J Med Internet Res. 2025 Apr 17;27:e67646. doi: 10.2196/67646.
ABSTRACT
BACKGROUND: The global COVID-19 pandemic’s mental health impact was primarily studied in the initial year of lockdowns but remained underexplored in subsequent years despite evolving conditions. This study aimed to address this gap by investigating how COVID-19-related factors, including nationwide COVID-19 deaths and incidence rates, influenced mental health indicators over time.
OBJECTIVE: This study aimed to examine the interplay among national COVID-19 pandemic deaths, incidence rates, stay-at-home behaviors, and mental health indicators across different income-level countries. Specifically, we assessed the mediating role of stay-at-home behaviors in the relationship between the COVID-19 pandemic deaths and mental health indicators.
METHODS: We analyzed data from 45 countries spanning March 2020 to October 2022. COVID-19-related factors included national COVID-19 pandemic deaths and incidence rates, obtained from publicly available datasets. Stay-at-home behaviors were assessed using Google Location History data, which captured residence-based cell phone activity as a proxy for mobility patterns. Mental health indicators were evaluated through Google Trends data, measuring changes in search volumes for “insomnia” and “suicide.” The interplay among these variables was assessed using mediation analysis to quantify the proportion mediated by stay-at-home behaviors in the association between COVID-19 deaths and mental health indicators.
RESULTS: In high-income countries, during the first pandemic year (March 2020 to February 2021), a higher monthly COVID-19 death count was associated with increased searches for “insomnia,” with a total effect estimate of 2.1×10-4 (95% CI 4.3×10-5 to 3.9×10-4; P=.01). Stay-at-home behaviors mediated 31.9% of this effect (95% CI 9.8% to 127.5%, P=.02). This association weakened and became nonsignificant in the second and third years (P=.25 and P=.54, respectively). For middle-income countries, a different pattern emerged regarding “suicide” searches. Higher COVID-19 death counts were linked to a decline in “suicide” searches in the first (estimate: -3.5×10-4, 95% CI -6.1×10-4 to -9.8×10-5; P=.006) and second years (P=.01). Mediation analysis indicated that this effect was not significantly explained by stay-at-home behaviors, suggesting the influence of other societal factors. In high-income countries, no significant association between COVID-19 deaths and “suicide” searches was observed in the first year (P=.86). However, a positive association emerged in the second year, approaching statistical significance (estimate: 2.2×10-4, 95% CI -9.5×10-7 to 4.2×10-4; P=.05), and became significant in the third year (estimate: 5.0×10-4, 95% CI 5.0×10-5 to 1.0×10-3; P=.03,), independent of stay-at-home behaviors.
CONCLUSIONS: Our findings highlight how the mental health impact of the pandemic varied across income groups and evolved over time. The mediating effect of stay-at-home behaviors was significant in the early phases but diminished in later stages, particularly in high-income countries. Meanwhile, middle-income countries exhibited unique patterns that suggest alternative protective factors. These insights can inform tailored mental health interventions and policy strategies in future public health crises.
PMID:40245400 | DOI:10.2196/67646
