Association of Post–COVID-19 Condition Symptoms and Employment Status

This survey study evaluates the association of post–COVID-19 condition symptoms with employment status.


Introduction
Symptoms of COVID-19 infection persist beyond 2 months in a subset of individuals, a phenomenon referred to as post-COVID-19 condition (PCC), also known as long COVID. 1 This syndrome has become prevalent, [2][3][4] but little is known about its association with functioning, with a 15-month follow-up study 5 describing an association with poorer quality of life and diminished function.

Key Points
Question Is post-COVID-19 condition (PCC), also known as long COVID, associated with differences in employment status that might suggest functional impairment?Findings Among 15 308 individuals with prior COVID-19 infection, those with PCC were less likely to be employed full-time and more likely to be unemployed.These differences persisted after adjustment for demographic differences between those with and without PCC.
Meaning These findings suggest that individuals with PCC are less likely to be working and to be working full time.A further question is whether particular illness features, such as neurocognitive symptoms, are associated with differential functioning, with a prior small study 6 associating these symptoms with decreased quality of life.A range of studies suggest that central nervous system changes may persist following acute infection, 7 and cognitive symptoms are common in PCC. 2 We therefore examined whether PCC, and neurocognitive symptoms in particular, are associated with differential rates of employment as a proxy for functional impairment, to better guide development of potential interventions.

Statistical Analysis
Primary analyses examined associations between PCC and lack of full employment or unemployment among those who were not retired, adjusted for age, sex, race and ethnicity, education, urbanicity (urban, suburban, or rural), and region, using multiple logistic regression with interlocking survey weights in R statistical software version 4.0 (R Project for Statistical Computing).All P values are 2-sided, with significance considered to be an uncorrected value of P < .05.

Results
The cohort included 15 308 survey respondents aged 18 to 69 years with test-confirmed COVID-19 at least 2 months before date of survey completion, of whom 2236 (14.6%) reported PCC symptoms (Table ) 2).

Discussion
Among 15 308 US adults surveyed between February 2021 and March 2022, PCC was associated with a greater likelihood of unemployment and lesser likelihood of working full time in adjusted models.Furthermore, among those with PCC, the presence of cognitive symptoms was associated with diminished likelihood of working full time, extending recent reports associating cognitive symptoms with poor quality of life among employed individuals. 6As unemployed individuals with PCC were more likely to have been employed before the pandemic, it is unlikely that employment status simply reflects a factor associated with risk for reporting PCC (ie, reverse causation), or confounding by some characteristic that contributes to unemployment more generally.

Limitations
A limitation in our analysis is the reliance on cross-sectional data, precluding determinations of causation.Moreover, the extent to which our results generalize remains to be determined, although prior work strongly suggests consistency with other approaches. 8

Conclusions
The results of this survey study underscore the importance of developing strategies to respond to PCC, and particularly the associated neurocognitive symptoms.Whether rehabilitation strategies drawn from neurology and psychiatry can help to ameliorate the impact of such symptoms merits investigation.More generally, although the true economic impact of the pandemic is difficult to estimate, these results underscore the importance of considering persistent effects of lost productivity.

Table .
2eatures of Survey Respondents With Prior COVID-19 Illness Who Did or Did Not Report Current PCC SymptomsIn this survey study, we analyzed a previously described cohort derived from 8 waves of a nonprobability-sample internet survey called the COVID States Project, conducted every 4 to 8 weeks between February 2021 and July 2022.2RespondentswereUSresidentsaged 18 years and older who provided written informed consent electronically before completing the survey online, according to a protocol approved by the institutional review board of Harvard University.We report Respondents were asked whether their acute symptoms had resolved; those who reported that they had not were then asked to complete a checklist of commonly reported symptoms.From this checklist of 25 items,2we combined 2 (memory problems and difficulty concentrating or focusing, or brain fog) for further evaluation.Respondents also self-reported sociodemographic variables, including race and ethnicity selected from a list according to US Census categories (African American, Asian, Hispanic, White, and other [Native American, Pacific Islander, other]), to allow for survey weighting to match the US adult population.For purposes of analysis and reporting, because they represented fewer than 5% of the total cohort, Native American, Pacific Islander, and other were combined into a single group.Other race includes Native American, Pacific Islander, and any other race.OR indicates odds ratio.
a Other race includes Native American, Pacific Islander, and "other" indicated on self-report race and ethnicity checklist.b Prepandemic employment status was available for only 1079 individuals without PCC and 260 with PCC; current job-seeking status was available for only 1124 individuals without PCC and 271 with PCC.Methods Figure 1.Association Between Post-COVID-19 Condition (PCC) and Likelihood of Working Full Time at Time of Survey in Logistic Regression Models Without and With Adjustment for Sociodemographic Features subset of currently unemployed individuals were asked about employment status before the pandemic (Table).Unemployed individuals with PCC were more likely to have been employed full Figure 2.Among Individuals With Post-COVID-19 Condition, Association Between Neurocognitive Symptoms and Likelihood of Working Full Time at Time of Survey in Logistic Regression Models Without and With Adjustment for Sociodemographic Features Other race includes Native American, Pacific Islander, and any other race.OR indicates odds ratio.timebefore the pandemic than those without PCC (103 of 260 respondents [39.6%] vs 306 of 1079 respondents [28.4%]; χ 2 1 = 12.5; P < .001).In survey-weighted regression models excluding retired respondents, the presence of PCC was associated with lower likelihood of working full time (odds ratio [OR], 0.71 [95% CI, 0.63-0.80];adjustedOR,0.84 [95% CI, 0.74-0.96])(Figure1)and with a higher likelihood of being unemployed (OR, 1.45 [95% CI, 1.22-1.73];adjusted OR, 1.23 [95% CI, 1.02-1.48]).We next examined whether specific symptoms of PCC were associated with likelihood of full-time employment.Among those A