In this first study to examine physical activity using objective measure, it was found that regardless of gender or age group classification, participants in the SHFS have mean pedometer values well below aggregated reference points.40
In fact, pedometer steps among male participants ranged from an average low of 3111 among men aged >70 years to a high of 5078 among men aged 18– 29 years () and among female participants in the SHFS, pedometer steps ranged from a mean of 3170 among women aged >70 years to 4833 among women aged 18–29 years. These low step counts would suggest that a large proportion of the sample is not meeting the current CDC and American College of Sports Medicine recommendations for physical activity.41
Other studies conducted among racially or ethnically diverse free-living samples with pedometer assessed physical activity have found that minority individuals tend to be less active than the suggested national recommendations. Bennett et al.42
examined pedometer step counts among multiethnic (50% African-American, 42% Hispanic) low income–housing residents aged 18 to >70 years. In this study, mean (SD) pedometer step counts ranged from 6587(4083.6) in participants aged <25 years to 3285(2873.3) in participants aged >70 years. This is in contrast to the findings of the SHFS study in which mean pedometer steps ranged from 5183.0 (95% CI=1548.7, 17,345.6) in participants aged 18–30 years to 3769.8 (95% CI=1036.6, 13710.1) in participants aged >70 years. Additionally, the Cross-Cultural Activity Participation Study38
found median daily step counts of 4783 (3009, 6987) and 4577 (3219, 6385) among 127 American Indian and 135 African American women (mean age 53.8±10.9 years), respectively. In comparison, female participants in the SHFS aged 50–60 years were found to have slightly lower median steps counts of 4568.4 (2620.0, 7066.2), compared to those reported in the Cross-Cultural Activity Participation Study. These findings confirm that the current sample of American Indian adults is at least as inactive as other minority samples.
When examining physical activity levels by gender in the SHFS, unlike in previous studies,16,43–45
no significant differences were found between men and women in age-adjusted, pedometer-determined physical activity levels. This finding is likely due to the lack of variability in physical activity levels across the entire SHFS population and the very low levels of activity among both men and women in the SHFS. In contrast, when the relationship between age and physical activity was examined categorically, it was found that physical activity declined with increasing age, which is often shown in population studies. Participants aged ≥70 years were found to accumulate approximately 2000 steps less than participants aged <30 years. These findings are similar to those of another study,42
which showed a 3000-step difference between participants aged >70 years when compared to participants aged <25 years.
In regard to BMI and physical activity, a significant line trend was noted in pedometer steps with increasing BMI in both men and women (P
= 0.05 for men, P
= 0.04 for women). These data indicate that SHFS participants with lower BMIs have higher step counts compared to those with larger BMIs. These findings are consistent with other studies that have shown decreasing levels of physical activity with increasing BMI.38,42,46
However, given the cross-sectional nature of this study, it was not possible to establish the causality of the association between BMI and daily step counts.
The Strong Heart Family Study provided the unique opportunity to examine physical activity levels in a large cohort of American Indian individuals using an objective measure, more specifically the pedometer. To date, most studies that have examined physical activity levels in American Indian populations have utilized subjective measures such as a questionnaire to assess physical activity in their population of interest. While this method of assessment is relatively reasonable in large population studies, it relies on participant recall and may not provide an adequate assessment of lower intensity, unstructured physical activities like walking and housework. By utilizing a pedometer, it was possible to eliminate some of the problems posed by the use of subjective measures and possibly obtain a truer representation of physical activity levels among American Indian adults.
However, despite the advantages of using the pedometer to capture unstructured and low-intensity physical activity in the SHFS, there are, unfortunately, limitations that need to be considered with its use as an assessment tool. First, the pedometer does not measure activities that are not ambulatory in nature such as resistance training and cycling. Additionally, many pedometers, such as the pedometer used in the current study, lack an internal clock and data storage capability; thus it was necessary to rely on the SHFS participants to accurately record their step counts from the pedometer in their 7-day activity diary. This process may have resulted in reporting errors or lack of data. Further, the pedometer used in this study is unable to discriminate between steps accumulated in walking, running, or stair climbing; therefore, it was not possible to determine intensity of activity. Finally, participant clothing or body habitus may have played a role in the accuracy of the pedometer. In order for a pedometer to accurately assess physical activity, it must be worn snug to the body and kept upright in a vertical plane, perpendicular to the ground. Although every effort was made to ensure that participants were properly instructed on how to wear the pedometer, there was no guarantee that this occurred. Therefore, if the pedometer was not worn in a correct manner, the pedometer may not have worked properly and may have resulted in an underestimation of physical activity levels for those specific individuals.
Other limitations that should be considered when interpreting these findings include the fact that the SHFS is made up of 96 large families selected across three geographic locations and not a random sample of American Indian communities. It has been suggested that physical activity levels may be similar in related individuals or those who share a related environment, therefore reducing inter-individual variation of physical activity levels. Individuals were treated as if they were completely independent of each other, and family structure and environmental influences were not taken into account when evaluating physical activity levels. The estimates of physical activity are believed to be reflective of their low levels of physical activity. However, the variability may be underestimated in this population.
In summary, this study is the first to objectively determine physical activity levels in a large sample of American Indian adults. The findings of this study suggest that based on pedometer steps, a majority of American-Indian participants in the Strong Heart Family Study are not meeting the minimum physical activity public health recommendations. Since physical activity has been shown to reduce the risk of developing many chronic diseases, efforts to increase physical activity levels in this population are warranted.