Evaluating Physical Activity Associations with Body Mass Index, and Related Psychosocial Factors Among Hospital Workers: A Cross-Sectional Study

Jemima C John, The University of Texas School of Public Health


Background: Over two-thirds of the United States population is overweight or obese, and in healthcare employees, the overweight and obesity prevalence mirrors the national prevalence. In literature, behavioral modifications such as physical activity and diet have been shown to have inverse associations with body mass index (BMI). This dissertation examines the relation between hospital employees’ physical activity engagement and BMI, and the role of psychosocial factors in physical activity engagement. Methods and Setting: Cross-sectional analyses were conducted using data from the multi-site hospital study, Shape Up Houston. 920 participants provided data to the study. Physical activity measures (occupation, commute, leisure and domestic) and levels of intensity (light, moderate, and vigorous) were collected using the International Physical Activity Question (IPAQ). BMI was measured by trained data collectors, and psychosocial factors- self-efficacy, social support, and the transtheoretical model (stages of change) were collected using validated surveys. T-tests and ANCOVA assessed group differences in mean minutes of physical activity at the p<0.05 level, and mixed effects linear regression analyses were conducted to evaluate associations between physical activity and BMI, and physical activity and psychosocial factors while controlling for covariates. Results: Over 78% of study participants were overweight or obese. Time spent in physical activity was higher in younger participants, Blacks, and in participants with lower income and education level. Mixed effects models showed inverse associations with BMI after adjusting for covariates: Light leisure activity: (β=-0.0183:p=0.003), moderate leisure activity: (β=-0.0182:p=0.005), vigorous leisure activity (β=-0.0248:p=0.000), moderate commute activity (β=-0.0315:p=0.02), moderate domestic activity (β=-0.0087:p=0.018), and vigorous occupation activity (β=-0.0158:p=0.001). Total leisure MET-minutes had a significant inverse relationship with BMI (β= -0.0006:p=0.000), and total vigorous MET-minutes had a significant inverse association with BMI (β= -0.0005:p=0.000) after adjusting for covariates. Across physical activity measures, family support had stronger associations with physical activity engagement than support from friends, and self-efficacy was significantly associated with all domains of physical activity except occupation physical activity. Conclusion: Individuals who want to lower their BMI should spend more time in vigorous activities per week. Interventions should also focus on all groups of persons, irrespective of race, income, education or gender. Leisure and domestic-based physical activities should be emphasized as in these two areas, individuals have autonomy in structuring their physical activity regimen. Walking meetings, instant recess, and standing work stations are recommendations that can be used to increase job-site energy expenditure. Additionally, the promotion of strong family support networks can help sustain the effects of physical activity interventions in and outside the home.

Subject Area


Recommended Citation

John, Jemima C, "Evaluating Physical Activity Associations with Body Mass Index, and Related Psychosocial Factors Among Hospital Workers: A Cross-Sectional Study" (2017). Texas Medical Center Dissertations (via ProQuest). AAI10616816.