Thomson Jessica L, Landry Alicia S, Zoellner Jamie M, Connell Carol, Madson Michael B, Molaison Elaine Fontenot, Yadrick Kathy
USDA Agricultural Research Service, Stoneville, MS, USA
University of Southern Mississippi, Hattiesburg, MS, USA.
Health Educ Behav. 2015 Feb;42(1):84-91. doi: 10.1177/1090198114537069. Epub 2014 Jul 1.
The objective of this secondary analysis was to evaluate the utility of several participant adherence indicators for predicting changes in clinical, anthropometric, dietary, fitness, and physical activity (PA) outcomes in a lifestyle intervention, HUB City Steps, conducted in a southern, African American cohort in 2010. HUB City Steps was a 6-month, community-engaged, multicomponent, noncontrolled intervention targeting hypertension risk factors. Descriptive indicators were constructed using two participant adherence measures, education session attendance (ESA) and weekly steps/day pedometer diary submission (PDS), separately and in combination. Analyses, based on data from 269 primarily African American adult participants, included bivariate tests of association and multivariable linear regression to determine significant relationships between seven adherence indicators and health outcome changes, including clinical, anthropometric, dietary, fitness, and PA measures. ESA indicators were significantly correlated with four health outcomes: body mass index (BMI), fat mass, low-density lipoprotein (LDL), and PA (-.29 ≤ r ≤ .23, p < .05). PDS indicators were significantly correlated with PA (r = .27, p < .001). Combination ESA/PDS indicators were significantly correlated with five health outcomes: BMI, percentage body fat (%BF), fat mass, LDL, and PA (r = -.26 to .29, p < .05). Results from the multivariate models indicated that the combination ESA/PDS indicators were the most significant predictors of changes for five outcomes--%BF, fat mass, LDL diastolic blood pressure (DBP), and PA--while ESA performed best for BMI only. For DBP, a one-unit increase in the continuous-categorical ESA/PDS indicator resulted in 0.3 mm Hg decrease. Implications for assessing participant adherence in community-based, multicomponent lifestyle intervention research are discussed.
这项二次分析的目的是评估几个参与者依从性指标在预测生活方式干预“中心城市步伐”中临床、人体测量、饮食、健身和身体活动（PA）结果变化方面的效用。该干预于2010年在一个美国南部非裔队列中进行。“中心城市步伐”是一项为期6个月、社区参与、多组分、非对照的干预措施，旨在针对高血压风险因素。描述性指标分别使用两种参与者依从性测量方法构建，即教育课程出席率（ESA）和每日步数/每周计步器日记提交量（PDS），并将两者结合使用。基于269名主要为非裔美国成年参与者的数据进行分析，包括双变量关联检验和多变量线性回归，以确定七个依从性指标与健康结果变化之间的显著关系，这些健康结果变化包括临床、人体测量、饮食、健身和PA测量。ESA指标与四个健康结果显著相关：体重指数（BMI）、脂肪量、低密度脂蛋白（LDL）和PA（-.29≤r≤.23，p<.05）。PDS指标与PA显著相关（r=.27，p<.001）。ESA/PDS组合指标与五个健康结果显著相关：BMI、体脂百分比（%BF）、脂肪量、LDL和PA（r=-.26至.29，p<.05）。多变量模型的结果表明，ESA/PDS组合指标是五个结果变化的最显著预测因素——%BF、脂肪量、LDL、舒张压（DBP）和PA——而ESA仅在BMI方面表现最佳。对于DBP，连续分类的ESA/PDS指标每增加一个单位，DBP就会降低0.3毫米汞柱。本文讨论了在基于社区的多组分生活方式干预研究中评估参与者依从性的意义。
