Sedentary behavior, screen time and mental health of college students: a Meta-analysis / 中华流行病学杂志

Objective: To evaluate the effects of sedentary behavior/screen time on mental health of college students by Meta-analysis based on the results of literature retrieval and provide theoretical basis for the improvement of college students' mental health. Methods: The original research literatures about sedentary behavior (including screen time) and college students' mental health published as of 14 July 2022 were retrieved from PubMed, Embase, Cochrane Library, CNKI, VIP and Wanfang data. Data were extracted from the included studies and scored by one author in accordance with the proposed programme, and quality score was reviewed by another author. The literature that met the inclusion criteria was systematically reviewed and Meta-analysis was carried out by software Stata 14.2 based on the data from the literatures. Results: A total of 36 studies met the inclusion criteria, including 35 observational studies and 1 interventional study. There are 4 papers about the effects of sedentary behavior and 9 papers about the effects of screening time on depression in college students and 4 papers about the effects of sedentary behavior/screening time on anxiety in college students were used for a Meta-analysis, and the other studies were also analyzed. The Meta-analysis on the effects of sedentary behavior on depression in college students showed that there was a significant positive correlation between higher level sedentary behavior and increased risk for depression (OR=1.07,95%CI:1.05-1.10). Subgroup analysis indicated that there was no significant correlation between higher level sedentary behavior and depression (OR=1.74, 95%CI:0.93-3.25) in the unadjusted confounding factor model, but there was significance positive correlation after adjusting confounding factors (OR=2.15, 95%CI:1.18-3.90). Meta-analysis on the effects of screen time on depression in college students showed that longer screen time were significantly positively correlated with higher depression level (OR=1.03, 95%CI: 1.02-1.05). The results of subgroup analysis showed that in both unadjusted confounding factor model and adjusted confounding factor model, longer screen time was significantly positively correlated with depression (OR=1.27, 95%CI: 1.13-1.42; OR=1.45, 95%CI: 1.18-1.79) , respectively. Meta-analysis on the effects of sedentary behavior on anxiety showed that longer screen time was significantly positively correlated with increased anxiety risk (OR=1.44, 95%CI: 1.31-1.58). The results of subgroup analysis showed that in both unadjusted confounding factor model and adjusted confounding factor model, there was a significant positive correlation between longer screen time and anxiety (OR=1.47, 95%CI: 1.31-1.65; OR=1.38, 95%CI:1.17-1.62). The analysis for the literatures which were not eligible for Meta-analysis found that sedentary behavior/screen time was significantly associated with stress and other mental health in college students. Conclusions: Sedentary behavior or screen time is significantly negatively correlated with college students' mental health, in particular, resulting in depression and anxiety. These effects might be be different between weekdays and weekend days.

DEVELOPMENT OF VO_2max PREDICTION MODELS FROM 3-MINUTE WALK TEST / 体力科学

The purpose of the study was to develop new VO_2max prediction models for Japanese men using a 3-minute walk test. One hundred and twenty-seven Japanese men aged from 20 to 69 years were recruited as subjects of the present study. Maximal oxygen uptake (VO_2max) was measured with a maximal incremental test on a bicycle ergometer. The prediction models were derived using data of age, 3-minute walking distance (3MWD), and either BMI, waist circumference (WC), or %Fat. This data was cross-validated by using PRESS cross-validation procedures. 3MWD was significantly related to VO_2max (<i>r</i> = 0.54, P<0.001). The multiple correlation coefficients for the BMI, WC, and %Fat models, respectively, were 0.81, 0.82, and 0.85. The standard error of estimate (SEE) was 4.5, 4.4, and 4.1 ml·kg^-1·min^-1, respectively, for the BMI, WC, and %Fat models. All regression models demonstrated a high level of cross-validity supported by the minor shrinkage of the coefficient of determination and increment of SEE in the PRESS procedure. This study demonstrated that 3MWD was useful for predicting VO_2max accurately using VO_2max prediction models for Japanese men. The new non-exercise prediction equations derived in this study are applicable to estimating VO_2max in Japanese adult men.