📄 other

Temporal changes and associated factors of depressive symptoms among primary school students in Eastern China: A repeated cross-sectional study with a nested longitudinal subsample

This repeated cross-sectional study with a nested longitudinal subsample of primary school students in Eastern China reveals a significant decline in depressive symptoms from 2021 to 2023, while identifying poor parent–child relationships, academic difficulties, excessive screen time, and socioeconomic disadvantage as persistent risk factors that warrant multilevel interventions.

Original authors: Haifan Bai, Yunming Lin, Tingting Wang, Yuting Yang, Xiaoxiao Chen, Na Wang, Haijiang Lin, Zeyu Yu, Qingzhi Zeng, Jingyi Wang, Chaowei Fu

Published 2026-07-27

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Haifan Bai, Yunming Lin, Tingting Wang, Yuting Yang, Xiaoxiao Chen, Na Wang, Haijiang Lin, Zeyu Yu, Qingzhi Zeng, Jingyi Wang, Chaowei Fu

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Mood Map: Why Some Kids Feel Blue and Others Don't

Imagine your mind is like a garden. Sometimes, the flowers bloom bright and happy; other times, they might wilt or turn a little gray. This "grayness" is what scientists call depressive symptoms—a feeling of persistent sadness that can make it hard to enjoy life, play, or learn. While we often think of these feelings as something that only teenagers or adults deal with, they can actually start much earlier, even in elementary school.

To understand why a garden wilts, we have to look at the whole environment, not just the plant itself. Scientists use a concept called "ecological systems" to explain this. Think of a child's life as a series of nested boxes: the smallest box is the child (their habits and feelings), the next box is the family (how parents and kids talk and get along), and the next is the school (how much homework and pressure there is). If one box is shaky, it can rattle the others. This study dives into these boxes to see what makes the garden grow or struggle, specifically looking at how things changed over time in a rapidly developing part of China.

The Big Picture: A Garden Getting Greener

Researchers in Taizhou, a bustling city on China's eastern coast, decided to take a massive snapshot of the mental health of primary school students. They didn't just take one picture; they took five, like a time-lapse video, between May 2021 and March 2023. They surveyed over 36,000 "person-moments" (meaning they asked thousands of kids at different times) and followed a special group of 9,212 students who showed up for multiple rounds of the survey.

The results were surprisingly hopeful. The study found that the number of students showing signs of depressive symptoms dropped like a stone. In May 2021, about 12.1% of the students were feeling the blues. By March 2023, that number had plummeted to just 4.1%. The biggest drop happened right between the first and second surveys (May to October 2021).

Why the sudden improvement? The authors suggest it lines up perfectly with two major events. First, the "Double Reduction" policy, a government rule introduced in July 2021 to slash homework and ban extra tutoring, went into full effect. Second, the strict lockdowns of the pandemic were easing, allowing kids to get back to normal school routines and socialize again. It seems that when the pressure valve was released, the kids' moods lifted.

The Trouble Spots: What Still Makes the Garden Wilt?

Even though the overall garden looked greener, the researchers found that certain "weeds" still made some kids much more likely to feel sad. They used a fancy statistical tool (a Generalized Linear Mixed-Effects Model) to figure out which factors were the real troublemakers, even after accounting for the general trend of improvement.

Here are the biggest culprits they identified:

The Family Connection: This was the heavyweight champion of all factors. If a child had a poor relationship with their parents, they were 6.00 times more likely to have depressive symptoms compared to kids with good relationships. It wasn't just about whether the parents were married or divorced; it was about the quality of the daily connection. A bad relationship with a father was linked to a 6.00 times higher risk, and with a mother, it was nearly 5.74 times higher.
The Homework Struggle: Kids who said they were having study difficulties were 3.71 times more likely to feel depressed. It's not just about how long they studied, but whether they felt stuck or overwhelmed by the work.
The Screen Trap: Spending 2 hours or more a day on screens (excluding online classes) was linked to higher sadness.
The Money Factor: Kids from families who felt they were economically disadvantaged were 1.48 times more likely to struggle.

Interestingly, being an only child was actually a protective factor, making kids slightly less likely to feel depressed (OR = 0.81). Regular physical exercise was also a hero, lowering the risk.

The Twist: The Gap is Getting Wider

Here is the most critical part of the story. While the whole garden got greener, the researchers noticed something worrying about the money factor. The link between being poor and feeling sad didn't just stay the same; it got stronger over time.

The study found that the association between poor family economic status and depressive symptoms strengthened as the years went on (interaction term OR = 1.20). This suggests that while the "Double Reduction" policy and the end of lockdowns helped almost everyone, the kids from poorer families didn't benefit as much as the others. It's as if the rain helped the whole field, but the dry soil in the poorer patches still couldn't soak it up as well. This hints that economic inequality might be becoming a bigger barrier to mental health than before.

What This Means

The paper concludes that while the overall mental health of primary school students in this region is improving, we can't just sit back and celebrate. The "weeds" of bad family relationships, academic struggles, and too much screen time are still pulling kids down. Most importantly, the study suggests that without specific help for families struggling financially, the gap between the "rich" and "poor" gardens might keep widening. The solution, the authors suggest, isn't just one thing; it requires fixing the family connections, easing the school pressure, and making sure economic disadvantages don't leave kids behind.

Technical Summary: Temporal Changes and Associated Factors of Depressive Symptoms Among Primary School Students in Eastern China

Problem Statement
Childhood depression is a significant public health challenge in China, with evidence suggesting an earlier onset of symptoms. However, existing literature often aggregates data across childhood and adolescence, limiting age-specific insights for primary school students. Furthermore, much of the available evidence relies on single cross-sectional surveys, which hinders the characterization of population-level temporal trends and the examination of whether the strength of associations between risk factors and depressive symptoms varies over time. While longitudinal studies exist, they are often constrained by small sample sizes and high attrition. This study addresses these gaps by investigating the temporal trends and associated factors of depressive symptoms among primary school students in Eastern China during a period of significant educational reform (the "Double Reduction" policy) and evolving COVID-19 prevention measures.

Methodology
The study employed a repeated cross-sectional design with a nested longitudinal subsample conducted in Taizhou City, Zhejiang Province, between May 2021 and March 2023.

Data Collection: Five survey waves (T1–T5) were conducted, corresponding to consecutive school semesters. The study utilized a multistage cluster sampling method covering six districts/counties, selecting schools and classes randomly.
Sample Size: The final dataset included 36,116 person-wave observations from the repeated cross-sectional design. A nested longitudinal subsample of 9,212 unique students (contributing 28,977 person-wave observations) was constructed from students who participated in at least two waves.
Measures:
- Outcome: Depressive symptoms were assessed using the 27-item Children's Depression Inventory (CDI), with a cutoff score of ≥19 indicating symptoms.
- Predictors: Data were collected via student and parent questionnaires to minimize same-source bias. Variables included sociodemographics (age, gender, residence, family structure), behavioral factors (screen time, physical activity), academic factors (study time, study difficulties), and family factors (parent-child relationship quality, economic status, parental education, marital status).
Statistical Analysis:
- Trend Analysis: Cochran-Armitage trend tests were used to evaluate unadjusted temporal trends in prevalence.
- Association Modeling: Generalized Linear Mixed-Effects Models (GLMMs) with a binomial distribution and logit link were applied to the longitudinal subsample. These models included student- and school-level random intercepts to account for repeated observations and clustering.
- Modeling Strategy: A three-step approach was used: (1) preliminary single-covariate models; (2) multivariable main-effects models; and (3) models incorporating interaction terms between survey waves and significant predictors to test for time-varying effects.

Key Results

Temporal Trends: The prevalence of depressive symptoms declined significantly from 12.1% in May 2021 (T1) to 4.1% in March 2023 (T5) ( $P$ for trend < 0.001). The steepest decline occurred between T1 and T2 (May to October 2021), coinciding with the implementation of the "Double Reduction" policy.
Associated Factors (Multivariable Analysis):
- Parent-Child Relationships: This was the strongest predictor. Poor relationships with fathers were associated with an OR of 6.00 (95% CI: 4.27–8.44), and poor relationships with mothers with an OR of 5.74 (95% CI: 3.82–8.62), compared to good relationships.
- Academic Factors: Study difficulties showed the strongest academic association (OR = 3.71, 95% CI: 3.23–4.26). Prolonged daily study time (≥4 hours) and excessive screen time (≥2 hours) were also significant risk factors.
- Behavioral Factors: Regular physical exercise was protective (OR = 0.76, 95% CI: 0.65–0.88).
- Socioeconomic Factors: Poor family economic status (OR = 1.48, 95% CI: 1.18–1.84) and non-married parental status were associated with higher odds of symptoms.
- Demographics: Higher grade levels were associated with increased odds of symptoms, while no significant gender differences were observed.
Time-Varying Effects: Most associations remained stable across waves. However, the association between poor family economic status and depressive symptoms strengthened over time (interaction term OR = 1.20, 95% CI: 1.04–1.39), indicating that the relative disadvantage for economically vulnerable students increased during the study period.

Key Contributions

Methodological Innovation: The study successfully integrated a large-scale repeated cross-sectional design with a nested longitudinal subsample, allowing for the simultaneous analysis of population-level trends and individual-level temporal dynamics while mitigating the attrition issues common in pure longitudinal cohorts.
Contextual Evidence: It provides timely data on the mental health of primary school students during a critical period of educational policy reform ("Double Reduction") and post-pandemic recovery in China.
Granular Risk Identification: The study isolates specific, high-impact correlates, identifying parent-child relationship quality as the most potent predictor, surpassing academic and socioeconomic factors in magnitude.

Significance and Claims
The paper claims that depressive symptoms among primary school students in Eastern China declined substantially between 2021 and 2023, a trend likely influenced by the reduction in academic burden via the "Double Reduction" policy and the normalization of social routines post-pandemic.

However, the authors emphasize that despite this overall improvement, multilevel interventions remain critical. The study highlights that:

Family dynamics are central: The quality of parent-child relationships is the most significant correlate, suggesting that interventions must target family emotional functioning.
Academic stress persists: Study difficulties and prolonged study time remain strong risk factors, indicating a need for continued academic support.
Inequality is widening: The strengthening association between socioeconomic disadvantage and depressive symptoms over time suggests that children from poorer families benefited less from the general improvements in mental health. This points to a widening mental health inequality, necessitating policies that address structural disparities in child developmental environments.

The authors conclude that while the overall trajectory is positive, targeted strategies are required to support vulnerable subgroups, specifically those facing family relational challenges, academic struggles, and socioeconomic disadvantages.