . Table 4 depicts the relationship between demographic characteristics and crisis events among children of frontline workers. African American children of workers had significantly higher odds of abuse and depression and lower odds of stress/anxiety and self-harm than White children of frontline workers. Self-reporting of stress/anxiety was significantly higher among Hispanic children and significantly lower among American Indian/Alaskan Native children compared to White children. Asian children of workers were less likely to report depression in relation to White children of workers.

Children of frontline workers (13 and under) were characterized by a higher crisis response for depression, self-harm, and suicidal thoughts than older children (14 to 24 years old) (Table 4) .

However, children (13 and under) had lower crisis responses for stress/anxiety and relationship issues than adolescents (14 to 24 years old).

Like frontline essential workers, children of workers had over a 50% increase in self-harm across all gender types (except non-binary children) compared to male children. Surprisingly, non-binary children of workers were also characterized by significantly lower odds of isolation, J o u r n a l P r e -p r o o f relationship issues, and depression than males. Children of workers who identified 'no response, other for gender' had lower odds of isolation and higher odds of abuse. Table 5 shows the association between CTL usage for the top eight crisis events for frontline essential workers and children of frontline essential workers at 7-, 14-, 21-, and 28-day lags in county-level COVID-19 cases counts. Results across the entire temporal period demonstrated little to no association between rates of COVID-19 at the county level and CTL usage for children of workers across the US. However, a significant association was observed for frontline workers in the US, particularly for 14-and 21-day lags. A sensitivity analysis was performed that included frontline essential workers and children of frontline essential workers in New York, which experienced high COVID-19 cases during our study period. We found even higher odds of CTL usage in frontline essential workers, particularly for the longer 21-day and 28-day periods. In New York, children of workers also had significantly higher CTL-usage for 21-day and 28-day lags.

Our study leverages a digital texting platform to investigate crisis response among frontline essential workers and their children during the early months of the COVID-19 pandemic. To date, our study is the first to use national crisis response data to examine the mental health response in frontline essential workers and the children of these workers. Results revealed that younger frontline essential workers (age 14 to 24 years) were more likely to connect with CTL for self-harm, suicidal thoughts, depression, and abuse than older workers (age 25 to 44 years) during the pandemic. Workers were more likely to report higher rates of suicidal thoughts than similar demographics with no association to workers. CTL-users whose parents were frontline essential workers were much more likely to experience isolation and abuse than children with J o u r n a l P r e -p r o o f no association to frontline essential workers. Children of workers reporting abuse were more likely to self-report being transgender or 'no response, other for gender,' and a higher proportion of children of workers who texted for abuse concerns were African American. Among children of workers, the risk for self-harm was particularly high for the youngest demographic (age 13 and under). Findings from this research demonstrate the significant mental health burden shouldered by frontline essential workers and an especially alarming trend in more severe crisis Our work parallels other work in the U.S., which has noted increases in psychological distress (e.g., depression, suicidal ideation), and notably higher distress for the younger demographics, Hispanics, and females during the COVID-19 pandemic (McGinty et al. 2020, Killgore 2020, Bruine de Bruin et al. 2020). Among frontline essential workers, workers identifying as nonbinary or female were more likely to report self-harm than male frontline essential workers. In contrast, male workers experienced increased substance abuse and suicidal thoughts during the pandemic. Among minorities, African American children of frontline essential workers were more likely to report increased abuse and depression, and Hispanic children were more likely to experience increased stress and anxiety. In the U.S., severe suicide ideation was reported in Although we could not differentiate between the types of essential frontline workers, we found an alarming increase in CTL usage for all crisis events for our cohort, whereby use increased with COVID-19 cases after 14-28 days. These trends suggest increased psychological distress among healthcare workers as they typically deal with the most severe patients hospitalized for a COVID-19 infection (Tenforde et al. 2020 ). In future infectious disease outbreaks, additional counseling and mental health interventions should be offered to the frontline essential workers throughout the intensive response effort and additional support may be needed even after response efforts have ceased.

Our results revealed some positive effects in frontline essential workers, evidenced by a decrease in abuse and isolation during the early pandemic period compared to non-workers ( Figure 1 ). Our results support findings from China, which discovered a reduction in psychological stress four weeks after the COVID-19 epidemic, though results were not clinically significant (Wang et al. 2020), and results in the UK, which found essential workers were less likely to experience depressive symptoms (Lob et al. 2020).

In addition to the demanding workload, high stress, and burnout facing many frontline essential workers, gaps in available and affordable childcare likely served as an important stressor for these workers. As a result, the pandemic has greatly increased the caretaking responsibilities of families (Power 2020). Women have been particularly impacted by the additional burden of managing multiple roles: work, childcare, and household responsibilities (Kantamneni 2020 ). In our sample, females comprised over 60% of CTL-users seeking crisis counseling. While additional funds through the CARES Act were allocated to childcare facilities, funds and policies 

An important strength of our study is that we were able to leverage data from a large national mental health repository to examine patterns in crisis response among this understudied and highly vulnerable group. Our study contributes to a growing body of evidence of the mental health impacts from COVID-19 by using a new indicator of mental health distress (e.g. Our results are subject to a few limitations. We could not discern which sectors were represented by frontline essential workers and could not infer socioeconomic vulnerability in these data. For example, we could not understand the mental health effects on low-income workers with presumably fewer resources to buffer the many challenges posed by the pandemic than workers who had more resources (e.g., could take paid time off). Another important limitation is our inability to determine if frontline workers in our sample were associated with the children simultaneously connecting with CTL services over the same temporal period. Our study also included multiple statistical comparisons across different sub-groups, which may elevate the risk for type I error. Lastly, our results are limited to specific populations that engage with CTL, and therefore, our results may not be generalizable to the larger population. However, our J o u r n a l P r e -p r o o f results point to the necessity of additional longitudinal studies for examining the wide-ranging psychological impacts of the pandemic on frontline essential workers and their children.

Our study is the first to examine the acute mental health impacts of frontline essential workers and their children in the early COVID-19 pandemic period (March to July 2020). Children of workers were characterized by higher CTL volume for isolation and abuse, whereas frontline workers experienced a higher proportion of suicidal thoughts. Our results varied across demographics. For instance, over a 100% increase in self-harm conversations was observed for female, transgender, and non-binary gender types. Elevated crisis events in frontline essential workers were observed 14-28 days following peaks in local COVID-19 cases. Findings reveal that more research is needed to understand the progression of these adverse mental health trajectories in workers and their families and provide insight into how supportive mental health resources can be incorporated into digital interventions. 

Workers and Children of Frontline and Essential Workers from March 23 to July 20, 2020. *refers to significance at p-value < 0.05 **refers to significance at p-value < 0.01 ***refers to significance at p-value < 0.001, n.s., refers to not significant. Table 1 . Descriptive Statistics of CTL users who responded to survey questions on their status as frontline or essential workers or a child of frontline or essential worker from March 23 to July 20, 2020. N/A * An active rescue is an event in which the CTL supervisor was unable to de-escalate and help a texter in crisis to disconnect from the means of harm and work towards a safety plan, which involves contact with emergency services. These CTL users are at the highest risk of harming themselves. Less than 1% of crisis conversations end in an active rescue. **means the texter has suicidal thoughts, plan, means, and timeframe and is at imminent risk. 

Stress/Anxiety Self-Harm Suicidal Thoughts

COVID-19 New Cases and Deaths Per Day. Mapping Support

Prevalence of depression symptoms in US adults before and during the COVID-19 pandemic

I'm kinda stuck at home with unsupportive parents right now": LGBTQ youths' experiences with COVID-19 and the importance of online support

Coronavirus disease 2019 (COVID-19) and mental health for children and adolescents

Adolescent mental health in crisis

Google searches for suicide and suicide risk factors in the early stages of the COVID-19 pandemic

Google searches for suicide and suicide risk factors in the early stages of the COVID-19 pandemic

Generalized Estimating Equations (GEE). Encyclopedia of Statistics in Behavioral Science

The R Package geepack for Generalized Estimating Equations

Multidisciplinary research priorities for the COVID-19 pandemic: a call for action for mental health science

Loneliness, mental health, and substance use among US young adults during COVID-19

Levels of severity of depressive symptoms among at-risk groups in the UK during the COVID-19 pandemic

Do suicide rates in children and adolescents change during school closure in Japan? The acute effect of the first wave of COVID-19 pandemic on child and adolescent mental health

Changes in intimate partner violence during the early stages of the COVID-19 pandemic in the USA

Impact on mental health and perceptions of psychological care among medical and nursing staff in Wuhan during the 2019 novel coronavirus disease outbreak -A cross-sectional study

The impact of the COVID-19 pandemic on marginalized populations in the United States: A research agenda

Loneliness: A signature mental health concern in the era of COVID-19

Factors associated with mental health outcomes among health care workers exposed to coronavirus disease 2019

Factors associated with mental health outcomes among health care workers exposed to coronavirus disease 2019

Geospatial suicide clusters and emergency responses: An analysis of text messages to a crisis service

Mental Health-Related Emergency Department Visits Among Children Aged< 18 Years During the COVID-19 Pandemic-United States

sjPlot: Data Visualization for Statistics in Social Science

SARS/MERS/SARS-CoV-2 outbreaks and Burnout Syndrome among healthcare workers. An umbrella systematic review

Economic vulnerability of households with essential workers

Psychological distress and loneliness reported by US adults

Who are essential workers? Economic Policy Institute

Risk of COVID-19 among front-line health-care workers and the general community: a prospective cohort study. The Lancet Public Health

Incidence of psychiatric disorders among accompanied and unaccompanied asylumin Denmark: a nation-wide register-based cohort study

Prevalence of depression, anxiety, and insomnia among healthcare workers during the COVID-19 pandemic: A systematic review and meta-analysis

Well-being of parents and children during the COVID-19 pandemic: a national survey

The COVID-19 pandemic has increased the care burden of women and families

R: A language and environment for statistical computing. R Foundation for Statistical Computing

Mental health outcomes among front and second line health workers associated with the COVID-19 pandemic in Italy

Quasi-Experimental Evaluation of Text-based Crisis Patterns in Youth following Hurricane Florence in the Carolinas

Initial Challenges of Caregiving During COVID-19: Caregiver Burden, Mental Health, and the Parent-Child Relationship

Sexual and Gender Minority Stress Amid the COVID-19 Pandemic: Implications for LGBTQ Young Persons' Mental Health and Well-Being

Effects of COVID-19 on college students' mental health in the United States: Interview survey study

Mental health problems faced by healthcare workers due to the COVID-19 pandemic -A review

Gender and sexual identities predicting patterns of co-occurring health risks among sexual minority youth: a latent class analysis approach

COVID-19, suicide, and femicide: Rapid Research using Google search phrases

COVID-19 and Psychological Distress-Changes in Internet Searches for Mental Health Issues in New York During the Pandemic

Crisis Support-Seeking Behavior and Temperature in the United States: Is there an Association in Young Adults and Adolescents? Science of the Total Environment

Mental health and its correlates among children and adolescents during COVID-19 school closure: The importance of parent-child discussion

Symptom duration and risk factors for delayed return to usual health among outpatients with COVID-19 in a multistate health care systems network-United States

Mental health of communities during the COVID-19 pandemic

Mental health status among children in home confinement during the coronavirus disease

geepack: Yet Another Package for Generalized Estimating Equations

Disparities in outpatient visits for mental health and/or substance use disorders during the COVID surge and partial reopening in Massachusetts

Package 'tableone'. R Foundation for Statistical Computing

The authors thank the Crisis Text Line for providing data and approving aggregated outputs for research use. This work would not be possible without their support. This work was supported