Active Workplace Study

The purpose of this study is to evaluate how a Total Worker Health intervention targeting sedentary behavior among call center employees improves the health, safety, and well-being of sedentary workers. Our primary hypothesis is that a multilevel integrated intervention that follows the Total Worker Health approach will have a stronger impact on primary outcomes (sedentary time and light physical activity at work, musculoskeletal pain, time lost due to injury or illness) compared to a control condition that matches typical worksite practices. To test our hypothesis, we will 1) tailor Total Worker Health intervention components to the call center environment, 2) determine the effects of a 6-month intervention, and 3) measure the durability of intervention effects at 12-month follow-up.

The study will use a randomized control trial design with two conditions: a control condition and a Total Worker Health intervention condition that includes organizational and individual intervention components. Study participants will include employees and supervisors from up to eight call centers. The call centers will be randomly assigned to the two conditions prior to the start of the study. During the study, we will conduct a baseline assessment; implement a six-month intervention; conduct a post-intervention assessment; and conduct a follow-up assessment at twelve months. During each assessment period, study participants in both conditions will complete a survey, physical measures, hemoglobin A1c measurement, and seven days of physical activity data collection with accelerometers. Study participants in both conditions could additionally be asked to complete a measurement for endothelial function during each assessment.

The baseline assessment will occur directly after participant enrollment and consent at each worksite. Both conditions will be given access to portable pedal stands. The intervention condition will have additional environmental alterations including hanging signs throughout the work environment that promote health and safety behaviors; and replacing high calorie and high sugar food and drink options in vending machines and break rooms with fresh fruits, vegetables, and other healthier choices.

Participants in the intervention condition will begin intervention activities. Intervention activities that participants could be asked to complete include: computer based training, goal setting and behavioral self-monitoring, scripted team-based health and safety discussions, and team competitions with pedal stands. Supervisor participants could additionally be asked to complete additional computer based training, goal setting and behavioral self-monitoring of supportive behaviors, and inter-supervisor observations. Participants in the control condition will receive no additional support beyond the provision of pedal stands.

Repeated measures will be collected at baseline, six months, and 12 months. Our primary hypotheses are that the intervention condition will produce greater reductions in sedentary behavior, increases in light physical activity, and reductions in musculoskeletal pain and sick day use than the control condition. To test these hypotheses, we will employ an intent-to-treat strategy using generalized estimating equations in order to use all available data to evaluate group differences in the magnitude of change over time in primary outcomes. The effect of interest for each outcome will be the interaction of condition X time. We will statistically control for confounders by measuring a wide range of demographic and psychological variables that could impact study outcomes. We will test for baseline differences between conditions on these variables using one-way ANOVAs (continuous variables) or chi-squares tests (categorical variables). Variables that differ significantly between groups at baseline will be included as covariates in the generalized estimating equation models of intervention effects over time.