Abstract：
OBJECTIVE: College students\' academic achievement has crucial implications for their future success. Students\' health may be a key determinant of academic performance, but more research is needed to understand this relationship.
METHODS: Secondary analysis of the American College Health Association-National College Health Assessment III pre-COVID-19 Spring 2020 dataset. N = 39 146 undergraduates at 75 higher education institutions (14% mean response rate, comparable with other large-scale national college health surveys).
METHODS: Self-reported grade point average (GPA) and 33 health behaviors in the categories of dietary behavior, physical activity, sedentary behavior, substance use, sexual risk behavior, violence-related behavior, mental health, and sleep behavior.
METHODS: Weighted cross-tabulations examining the association between GPA and health behaviors; multinomial logistic regressions assessing if behaviors predicted GPA, controlling for year, sex/gender, and race/ethnicity. Individual GPA categories were also compared to a D/F referent group.
RESULTS: There were gradient trends across GPA categories for A through D/F (18 behaviors) or A through C (12 behaviors) (P < .001). Each health behavior predicted GPA differences (P < .001), except heroin use (P = .052). The A GPA group was significantly different from the D/F GPA group for 27 behaviors (P < .001). In general, protective behaviors corresponded with higher GPAs and most risk behaviors were associated with lower GPAs.
CONCLUSIONS: There is a link between numerous health behaviors and academic performance. Stakeholders invested in college students\' health and academics should engage in mutually beneficial strategies to safeguard students\' current and future well-being and success.
METHODS: Secondary analysis of the American College Health Association-National College Health Assessment III pre-COVID-19 Spring 2020 dataset. N = 39 146 undergraduates at 75 higher education institutions (14% mean response rate, comparable with other large-scale national college health surveys).
METHODS: Self-reported grade point average (GPA) and 33 health behaviors in the categories of dietary behavior, physical activity, sedentary behavior, substance use, sexual risk behavior, violence-related behavior, mental health, and sleep behavior.
METHODS: Weighted cross-tabulations examining the association between GPA and health behaviors; multinomial logistic regressions assessing if behaviors predicted GPA, controlling for year, sex/gender, and race/ethnicity. Individual GPA categories were also compared to a D/F referent group.
RESULTS: There were gradient trends across GPA categories for A through D/F (18 behaviors) or A through C (12 behaviors) (P < .001). Each health behavior predicted GPA differences (P < .001), except heroin use (P = .052). The A GPA group was significantly different from the D/F GPA group for 27 behaviors (P < .001). In general, protective behaviors corresponded with higher GPAs and most risk behaviors were associated with lower GPAs.
CONCLUSIONS: There is a link between numerous health behaviors and academic performance. Stakeholders invested in college students\' health and academics should engage in mutually beneficial strategies to safeguard students\' current and future well-being and success.
摘要：
目的：大学生的学业成绩对他们未来的成功有着至关重要的影响。学生的健康可能是学习成绩的关键决定因素，但是需要更多的研究来理解这种关系。
方法：美国大学健康协会-国家大学健康评估III2020年春季COVID-19数据集的二次分析。N=75所高等教育机构的39.146名本科生(平均应答率14%，与其他大规模的全国大学健康调查相当)。
方法：饮食行为类别中的自我报告平均成绩（GPA）和33种健康行为，身体活动,久坐的行为，物质使用,性危险行为，与暴力有关的行为，心理健康,和睡眠行为。
方法：加权交叉表检查GPA与健康行为之间的关联；多项逻辑回归评估行为是否能预测GPA，控制一年，性别/性别,和种族/民族。还将各个GPA类别与D/F参考组进行了比较。
结果：从A到D/F（18种行为）或从A到C（12种行为）的GPA类别存在梯度趋势（P<.001）。每种健康行为预测GPA差异(P<.001)，除了使用海洛因（P=0.052）。AGPA组与D/FGPA组的27种行为差异有统计学意义(P<.001)。总的来说,保护行为与较高的GPA相对应，大多数风险行为与较低的GPA相关。
结论：许多健康行为与学业成绩之间存在联系。投资于大学生健康和学术的利益相关者应该参与互惠互利的策略，以保护学生当前和未来的福祉和成功。
方法：美国大学健康协会-国家大学健康评估III2020年春季COVID-19数据集的二次分析。N=75所高等教育机构的39.146名本科生(平均应答率14%，与其他大规模的全国大学健康调查相当)。
方法：饮食行为类别中的自我报告平均成绩（GPA）和33种健康行为，身体活动,久坐的行为，物质使用,性危险行为，与暴力有关的行为，心理健康,和睡眠行为。
方法：加权交叉表检查GPA与健康行为之间的关联；多项逻辑回归评估行为是否能预测GPA，控制一年，性别/性别,和种族/民族。还将各个GPA类别与D/F参考组进行了比较。
结果：从A到D/F（18种行为）或从A到C（12种行为）的GPA类别存在梯度趋势（P<.001）。每种健康行为预测GPA差异(P<.001)，除了使用海洛因（P=0.052）。AGPA组与D/FGPA组的27种行为差异有统计学意义(P<.001)。总的来说,保护行为与较高的GPA相对应，大多数风险行为与较低的GPA相关。
结论：许多健康行为与学业成绩之间存在联系。投资于大学生健康和学术的利益相关者应该参与互惠互利的策略，以保护学生当前和未来的福祉和成功。
