Pandemic Recovery Is Not Linear: Mental Health Trajectories Among Disaster-Affected Children across COVID Recovery Waves

Background. Children exposed to cumulative disasters, including the COVID-19 pandemic and environmental disasters such as hurricanes, are at elevated risk for adverse mental health outcomes including depression, anxiety, and behavioral difficulties. Little is known, however, about how symptoms evolve over time. Most studies treat the pandemic as fixed exposure, rather than examining how its effects on children’s well-being unfold across the recovery period. This leaves gaps in knowledge about whether natural recovery occurs, and what factors influence mental health outcomes. This study addressed these gaps by examining whether child mental health outcomes differ across early, mid, and late pandemic recovery waves and identifying predictors of elevated symptoms in a multi-state sample of hurricane and pandemic-affected children and youth.

Methods. Data come from the first wave of a rolling enrollment randomized controlled trial (N = 670) conducted between spring 2023 and spring 2026 across Florida, Louisiana, and Texas, in areas previously affected by hurricanes. Participants were grouped into three cohorts reflecting pandemic recovery phase: early (2023, n = 188), mid (2024, n = 234), and late (2025-6, n = 248). Depression and anxiety symptoms (i.e., internalizing symptoms) were assessed using the RCADS-25 parent report (T ≥ 65). Behavioral difficulties were measured using the Strengths and Difficulties (SDQ) scale. Pandemic exposure was captured using the COVID-19 Exposure and Family Impact Survey. Binary logistic regression models examined predictors of elevated internalizing symptoms, and linear regression models examined behavioral difficulties, adjusting for age, gender, state, race/ethnicity, income, and pandemic exposure. Polynomial contrasts were used to test whether symptom patterns followed linear or non-linear trajectories across recovery phases.

Results. Clinically elevated RCADS symptoms were highest in early recovery (19%), declined in mid recovery (11%), and increased again in late recovery (14%). Polynomial contrasts indicated a non-linear (quadratic) trend (OR = 1.46, p = .073), suggesting that symptom change over time did not follow a steady pattern of improvement. COVID-19 exposure emerged as the most consistent predictor of both internalizing symptoms (OR = 1.14, 95% CI [1.08, 1.19], p < .001) and behavioral difficulties (B = 0.21, 95% CI [0.11, 0.32], p < .001). Age was also associated with both outcomes, with older youth showing higher levels of internalizing symptoms and behavioral difficulties. Gender differences were observed for behavioral outcomes only, with boys exhibiting higher levels of difficulty. Recovery phase, state, race/ethnicity, and income were not significant predictors of outcomes.

Conclusions and Implications. Findings suggest that child mental health does not improve uniformly over time following disaster exposure. Late recovery showed a partial rebound in symptoms, indicating uneven trajectories. Pandemic exposure severity, rather than elapsed time, was the strongest predictor of internalizing and behavioral outcomes, underscoring the need to assess and respond to children’s mental health needs regardless of recovery phase. Age and gender remained important, with older children and boys showing higher behavioral difficulties. Age-related findings underscore the importance of ongoing mental health support across the developmental spectrum, particularly for children approaching adolescence where untreated symptoms may compound over time.