Relationships of physical activity and sedentary time in obese parent-child dyads: a cross-sectional study

The findings of this study support our hypothesis that there are significant relationships
between obese parents and their obese children in both sedentary behavior and MVPA,
with a stronger association on weekend days than weekdays. The weaker relationship
during weekdays makes sense when one considers that on weekdays, children spend the
majority (~60 %) of their waking hours away from their parents, at school, traveling
to and from school, or in after-school activities. In support, Thompson et al. 42] found that during the week families spend little or no time together being physically
active. The exploratory analyses also revealed that parent-child dyad relationships
for sedentary time were stronger for girls, while the parent-child dyad relationships
for MVPA were stronger for boys. Other studies also supported these results 15], 21], 42].

These findings are novel in that the entire sample consisted of obese parents and
their obese children, were mostly minority (African American) with low-income, and
were living in rural areas. In addition, accelerometry was used for simultaneous measurement
of physical activity for both the parent and the child, and the results were determined
by weekday versus weekend days. Previous studies have shown significant relationships
between parents’ and their children’s activities, including both sedentary and MVPA
8], 13], 43], but not all studies agree 11], 12], 25]. The controversy could be related to differing sample sizes, differing locations
(southeastern versus mid-western U.S. sites, urban versus rural settings), differing
parental participation (mother or father), differing obesity status, and failure to
separate weekdays from weekend days; although the literature clearly shows differences
in physical activity for these two differing portions of the week 26], 28].

The relationships between the sedentary behavior and MVPA levels of our obese parents
and their obese children were, in general, weak, accounting for only 2-12 % of the
total variance. The relationships were stronger than those of Dunton et al. 21] or Jago et al. 22]; however, Dunton et al. 21] compared only activities completed with both parent and child in proximity of each
other, while Jago and associates 22] used a sample of children older than ours. Conversely, our results are weaker than
other previous reports of normal weight children and their parents 8], 43], 44]. Our results may reflect the fact that our sample were all obese individuals living
in rural areas. From the parent’s perspective, some of the obese mothers may have
had low perceptions of their ability to exercise and felt more obligated toward family
responsibilities 22], which may have reduced their physical activity. In addition, many of the parents
in our study mentioned concerns for their children’s safety during exercise, suggesting
that for obese children, factors other than parent modeling of the behavior are important;
including enjoyment of physical activity, self-efficacy for physical activity, parental
support, and availability of home physical activity equipment 11]–13], 45]. However, accelerometry does not make it possible to determine the influence of these
factors.

We found that similarities between obese parents’ and obese children’s physical activity
behaviors were not associated with any characteristic of the parents that we explored
(age, BMI, gender, or ethnicity), but were associated with certain characteristics
of the children. Sedentary behavior correlations were strongest for girls and for
non-African American children, who also had the highest levels of sedentary behavior.
The literature on sedentary behaviors support stronger correlations between girls
and parents than boys and parents 21], 22]. In addition, older children (9-10 year) on weekend days had lower activity levels
than younger children, a fact also supported in the literature 15], 21], 22]. This could be related to older children asserting their independence or parents
providing less supervision 46]. MVPA levels of boys appeared to be most influential in driving parent-child correlations
for physical activity. The stratified correlations are particularly interesting in
view of the fact that the vast majority of parents in this study were women. One could
speculate that mothers who have sons tend to be more active than mothers who have
daughters. However, this would need to be verified in a larger sample of obese parents
and their obese children.

The weekday and weekend day patterns of activity were noteworthy. Previous studies
of adults 26] and children 28], 29], 31] all suggest that there is less activity during weekends than on weekdays and that
sedentary behavior increases on weekends. However, Trost and colleagues have suggested
that the physical activity levels of children (in grades 1-6) actually increase on
weekends 47]. Our data on the proportion of time spent in MVPA show similar percentages of time
for both types of days for both obese parents (weekdays?=?1.6 % and weekend days?=?1.2 %)
and obese children (weekdays?=?4.3 % and weekend days?=?4.2 %). However, total wear
time was less on weekend days, resulting in fewer actual minutes of MVPA on weekend days
than on weekdays for parents (weekend days?=?9 minutes and weekdays?=?14 minutes)
and for children (weekend days?=?31 minutes and weekdays?=?37 minutes) (Fig. 1). Thus, our data are consistent with the majority of previous findings 26], 28], 29], 31].

An intriguing trend was observed in mean counts per hour, or the equivalent counts
per minute (Fig. 1). As expected, parents had fewer counts per hour on weekend days (8,463) than on
weekdays (9,449) and fewer minutes of MVPA on weekend days. In contrast, the counts
per hour for children were actually a little higher on weekend days (17,160) than
on weekdays (16,571), though they had lower proportions of MVPA. This suggests that
on weekend days the children were spending fewer minutes in MVPA, but the time appeared
to be spent at a slightly higher intensity. The percentages of time spent in sedentary
behaviors did not change appreciably between weekdays and weekend days for either
children or parents, but actual minutes of sedentary time decreased on the weekend days,
probably related to less wear time and thus, fewer hours of recorded data on weekend days.
Alternatively, the declines in both MVPA and sedentary minutes for both groups could
suggest that slightly more time was spent in low intensity physical activity (LPA)
on the weekend days than on weekdays. One can estimate the proportion of time spent
in LPA by subtracting sedentary and MVPA from total time. In doing so post-hoc we noticed a little change in the proportion of time spent in LPA for both parents
(25.8 % weekend days versus 25.4 % weekdays) and children (32.8 % weekend days versus
30.7 % weekdays). These estimated small differences suggest that the results were
actually related to wear-time differences rather than an increase in LPA.

Overall, 50-80 % of the day was spent in sedentary behaviors by both parents and children,
similar to the proportion reported by other research on adults 39] and children 39], 42]. The stronger relationship seen on weekend days than on weekdays is reasonable considering
the time that children spend away from their parents and the home environment, during
weekdays. These results suggest that the influence of obese parents on sedentary behaviors
and MVPA of their obese children is greater on weekend days; thus, programs should
consider putting greater emphasis on increasing parents’ MVPA not only on weekdays
but also on weekends.

The 1-2 % of time the parents spent in MVPA was expected 26], 44]. However, we had anticipated that the parents would be more active on weekend days
versus weekdays but the reverse occurred (~14 minutes on weekdays and ~9 minutes on
the weekend days). One possible explanation could be that the majority of the sample
had low-income jobs that were somewhat physically demanding during the week, causing
them to be more fatigued on the weekend; thus, they had higher activity levels on
weekdays than on weekend days. The children participated in more MVPA than their parents,
as expected; but they also had fewer minutes of MVPA on weekend days than on weekdays;
thus, there was some parallelism between parents and their children 48]. Although the children were more active than parents, these obese children’s activity
levels fell short of meeting the guidelines of 60 minutes of MVPA per day 49] and were considerably less than those reported by some studies 18], 22], 28], 44]. These obese children are at risk for greater weight gain because of their low levels
of MVPA.

This study had several strengths. First, it is one of the few studies that used accelerometry
to objectively measure activity in both obese parents and their obese children on
the same days. Second, the sample consisted of only obese parents and their obese
children. Third, the study was one of a very few that have examined low socioeconomic
status families from rural areas. Finally, the sample was larger than in most previous
studies. There were also a few limitations. The proportion of the total sample with
complete accelerometry data for both parent and child was less than expected (56 %),
though the dyads that had complete data were similar in gender, age and BMI status
to the other participants. A concern may be that 94 % of the parents were female,
which could have biased the generalizability of the findings to all parent-child dyads.
Thus our findings may be better suited for mother-child relationships. Previous studies
have reported a similar issue 15], 21], 22], 50]. Finally, other data were not obtained on other factors, like home environment, geographical
environmental issues, and enjoyment of physical activity, which could have provided
more insight into the reasons for the low relationships, particularly since gender,
ethnicity, and BMI of the parent or child minimally influenced the relationships.
Therefore, these other factors could be considered biases in this study focusing on
baseline accelerometer data. Suggestions for future studies include the need to measure
both parents and children for a full seven days and include an assessment of the home
environment, geographical environmental issues, and parent and child enjoyment of
physical activity. This could provide additional information that could assist in
designing future studies.