Our findings indicate persistent, significant differences in mental health symptoms one and two years into the pandemic across several demographic, occupational, and educational characteristics. Specifically, lower income individuals and those unemployed at baseline reported worsening depression symptoms from their baseline scores. Essential workers, healthcare providers, and people with a history of depression experienced a similar fate. We also found education appears to have acted as a protective factor and led to improved mental health at follow up.
Just as the acute COVID-19 illness and disease severity was disproportionately experienced, the long-term social and emotional burden of the pandemic will likely be explained by differences in employment, education, housing, and other critical social determinants of health (SDOH). The characteristics of low-wage work that predispose workers to poor mental health when shelter-in-place orders and other public health measures were implemented included job loss or, in the case of essential workers, a dramatically increased workload and burnout.6 In fact, changes to work situations related to COVID mitigation strategies have been identified as the single greatest occupational source of anxiety and depression.7 Low income is a risk factor for poor mental health even outside the pandemic context, and people with previous history of depression are at greater risk for relapse.3 The economic downturn that has followed has further exacerbated social risks such as homelessness, hunger, and displacement, as many continue to lose employment and associated financial security.8
Our findings support reports of the how the global pandemic has amplified inequalities. Differences in education and income are better predictors of poor health than any other factors. Adults without a high school diploma are three times as likely as those with a college education to die before age 65.9 Education plays a significant role in health literacy skills, and is critical for understanding basic health education communications and messaging. Health related behaviors are almost entirely driven by education and income, with low education status accounting for half of all avoidable deaths among working age adults.10
This study has several limitations that should be considered. As with any survey, self-selection bias is a potential concern, and, since this was an online survey administered only in English, people with limited internet access and/or computer skills, and non-English speakers are likely under-represented. Our data on current or past mental health conditions are based on self-report and measures of depression employed symptom screening tools rather than clinical diagnoses. Lastly, household income information was collected in ranges, and so we could not calculate relative to benchmarks such as the federal poverty level, or median national household income.
Public Health Implications
Low education level, racial segregation, low social support, and poverty are all social determinants which contribute to increased death rates.11 Financial and educational assets have been identified as particularly important for reducing symptoms of persistent depression in the absence of stressors, and people with the lowest risk of depression are those with high assets.4 Since virtually all of the data in SDOH research has been observational, establishing mechanisms and pathways of causality, while accounting for bias in the limitations of the variables, is critical.12 Methodological advances are needed to elucidate the protective factors and mechanisms of resilience for withstanding adverse events and to inform public health policy and population health interventions.12 Ultimately, existing public health policy, even that created with emergency legislation, is likely inadequate to address the long term disproportionate hardship associated with low income and low wage occupations.