Physical activity (PA) is a critical factor in maintaining mental health, particularly among college students who are vulnerable to stress, anxiety, and mood disturbances. The COVID‐19 pandemic introduced unprecedented disruptions to daily routines. The purpose of this cohort study was to examine the longitudinal PA behaviors before and during COVID‐19 using device‐based assessment in a sample of college students. The participants were a convenience sample of 1‐year college students from the University of Vermont Wellness Environment study. A daily survey was distributed to the participants every night on a study app measured three mental health outcomes of mood, anxiety, and stress for 16 weeks. Participants wore Apple Watches to monitor PA. A total of 167 participants completed at least 50% of daily surveys and had 20 days of valid Apple Watch data, which resulted in 11 387 participant‐days' of observations. Changes in average daily moderate‐to‐vigorous physical activity (MVPA) and step counts were examined week‐over‐week from an 8‐week period before COVID to an 8‐week period during COVID using cluster‐robust piecewise regression (16‐weeks total). Linear mixed models examined the association between PA and mental health outcomes, while also examining the moderating influence of COVID phase. Significantly lower MVPA was observed from the end of pre‐COVID to start of COVID by −18.2 min/day (p < 0.001) and significantly fewer steps/day was observed from end of pre‐COVID to start of COVID by −3277 steps/day (p < 0.001). An MVPA “catch‐up” effect was observed as there were small but positive week‐over‐week improvements during COVID for MVPA (b = 1.32 min/day, p < 0.001). The influence of COVID‐19 phases had a notable impact on the relationships between PA/exercise and mental health outcomes. A discernible trend emerged, indicating stronger connections during the COVID period for anxiety and stress compared to the pre‐COVID era. Interestingly, the moderating effect of COVID was opposite for mood and exercise. The COVID‐19 pandemic led to a dramatic decline in PA among college students, coinciding with a period of heightened stress and anxiety. Despite a slight recovery in PA levels during the pandemic, the strengthened association between exercise and anxiety/stress during this time underscores the vital role of PA in promoting mental health. These findings highlight the importance of implementing behavior change strategies to maintain and promote student wellbeing.

中文翻译：

---

追踪幸福感：使用生态瞬时评估对 COVID-19 阶段大学生的身体活动和心理健康进行全面分析


体育活动 （PA） 是维持心理健康的关键因素，尤其是在容易受到压力、焦虑和情绪障碍影响的大学生中。COVID-19 大流行对日常生活造成了前所未有的干扰。这项队列研究的目的是在大学生样本中使用基于设备的评估来检查 COVID-19 之前和期间的纵向 PA 行为。参与者是来自佛蒙特大学健康环境研究的 1 年制大学生的便利样本。每天晚上在研究应用程序上向参与者分发每日调查，测量了 16 周的情绪、焦虑和压力三种心理健康结果。参与者佩戴 Apple Watch 来监测 PA。共有 167 名参与者完成了至少 50% 的每日调查，并拥有 20 天的有效 Apple Watch 数据，从而产生了 11387 个参与者天的观察结果。使用整群稳健分段回归（共 16 周）检查从 COVID 前 8 周到 COVID 期间 8 周的每周平均每日中度至剧烈身体活动 （MVPA） 和步数的变化。线性混合模型研究了 PA 与心理健康结果之间的关联，同时还研究了 COVID 阶段的调节影响。从 COVID 前结束到 COVID 开始，观察到 MVPA 显著降低了 -18.2 分钟/天 （p < 0.001），从 COVID 前结束到 COVID 开始，观察到的步数/天显著减少 -3277 步/天 （p < 0.001）。观察到 MVPA“追赶”效应，因为在 COVID 期间，MVPA 的每周改善幅度虽小但积极 （b = 1.32 分钟/天，p < 0.001）。 COVID-19 阶段的影响对 PA/运动与心理健康结果之间的关系产生了显着影响。出现了一个明显的趋势，表明与 COVID 之前时代相比，COVID 时期焦虑和压力的联系更紧密。有趣的是，COVID 的调节作用对情绪和运动的作用是相反的。COVID-19 大流行导致大学生的 PA 急剧下降，恰逢压力和焦虑加剧的时期。尽管大流行期间 PA 水平略有恢复，但在此期间运动与焦虑/压力之间的关联加强凸显了 PA 在促进心理健康方面的重要作用。这些发现强调了实施行为改变策略以维护和促进学生福祉的重要性。