Background: Previous studies have reported conflicting results as to whether an association exists between sedentary time and cardiovascular disease (CVD) risk among African Americans. These studies, however, were limited by lack of consideration of sedentary behavior in leisure versus non-leisure settings. To elucidate this relation, we investigated the associations of television (TV) viewing time and occupational sitting with carotid intima-media thickness (CIMT), a subclinical atherosclerosis measure, in a community-based sample of African Americans.

Methods: We studied 3410 participants from the Jackson Heart Study, a single-site, community-based study of African Americans residing in Jackson, MS. CIMT was assessed by ultrasonography and represented mean far-wall thickness across right and left sides of the common carotid artery. TV viewing time, a measure of leisure sedentary behavior, and occupational sitting, a measure of non-leisure sedentary behavior, were assessed by questionnaire.

Results: In a multivariable regression model that included physical activity and CVD risk factors, longer TV viewing time (2–4 h/day and >4 h/day) was associated with greater CIMT (adjusted mean ± SE difference from referent [<2 h/day] of 0.009 ± 0.008 mm for 2–4 h/day, and 0.028 ± 0.009 mm for >4 h/day; P-trend =0.001). In contrast, more frequent occupational sitting (‘sometimes’ and ‘often/always’) was associated with lower CIMT (adjusted mean ± SE difference from referent [‘never/seldom’]:−0.021 ± 0.009 mm for ‘sometimes’, and−0.018 ± 0.008 mm for ‘often/always’; P-trend = 0.042).

Conclusions: Longer TV viewing time was associated with greater CIMT, while occupational sitting was associated with lower CIMT. These findings suggest the role of sedentary behaviors in the pathogenesis of CVD among African Americans may vary by whether individuals engage in leisure versus non-leisure sedentary behaviors.

Background
This study investigated the number of pedometer assessment occasions required to establish habitual physical activity in African American adults.
Methods
African American adults (mean age 59.9 ± 0.60 years; 59 % female) enrolled in the Diet and Physical Activity Substudy of the Jackson Heart Study wore Yamax pedometers during 3-day monitoring periods, assessed on two to three distinct occasions, each separated by approximately one month. The stability of pedometer measured PA was described as differences in mean steps/day across time, as intraclass correlation coefficients (ICC) by sex, age, and body mass index (BMI) category, and as percent of participants changing steps/day quartiles across time.
Results
Valid data were obtained for 270 participants on either two or three different assessment occasions. Mean steps/day were not significantly different across assessment occasions (p values > 0.456). The overall ICCs for steps/day assessed on either two or three occasions were 0.57 and 0.76, respectively. In addition, 85 % (two assessment occasions) and 76 % (three assessment occasions) of all participants remained in the same steps/day quartile or changed one quartile over time.
Conclusion
The current study shows that an overall mean steps/day estimate based on a 3-day monitoring period did not differ significantly over 4 – 6 months. The findings were robust to differences in sex, age, and BMI categories. A single 3-day monitoring period is sufficient to capture habitual physical activity in African American adults.