*3.4. Work Status and Elevated Blood Pressure*

Table 3 presents the results of the statistical model constructed to estimate the association between work status and elevated blood pressure at baseline and the first and second years of follow-up. Based on variables associated with elevated blood pressure at the bivariate level from Table 2, the following were included in the model as potential confounders: school, gender, age, AN, BMI, and number of days physically active in the past seven days. The adjusted mean probability of having elevated blood pressure at baseline among currently working adolescents was approximately twice the probability observed for adolescents who were not currently working. The difference in adjusted mean probability between the two groups was 27% and statistically significant (*p* = 0.01). Although the differences in the adjusted mean probabilities for follow-up year 1 and follow-up year 2 were higher among adolescents who were currently working, the differences were much smaller (4–9%) and not statistically significant (*p* > 0.05).


**Table 3.** Adjusted <sup>a</sup> mean probability of high or high normal blood pressure at each time point among students who were not working and who were currently working.

<sup>a</sup> Adjusted means from generalized linear mixed models with a logit link function, model covariates include: age, gender, AN, body mass index (BMI), number of days physically active in the past seven days, school participant enrolled, work status, year of survey, and the interaction of work status with year of survey.
