📄 medicine

Dynamic Interrelationships Among Physical Activity, Cognitive Function, and Depression in Older Adults: A Random-Intercept Cross- Lagged Panel Analysis

Using a Random-Intercept Cross-Lagged Panel Model on data from 2,435 older adults, this study reveals that while stable individual differences link physical activity and depression to cognitive function, within-person temporal dynamics show that depressive symptoms significantly predict subsequent cognitive decline whereas physical activity does not, suggesting depression screening is a critical priority for dementia prevention.

Original authors: Jiatao Zhang, Chenhao Yu, Dongsheng Bian, Xiao Han, Chenshu Shi, Guohong Li

Published 2026-07-23

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Jiatao Zhang, Chenhao Yu, Dongsheng Bian, Xiao Han, Chenshu Shi, Guohong Li

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

Imagine your brain as a bustling city. In this city, there are two main neighborhoods that often get into trouble together: the "Mood District" and the "Thinking District." Scientists have long known that when the Mood District gets gloomy (depression), the Thinking District often starts to lose its power (cognitive decline). They also know that keeping the city's streets active with movement (physical activity) is supposed to be good for everything. But here's the tricky part: when you look at the whole city, it's hard to tell if the gloominess causes the thinking to slow down, or if the slow thinking causes the gloominess. It's like trying to figure out if a cloudy sky causes the grass to be wet, or if wet grass makes the sky look cloudy. This is especially important for older adults, where keeping the city's thinking districts sharp is the key to preventing dementia, a condition that makes the city's map fade away. Researchers are desperate to know: if we have limited resources to fix the city, should we focus on cheering up the Mood District first, or should we force everyone to run more laps to wake up the Thinking District?

This study dives into that exact question using data from over 2,400 older adults in China, tracking them over several years. The researchers used a special statistical tool called a "Random-Intercept Cross-Lagged Panel Model," which you can think of as a super-powered magnifying glass. Most studies just look at the whole group and say, "People who run more tend to think better." But that's like saying, "People who live in sunny cities tend to be happy," without realizing that maybe the people who are naturally happy just choose to live in sunny cities. This new tool separates the "sunny city" people (stable personality traits) from the actual daily changes. It asks: "When this specific person feels sadder than usual, do they think worse later? And when this specific person moves more than usual, do they think better later?"

The results are a bit of a plot twist. When the researchers looked at how a person's own mood changed over time, they found a clear, one-way street: when an older adult felt more depressed than their usual self, their thinking skills dropped in the following years. It's as if a heavy fog rolling into the Mood District actually chokes the Thinking District. However, the story for physical activity was different. The study found that when a person moved more than their usual self, it did not predict an immediate boost in their thinking skills later on. In fact, the relationship seemed to go the other way: when a person's thinking skills were sharper than usual, they were more likely to get moving later. It seems that having a clear head helps you get off the couch, but just getting off the couch doesn't automatically clear the head in the short term.

So, what does this mean for our city? The paper suggests that for community programs trying to stop dementia, keeping an eye on depression might be the most urgent priority. If you spot a neighbor whose mood has dipped below their normal baseline, that might be the earliest warning sign that their thinking city is about to face a storm. While exercise is still great for the heart and the body, this study suggests it might not be the magic bullet for preventing memory loss on its own, at least not in the way we thought. Instead, the path might be: fix the mood, keep the thinking sharp, and then the physical activity will naturally follow. The authors are careful to say this is an observation, not a final proof of cause-and-effect, but it gives us a very strong clue about where to start our efforts.

Technical Summary: Dynamic Interrelationships Among Physical Activity, Cognitive Function, and Depression in Older Adults

Problem Statement
Depression and physical inactivity are recognized as modifiable risk factors for cognitive decline and dementia, as highlighted by the Lancet Dementia Commission. However, a critical gap exists in understanding the relative within-person predictive impacts of these factors on cognitive trajectories. Previous research has largely relied on traditional Cross-Lagged Panel Models (CLPM) or similar approaches that conflate stable between-person differences (e.g., socioeconomic status, education, genetic predisposition) with temporal within-person fluctuations. This conflation risks the "ecological fallacy," where population-level correlations are incorrectly assumed to represent causal mechanisms operating within individuals. Consequently, it remains unclear for primary care practitioners and policymakers which intervention—depression management or physical activity promotion—should take priority to effectively slow cognitive decline when resources are constrained.

Methodology
This study utilized longitudinal data from 2,435 older adults (mean age 60.68 ± 7.27 years) drawn from the China Health and Retirement Longitudinal Study (CHARLS). The analysis covered three waves of data collection spanning from 2015 to 2020.

Measures:
- Physical Activity: Assessed via a modified International Physical Activity Questionnaire (IPAQ), quantified as metabolic equivalent (MET)-minutes per week.
- Cognitive Function: Evaluated using the Telephone Interview of Cognitive Status (TICS), encompassing episodic memory and mental status domains (scores 0–31).
- Depression: Measured using the 10-item Center for Epidemiologic Studies Depression Scale (CESD-10), with scores ranging from 0 to 30.
- Covariates: Included age, gender, marital status, residence, household income, health insurance, chronic disease comorbidity, alcohol consumption, and tobacco smoking.
Statistical Approach:
The core analytical framework was the Random-Intercept Cross-Lagged Panel Model (RI-CLPM). This methodology was selected to partition time-invariant between-person heterogeneity (random intercepts) from within-person temporal fluctuations. By isolating intraindividual deviations, the model aimed to estimate true longitudinal effects net of stable confounders.
- Given unequal time intervals between waves (3 years between T1-T2; 2 years between T2-T3), equality constraints were imposed on autoregressive and cross-lagged paths across intervals, following standard practices for three-wave RI-CLPM.
- Model fit was assessed using $\chi^2/df$ , CFI, TLI, RMSEA, and SRMR.
- Analyses were conducted using R (version 4.1.0) and the lavaan package.

Key Results
The study distinguished between between-person associations (stable trait differences) and within-person temporal dynamics (changes over time).

Between-Person Level:
- Cognitive Function was significantly negatively associated with Depression ( $\beta = -0.203, p < 0.01$ ) and positively associated with Physical Activity ( $\beta = 0.135, p < 0.01$ ).
- The association between Physical Activity and Depression was negative but not statistically significant ( $\beta = -0.070$ ).
- These findings suggest that individuals with better baseline cognition tend to have fewer depressive symptoms and higher activity levels, likely reflecting stable individual characteristics.
Within-Person Level (Temporal Dynamics):
- Depression $\rightarrow$ Cognition: Depressive symptoms significantly predicted subsequent cognitive decline across time intervals ( $\beta = -0.046$ to $-0.043, p < 0.05$ ). When an individual's symptoms increased relative to their own baseline, their cognitive function subsequently declined.
- Cognition $\rightarrow$ Physical Activity: Cognitive function significantly predicted subsequent physical activity engagement ( $\beta = 0.059$ to $0.073, p < 0.05$ ). Improvements in cognition facilitated later activity participation.
- Physical Activity $\rightarrow$ Cognition/Depression: Physical activity did not significantly predict subsequent cognitive function or depression at the within-person level. The cross-lagged paths were non-significant.
- Autoregressive Effects: Cognitive function and depression showed significant stability over time, whereas physical activity did not.
Model Fit:
The RI-CLPM demonstrated excellent fit indices ( $\chi^2/df = 1.911$ , CFI = 0.995, TLI = 0.992, RMSEA = 0.015, SRMR = 0.024), supporting the validity of the specified model structure.

Significance and Claims
The primary contribution of this study is the direct comparison of depressive symptoms and physical activity as predictors of cognitive change within a framework that controls for stable individual differences.

Depression as a Priority Predictor: The findings indicate that, at the within-person level, depressive symptoms are a stronger temporal predictor of cognitive decline than physical activity. This supports the hypothesis that managing depressive symptoms may be a critical component of dementia prevention strategies.
Re-evaluating Physical Activity: While physical activity is positively associated with cognition at the population level, this association appears to be driven by stable between-person factors rather than a causal within-person effect where increased activity leads to immediate cognitive gains. Conversely, the study found a reverse causal pathway: cognitive function drives physical activity engagement. This suggests that maintaining or improving cognition may be a prerequisite for sustaining physical activity in older adults.
Methodological Advancement: By employing the RI-CLPM, the study addresses the ecological fallacy common in longitudinal aging research, providing a more nuanced understanding of how these variables interact over time within individuals.

Limitations Acknowledged by the Authors
The authors note several constraints:

Measurement Bias: Physical activity was self-reported, susceptible to recall and social desirability biases, which may attenuate observed associations.
Time Intervals: The unequal intervals between waves required equality constraints in the model, potentially averaging effects that might vary across different time scales.
Attrition: Selective attrition may have influenced results, as those with severe symptoms might have dropped out.
Causality: While RI-CLPM controls for stable confounders better than traditional CLPM, the cross-lagged effects represent temporal predictive relationships, not definitive causal effects. Unmeasured time-varying confounders may still exist.

Conclusion
The paper concludes that depression screening warrants particular attention in dementia prevention initiatives due to its robust within-person predictive power for cognitive decline. While physical activity remains vital for overall health, its specific cognitive benefits may be more effectively realized when cognitive function is maintained or improved first, potentially necessitating integrated intervention programs. The authors emphasize that these observational findings highlight the need for future intervention studies to establish causal effects.

Technical Summary: Dynamic Interrelationships Among Physical Activity, Cognitive Function, and Depression in Older Adults

More like this