Abstract
The COVID-19 outbreak has clear clinical and economic impacts, but also affects behaviors e.g. through social distancing, and may increase stress and anxiety. However, while case numbers are tracked daily, we know little about the psychological effects of the outbreak on individuals in the moment. Here we examine the psychological and behavioral shifts over the initial stages of the outbreak in the United States in an observational longitudinal study. Through GPS phone data we find that homestay is increasing, while being at work dropped precipitously. Using regular real-time experiential surveys we observe an overall increase in stress and mood levels which is similar in size to the weekend vs. weekday differences. As there is a significant difference between weekday and weekend mood and stress levels, this is an important decrease in wellbeing. For some, especially those affected by job loss, the mental health impact is severe.
Introduction
While the clinical and economic tolls of COVID-19 were tracked in real-time through case counts and unemployment filings, the immediate psychological impact on individuals was harder to quantify. Most mental health studies rely on retrospective surveys, which are prone to recall bias. This study sought to fill this gap by capturing "in-the-moment" data during the critical transition period of March 2020. By leveraging smartphone sensor data and real-time surveys, researchers aimed to understand how mandated social distancing and the national state of emergency shifted not just where people spent their time, but how they felt as they went about their newly restricted daily lives.
Methods
The study employed a longitudinal observational approach using two primary digital phenotyping tools:
- Passive Sensing: Using the "Passive Data Kit" (PDK) mobile app, researchers collected real-time GPS coordinates at 15-second intervals to calculate "semantic location duration"—specifically, the percentage of time spent at "home" vs. "work".
- Ecological Momentary Assessments (EMAs): The app prompted participants three times a day to rate their mood and stress on 9-point Likert scales.
- Standardized Inventories: The PHQ-8 depression scale was administered at three-week intervals to track changes in symptom severity.
Participant Demographics
The study followed 127 U.S. adult participants recruited through a national research panel. Notably, over half of the participants had a previous diagnosis of major depressive disorder, representing a population potentially more vulnerable to psychological stressors.
Results
Behavioral and Psychological Shifts
The researchers compared data from the "pre-COVID" week (March 3–9, 2020) and the "post-mandate" week (March 24–30, 2020). There was a significant decrease in time spent at work (95% CI [-5.26, -2.12], p<0.001) and a significant increase in time spent at home (95% CI [7.07, 13.83], p<0.001).
Psychologically, stress levels increased (Cohen’s d=0.15, p=0.03) and mood worsened (Cohen’s d=0.17, p<0.01). While these population-level shifts were statistically significant, they were relatively small in magnitude. However, the most striking finding was the erosion of the "weekend boost."
The "Death of the Weekend"
In normal times, there is a statistically significant improvement in mood and a drop in stress during weekends. The study found that the pandemic’s impact on mood and stress was roughly equal in magnitude to this weekend/weekday difference. Essentially, the pandemic turned every day into a "weekday," psychologically speaking. For the average person, the pandemic made every day feel as stressful as a typical Monday.
Concentrated Impact: The Job Loss Factor
The study suggests that while the general population experienced a mild decline in well-being, severe impacts were concentrated in high-risk subgroups. Specifically, for the seven participants who reported losing their jobs, 100% showed an increase in depression severity, with a median increase of 3 points on the PHQ-8. The probability of this occurring by chance is a mere 1/128.
Discussion
The study's findings highlight that for many, the "work-from-home" shift was a trade-off of stressors rather than a total psychological collapse. However, the results are limited by sampling bias, as the cohort over-represents females and those with pre-existing depression. Additionally, a drop in survey response rates after the emergency declaration suggests that the most distressed individuals might have been less likely to participate, potentially underestimating the negative impact.
Future research will focus on monitoring this cohort to report on longer-term trends and identifying other specific at-risk groups beyond those affected by immediate unemployment.
Summary updated