UBC Faculty Research and Publications

Parent-child relationship of directly measured physical activity Fuemmeler, Bernard F; Anderson, Cheryl B; Mâsse, Louise C Mar 8, 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12966_2010_Article_413.pdf [ 365.88kB ]
JSON: 52383-1.0223490.json
JSON-LD: 52383-1.0223490-ld.json
RDF/XML (Pretty): 52383-1.0223490-rdf.xml
RDF/JSON: 52383-1.0223490-rdf.json
Turtle: 52383-1.0223490-turtle.txt
N-Triples: 52383-1.0223490-rdf-ntriples.txt
Original Record: 52383-1.0223490-source.json
Full Text

Full Text

RESEARCH Open AccessParent-child relationship of directly measuredphysical activityBernard F Fuemmeler1*†, Cheryl B Anderson2†, Louise C Mâsse3†AbstractBackground: Studies on parent-child correlations of physical activity have been mixed. Few studies have examinedconcurrent temporal patterns of physical activity and sedentary behaviors in parents and children using directmeasures. The purpose of this study was to examine parent-child activity correlations by gender, day of week, andtime of day, using accelerometers - a method for direct assessment of physical activity.Methods: Accelerometers were used to assess physical activity and sedentary time in 45 fathers, 45 mothers andtheir children (23 boys, 22 girls, mean age 9.9 years) over the course of 4 days (Thursday - Sunday). Participantswere instructed to wear accelerometers for 24 hours per day. Data from accelerometers were aggregated intowaking hours on weekdays and weekends (6:00 am to midnight) and weekday after-school hours (3:00 - 7:00 pm).Results: Across the 4 days, the mean minutes per day of moderate-to-vigorous physical activity (MVPA) for fatherswas 30.0 (s.d. = 17.3), for mothers was 30.1 (s.d. = 20.1) and for children was 145.47 (s.d. = 51.64). Mothers’ andfathers’ minutes of MVPA and minutes of sedentary time were positively correlated with child physical activity andsedentary time (all ps < .05, with the exception of mothers’ and children’s sedentary time on weekdays from 6 amto 12 am). Multivariate linear regression analyses resulted in significant effects between parents and children forMVPA across all time segments. For sedentary activity, significant associations were observed only between fatherand child on the weekend. Sedentary activity of parents and children were not related for other time segments.Models examining the associations of one or two parents with high levels of MVPA or sedentary time indicated adose response increase in child activity relative to parent.Conclusions: Greater parental MVPA was associated with increased child MVPA. In addition, having two parentswith higher levels of MVPA was associated with greater levels of activity in children. Sedentary time in children wasnot as strongly correlated with that of their parents. Findings lend support to the notion that to increasechildhood activity levels it may be fruitful to improve physical activity among parents.BackgroundThe high rates of obesity among children in the U.S.,and globally, are a significant public health concern[1,2]. Although the causes for obesity in society are mul-tifactorial, minimal physical activity, high levels ofsedentary time, and excess consumption of energy densefoods are lifestyle factors believed to be contributing toweight gain and risk of obesity in youth [3,4]. Reducingtime spent in sedentary activity and increasingmoderate-to-vigorous activity (MVPA) has numerousbenefits to children’s physical and psychological health,including being a promising strategy to prevent obesityin children. Yet, large percentages of children do notmeet recommended and optimal levels of regular physi-cal activity [5].Parents may exert a great degree of influence ontheir children’s physical activity through genetic influ-ence [6] and social learning [7,8]. Within the realm ofsocial learning, parents can serve as role models,encourage their children, or may instrumentally sup-port their children’s activity by taking them to eventswhere they can be active [9,10]. The extent to whichparents and their children have similar patterns ofphysical activity levels has been the subject of anincreasing body of research because such information* Correspondence: bernard.fuemmeler@duke.edu† Contributed equally1Duke University Medical Center, Department of Community and FamilyMedicine, Durham, NC, USAFull list of author information is available at the end of the articleFuemmeler et al. International Journal of Behavioral Nutrition and Physical Activity 2011, 8:17http://www.ijbnpa.org/content/8/1/17© 2011 Fuemmeler et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.would be useful to intervention development [9]. Pre-vious studies have shown that families tend to aggre-gate on activity patterns, especially in the extremes(e.g., sedentary or vigorous activity) [11-13]. Other stu-dies, however, have not supported parent-child covar-iation of leisure-time activity [5,14]. A review on thissubject by Gustafson and Rhodes (2006) concludedthat the results of extant studies on parent-child physi-cal activity (or inactivity) correlations are largely mixed[9]. A possible explanation for this equivocality is thatmany studies have relied on self-report or parent-report measures rather than more direct measures ofphysical activity, such as accelerometers [15]. A recentstudy of children in the UK, found that accelerometerderived parent-child sedentary activity was significantlycorrelated, but moderate-vigorous activity was not[15]. However, others using accelerometer derivedmeasures of physical activity have found that parents’physical activity levels predict those of their children[16]. In sum, although the use of accelerometers forassessing parent-child correlation in physical activityhas been increasing, there are very few studies usingthis more robust methodology for determining parent-child correlations of physical activity [12,15-17].There are a number of relevant correlates of children’sphysical activity engagement [9,18]. A well documentedrelationship exists between less physical activity engage-ment and older age [18]. Although among young chil-dren (2-5 years), this association appears to be thereverse [19]. In general, girls of any age have lowerlevels of physical activity than boys [18]. The relation-ship between race/ethnicity in children and physicalactivity is not always consistent. Some studies indicateless physical activity engagement among children ofracial/ethnic minority groups (especially among girls)[20-22], whereas others indicate greater physical activityor no differences [23,24]. Similarly, mixed findings havealso been reported with respect to the associationbetween socio-economic status and physical activity inchildren; however, higher maternal education and familyincome appear to be related to greater physical activityengagement, especially among older children [25,26].Adiposity and overweight status have also been shownto be inversely related to physical activity [27-29], butthis is not always found [19]. Finally, although relativelyunderstudied, lower paternal BMI has been reported tobe associated with greater physical activity; however, thiswas observed among boys who were obese [30].In addition to predictors of childhood physical activityengagement, other factors can be potentially relevant tothe parent-child physical activity correlation. Forinstance, observational studies have shown that dailypatterns of activities differ between weekends and week-days, with less sedentary and more active behavior onweekdays versus weekends among both children andadolescents [31,32] and adults [33]. More fine grainedanalyses have shown that time of day (e.g., late after-noons) may be an important factor in determining whenadolescents are more active [31,34]. Thus, it is reason-able to suspect that the parent-child physical activityrelationship could vary with respect to day of the weekor time of day. Another potential influential factor onparent-child physical activity correlations could be thegender of either the parent or child. Boys tend to receivemore parental support for physical activity than girls [9].At least one study has found that having physicallyactive parents is more strongly associated with physicalactivity among boys compared to girls [17]. Patterns ofgender-related differences have received little attentionin the extant literature [9,35].The purpose of this study was to measure activity pat-terns using accelerometers to determine the degree towhich physical activity and sedentary time correlateamong parents and children. We examined correlationsduring the weekend, weekday and late afternoon weekdaytime periods. Associations were further examined usingmultivariate linear regression models, which included anumber of potential covariates of interest. In addition, anexploratory aim examined the overall effect of havingsedentary or active parents on their children’s overallactivity level and if this varied by gender of the child.MethodsParticipantsA sub-sample of 57 parent-child triads were recruitedfrom a larger measurement validation study of familieswith children in 4th and 5th grade from 12 elementaryschools in the Southwest US. The current sample pro-vided additional data for the primary study, which devel-oped a measure of parental beliefs about child physicalactivity [36]. Participants agreed to wear accelerometers24 hrs per day for 4 consecutive days (Thursday throughSunday). Most of the data were collected in monthsfrom August until March when the weather in theSouthwest U.S. was temperate. Twelve families wereexcluded from the analysis: Data from 1 family wasexcluded due to a malfunctioning accelerometer; 3families had at least one member with less than 4 validdays of data; 5 families were excluded because only oneparent participated, and 3 families were excludedbecause a member from the “parental” dyad was not anactual parent, but a relative. Complete data on the 4consecutive days (Thursday - Sunday) for 45 familieswas available for the analyses (23 boys and 22 girls of 45parent dyads). The Baylor College of Medicine Institu-tional Review Board approved this study, and writteninformed parental consent and child assent was obtainedfor all participants.Fuemmeler et al. International Journal of Behavioral Nutrition and Physical Activity 2011, 8:17http://www.ijbnpa.org/content/8/1/17Page 2 of 9Measures and ProceduresThe MTI Actigraph accelerometer (model 7164; Manu-facturing Technologies Inc., Fort Walton Beach FL) wasused to objectively measure physical activity. The moni-tors were set to capture data at 60 second epochs. Acti-graph has been shown to provide valid and reliableestimates of physical activity in both adults [37,38] andchildren [39-41]. Actigraphs were worn on the waistabove the right hip using an elastic belt. They wereplaced on children and their parents on Wednesdayafternoon, usually at the participants’ home, andremoved the following Monday by staff. Participantswere instructed to wear the Actigraph continuously dur-ing the 24 hour day, except while bathing or swimmingfor 4 consecutive days. Fathers, mothers, and childrenwore the Actigraph for approximately 90% of the weartime (96 hr).Data ReductionA SAS program was modified slightly from its originaluse to read downloaded Actigraph data and producethe necessary outcomes [42]. The data was reduced toinclude waking hours between 6 am and 12 am. Dur-ing this 18 hour period of observation the averagemedian hours over the 4 days of observation that chil-dren wore the monitor was 17 hours (mean = 17hours, s.d. = .5 hours). The median hours that fatherswore the monitor was 17.4 hours (mean = 17, s.d. = .9hours) and the median hours that mothers wore themonitor was 17.3 hours (mean = 17 hours, s.d. = 1.2hours). The algorithm used 20 minute blocks of conse-cutive zero counts to identify the non-wear time on agiven day. A day was considered valid if the participantwore the accelerometer for at least 10 hours between6 am and 12 am. Child-specific cut-points [43] andadult cut-points [38] were used to categorize physicalactivity into minutes spent in the outcome categoriesof interest, namely Sedentary (< 1.5 METs) and MVPA(> 3 METs). Light METs (1.5 - 3 METs) were notincluded in the analyses. The decision to merge mod-erate and vigorous categories was made due to lowlevels of vigorous activity in the sample. As mentionedabove, this study focused on weekend, weekday, andlate afternoons (3 pm to 7 pm) and thus data weresegmented accordingly and mean minutes of activityper time period were calculated for each of these inter-vals. To evaluate the overall effect of parents’ activitylevel on children’s activity level, the mean minutes ofMVPA or sedentary time per day for the 18-hour day(6 am to 12 am) were also examined.Data AnalysisThe initial analyses included summary statistics ofmeans, standard deviations and ranges of minutes perhour spent sedentary and in MVPA for time segmentsof weekends, weekdays, and weekday afternoons (3 pm -7 pm). Bivariate correlations were conducted examiningthe association between the physical activity level ofmothers, fathers, and children (daughters and sons) forthe selected time segments. Initial adjusted modelsusing linear regression analyses were performed toexamine the effect of mothers’ and fathers’ sedentaryand MVPA on children’s sedentary and MVPA. Modelsincluded a number of covariates and potential confoun-ders, including maternal and paternal educational attain-ment, child age, gender, BMI (of children and parents),minority status, and accelerometer wear time. Prelimin-ary analyses showed that child age and minority statuswere related to at least one of the MVPA and sedentarytime outcomes. In general, older age was related tolower MVPA and higher sedentary time on the weekend(MVPA on the weekend = -.42, MVPA on weekday =-.37, MVPA on weekday 3 pm - 7 pm = -.34, andSedentary on weekend = .39, all ps < .05) and comparedto children in a minority ethnic group, Caucasiansattained significantly greater MVPA on the weekday(160.2 vs. 117.3) and significantly lower Sedentary activ-ity on the weekend (643.7 vs. 706.0). Thus, these vari-ables were entered as covariates in the adjusted linearregression models. Although accelerometer wear timewas fairly uniform, it was associated with sedentary timeon the weekend only (r = .51, p < .05), and thus, it wasalso included in linear regression models. There wereno significant associations between activity levels andeither mothers’ or children’s BMI. Fathers’ BMI was cor-related with children’s MVPA on the weekends (r = .33,p < .05). The educational attainment of parents (collegegraduate or higher versus not having graduated college)was unrelated to MVPA or Sedentary activity among thechildren. However, because BMI and educational attain-ment have been found to be relevant to child physicalactivity in other samples they were included in the lin-ear regression models. Finally, using the observed min-utes of MVPA or sedentary time across all days, weevaluated the combined influence of parent activitylevels (i.e., having one or two parents who are active orsedentary) on their children’s MVPA and sedentarytime. To do this, we created categorical variables fromparents’ MVPA and sedentary time based on a mediansplit. The categorical variable then became 1) havingboth parents in the low category of MVPA (or lowsedentary), 2) at least one parent in the high category ofMVPA (or high sedentary), or 3) having two parents inthe high MVPA category (or high sedentary). A 3 (bothlow, one high, both high) by 2 (child gender) Analysis ofVariance (ANOVA) was performed to evaluate the maineffect of category of exposure to one or two parents,child gender, and the interaction.Fuemmeler et al. International Journal of Behavioral Nutrition and Physical Activity 2011, 8:17http://www.ijbnpa.org/content/8/1/17Page 3 of 9ResultsCharacteristics of the sample are presented in Table 1.Participants were approximately 70% white and 30% min-ority, and represented a fairly homogenous sample ofmedium to high socio-economic status. Table 2 presentsthe means for minutes per day of MVPA and sedentary,as well as mean counts per minute, for weekends, week-days, and weekday afternoon time periods. Across the 4days, the mean minutes per day of MVPA for fathers was30.0 (s.d. = 17.3), for mothers was 30.1 (s.d. = 20.1) andchildren was 145.47 (s.d. = 51.64). The mean minutes perday of sedentary time was 769.9 (s.d. = 90.1) for fathers,739.9 (s.d. = 86.9) for mothers, and 654.4 (s.d. = 77.7) forchildren.Correlations between Parents and ChildrenBivariate correlations for sedentary and MVPA stratifiedby child and parent gender are presented in Table 3.According to Cohen (1992), correlation coefficientsof .10 are considered small, .30 considered medium, and.50 considered large [44]. Fathers’ and sons’ MVPA weresignificantly and positively correlated during the week-end and during the weekday afternoon hours (r = .43and .55, respectively). Mothers’ and sons’ MVPA werenot significantly correlated during any of the segmentedtimes. Fathers’ and daughters’ MVPA were significantlycorrelated during the weekdays (r = .42), but not duringthe after-school period. However, there was a fairlyrobust correlation between mothers’ and daughters’MVPA for all time segments (r = .67 on the weekends,.70 on weekdays, and .62 for after-school). With regardto sedentary counts, significant correlations were foundfor sons’ during the weekend and after-school periodwith both of their parents (r = .44 and .46 with mothers’and fathers’ on weekends, respectively, and r = .60 and.45 with mothers’ and fathers’ for after-school, respec-tively). Daughters’ sedentary activity was significantlycorrelated with mothers’ (r = .52) and fathers’ (r = .65)during the weekend, and with fathers’ during the week-day (r = .61). Partial correlations adjusting for minoritystatus, age, BMI (of children and parents) and educa-tional attainment of both mothers and fathers did notmarkedly affect the statistical significance for many ofthe correlations (data not shown). However, the signifi-cant correlation between fathers’ and daughters’ MVPAand sedentary activity during the weekday time periodwas reduced to non-significance, as was the significantcorrelations between both parents and sons’ sedentaryactivity on the weekend.Linear Regression AnalysesResults of the linear regression analyses represent therelative association of mothers ’ or fathers’ activity(either MVPA or sedentary) with children’s, adjustedfor minority status, age, gender, BMI (of child and par-ents), educational attainment (of both parents), andaccelerometer wear time (Table 4). Fathers ’ andmothers’ MVPA were each statistically significantlyassociated with children’s MVPA during the weekend(p = .01 and p = .02, respectively), during the after-noon hours (p < .01 and p = .01, respectively), andduring the weekday (p = .03 and p = .04, respectively).Results of the linear regression analyses for children’ssedentary time (Model 1b) indicated that fathers ’sedentary time was significantly associated with theirchildren’s during the weekend (p < .01), but not duringany of the other time segments.Effect of one or two parents active or inactiveAcross all days of monitoring, children’s MVPA andsedentary time were subjected to a 3 by 2 ANOVA; 3parental levels of MVPA (both parents have high levelsof MVPA, both parents have low levels of MVPA, andTable 1 Sample characteristics (percentages or meansand standard deviations)Parent Variables Mothers(n = 45)Fathers(n = 45)% or Mean (sd)Mean age in years (sd) 40.6 (5.6) 42.8 (6.2)Education LevelLess than High School 0 4High School or Equivalent 4 4Some College 9 13College Graduate 44 31Post Graduate Professional Degree 42 47Marital StatusNever married 7 4Married 91 93Separated/Divorced/Widowed 2 2RaceHispanic 11 9Black 4 4White 71 73Other 13 13BMI (kg/m2)†Normal 82 44Overweight 11 49Obese 7 7Child Variables Girls (n = 22) Boys (n = 23)Mean Age in years (SD) 10.6 (.63) 10.6 (.76)BMI†Normal 82 83At Risk for Overweight (> 85thpercentile)14 13Overweight (> 95th percentile) 5 4Fuemmeler et al. International Journal of Behavioral Nutrition and Physical Activity 2011, 8:17http://www.ijbnpa.org/content/8/1/17Page 4 of 9one parent with high levels while the other with lowlevels of MVPA) and 2 categories for gender. WithMVPA among the children as the dependent variable,the main effect of parental MVPA yielded an F ratio ofF (2, 44) = 105.48, p < .01, such that the children’sMVPA mean was significantly greater with two parentshaving high levels of MVPA (M = 195.83, SD = 14.38)compared to having both parents with low levels ofMVPA (M = 107.09, SD = 10.17) (Figure 1). The meanMVPA for children with only one parent having highlevels of MVPA (M = 130.73, SD = 17.62) was not sig-nificantly different from the other two parental levels ofMVPA. There was no significant main effect for genderF (1, 44) = 0.664, p > .05 and no significant interactionbetween parental levels of MVPA and gender F (1, 44) =0.219, p > .05.An ANOVA for children’s sedentary time was per-formed in a similar fashion as above. Results indicatedthat the average time children were sedentary was M =600.19, SD = 20.1 when both parents were low in seden-tary, was M = 657. 3, SD = 15.6 when one parent washigh in sedentary and the other was low, and was M =700.11, SD = 19.3 when both parents were high insedentary (Figure 1). Despite the increasing means rela-tive to increasing parental levels of sedentary time, therewas no significant main effect for parental sedentaryTable 2 Means and standard deviations for minutes per day of MVPA, sedentary and counts/minute for each timesegmentSedentary MVPA CountsMean (s.d.) Mean (s.d.) Mean (s.d.)MothersWeekend (6 am to 12 am) 746.2 (95.2) 26.3 (20.4) 292.0 (107.4)Weekday (6 am to 12 am) 733.5 (101.0) 33.8 (25.2) 330.9 (130.9)Weekday (3 pm to 7 pm) 162.0 (25.5) 7.7 (7.7) 344.7 (152.5)FathersWeekend (6 am to 12 am) 744.1 (114.3) 29.5 (18.8) 294.9 (100.2)Weekday (6 am to 12 am) 795.7 (79.3) 30.5 (23.2) 284.3 (114.8)Weekday (3 pm to 7 pm) 169.1 (31.5) 8.7 (11.4) 345.2 (229.7)Children (all)Weekend (6 am to 12 am) 651.3 (86.8) 141.8 (55.5) 491.0 (214.1)Weekday (6 am to 12 am) 657.6 (86.2) 149.1 (56.5) 475.0 (178.5)Weekday (3 pm to 7 pm) 125.5 (29.9) 50.1 (28.0) 672.7 (427.9)DaughtersWeekend (6 am to 12 am) 644.5 (93.0) 138.5 (60.1) 486.3 (240.8)Weekday (6 am to 12 am) 672.0 (77.6) 128.7 (45.5) 407.9 (113.3)Weekday (3 pm to 7 pm) 126.3 (24.1) 43.5 (22.0) 596.9 (233.8)SonsWeekend (6 am to 12 am) 657.9 (82.0) 145.0 (51.9) 495.6 (190.5)Weekday (6 am to 12 am) 643.7 (93.3) 168.7 (59.9) 539.2 (206.6)Weekday (3 pm to 7 pm) 124.8 (35.1) 56.5 (31.9) 771.0 (541.7)Table 3 Bivariate correlations between mothers’, fathers’ and children’s sedentary and MVPA levelsSedentary MVPADaughters Sons All Daughters Sons AllWeekend (6 am to 12 am)Mothers .52* .44* .43** .67** .10 .45**Fathers .65** .46* .56** .37 .43* .40**Weekday (6 am to 12 am)Mothers .30 .13 .23 .70** .09 .39**Fathers .61* .09 .31* .42* .38 .41**Weekday (3 pm to 7 pm)Mothers .19 .60** .39** .64** .13 .34*Fathers .33 .45* .39** .19 .55** .46*** p < .05, ** p < .01.Fuemmeler et al. International Journal of Behavioral Nutrition and Physical Activity 2011, 8:17http://www.ijbnpa.org/content/8/1/17Page 5 of 9time F (2, 44) = 4.14, p > .05, gender F (1, 44) = .11, p >.05, or the interaction F (2, 44) = 1.63, p > .05.DiscussionTo better understand and affect children’s physicalactivity levels, recent research on the determinants ofphysical activity has expressed the need for studies thatmore precisely investigate the varying contexts in whichparental activity is related to children’s activity [9,45].The data presented in the current study add to the lit-erature examining parent-child correlations of activityby examining gender-related differences as well asexamining patterns during relevant weekly segments:weekend, weekday, and weekday after-school hours. Thenumber of studies using accelerometer derived measuresof physical activity to assess parent-child correlations inactivity has been increasing, and this study is in linewith this small, but growing, body of research.In these data, we found that, overall, MVPA of parentsand children were significantly correlated for many ofthe observed time segments (weekend, weekday, andweekday 3 to 7 pm). These associations remained signif-icant in multivariate regression models. However, theassociation between parents’ and children’s sedentaryactivity was not as consistent. In the regression models,the only statistically significant finding was that fathers’sedentary activity was associated with children’s on theweekend. We also found that children of two highlyactive parents engaged in more MVPA than children ofparents who engaged in very little MVPA. The MVPAcorrelations stratified by gender of parent and childwere interesting in that they appear to be gender speci-fic (i.e., mothers’ were correlated with daughters’ andfathers’ with sons’), especially for the weekend andweekday after-school time segments. Greater attentionto gender specific association in future studies seemswarranted.Comparing our findings to those few existing studiesof parent-child correlations using accelerometer derivedmeasures of physical activity [12,15-17] is difficultbecause; 1) the age ranges of children differ across stu-dies, 2) some studies had data from only one parent,whereas others had data from both mothers and fathers,and 3) slightly different analytic methods have beenemployed across studies. However, in general, our find-ings comport with those that have found that parentalphysical activity is positively associated with an increasein children’s physical activity [16] and those that findthat children are more likely to be active when theirparents are active [12,17]. Notably, the one other studyusing accelerometer derived measures of physical activ-ity found parents’ sedentary activity was significantlycorrelated with their children’s sedentary activity, butMVPA was not [15].Table 4 Linear Regression Analysis for Variables Predicting Children’s Activity LevelVariable B SE B b p-value B SE B b p-value B SE B b p-valueChild MVPA Weekend (6 am to 12 am) Weekday (6 am to 12 am) Weekday (3 pm to 7 pm)Model 1aGender (male) -4.75 15.32 -0.04 0.76 21.60 15.80 0.19 0.18 0.05 6.98 0.00 0.99Fathers’ MVPA 1.03 0.38 0.35 0.01 0.76 0.33 0.31 0.03 1.30 0.29 0.53 0.00Mothers’ MVPA 0.88 0.36 0.32 0.02 0.63 0.30 0.28 0.04 1.29 0.46 0.35 0.01Child SedentaryModel 1bGender (male) 39.28 25.85 0.23 0.14 -0.49 32.00 0.00 0.99 5.88 10.46 0.10 0.58Fathers’ Sedentary 0.37 0.12 0.50 0.00 0.19 0.21 0.18 0.38 0.23 0.18 0.24 0.22Mothers’ Sedentary -0.01 0.16 -0.01 0.96 0.05 0.16 0.06 0.76 0.41 0.23 0.34 0.08b = standardized beta; Models controlled for minority status, child age, gender, BMI of parents and child, maternal and paternal education, and wear time.Figure 1 Mean minutes of MVPA and sedentary time byparental activity status.Fuemmeler et al. International Journal of Behavioral Nutrition and Physical Activity 2011, 8:17http://www.ijbnpa.org/content/8/1/17Page 6 of 9The study findings suggest several key points relevantto the ongoing efforts to better understand parent-childcorrelations in physical activity patterns. First, duringtimes when most parents could potentially have themost direct influence on child activity (i.e., after-schooland on weekends) both mothers’ and fathers’ MVPAwere positively related to their children’s. Although ourdata does not allow us to know whether parental model-ing, support, shared activities, or combinations of suchfactors were responsible for the parent-child activityaggregation found, it seems clear that time periods out-side of work or school are crucial targets for interven-tions that aim to involve parents.Second, the results add to the increasing evidenceregarding the importance of the after-school hours inyouth behavior and health. Higher levels of physicalactivity have been found in the time period after-schoolin previous studies [31,32,34], and interventions thathave targeted this time period have been successful inincreasing physical activity and decreasing overweight inchildren and adolescents [46-48]. Previous studies havealso reported distinct gender differences (boys > girls)in levels of MVPA during the after-school period thatwere not clearly replicated in our study [31,34]. Theboys in our sample had higher MVPA means during theafter-school hours than the girls, but they were not sta-tistically significantly higher (data not reported). Impor-tantly, the MVPA levels of parents’ were predictive ofchildren’s activity after-school, controlling for gender;although the correlations stratified by gender hint thatthese associations may be gender-specific. This deservesfurther study in other samples. Interestingly, in multi-variate models we did not find that parents’ sedentarytime during the after-school segment predicted that oftheir children’s.Third, our study extends findings on the importanceof having active parents as a predictor of physical activ-ity in children. Although parental modeling has beenwell accepted as a possible mechanism for parent-childaggregation of physical activity, there have been veryfew studies that have looked at the impact of one versustwo active parents using an accelerometer derived mea-sure to quantify activity [9]. Similar to findings byMoore et al. [17], we found a strong, linear relationshipbetween the number of active parents and the activitylevels of children. Children with two active parentsengaged in greater MVPA than children where bothparents were low in MVPA. We did not find any childgender effects, as Moore et al. did in their study, inwhich they found that parental activity was stronger forboys. In our study, in which the children were slightlyolder, children appeared to benefit substantially morefrom two active parents and this was true for both boysand girls. With respect to sedentary activity, children’ssedentary activity time did increase in a graded fashionwith the number of sedentary parents, but these werenot statistically significant increases in mean sedentaryactivity time. In other words, high levels of sedentarytime existed for all children, regardless of the sedentaryclassifications of the parents.Our data, and the multiple analytical approaches wehave used to interpret it, demonstrate the independenceof MVPA and sedentary time [49] in parent-child corre-lations. Biddle and others have argued that even highlyactive people spend considerable time being sedentary[50], and this may have been the case among our parti-cipants. In sum, our data lend support to the notionthat to increase childhood activity levels it may be fruit-ful to focus on improving the MVPA among the wholefamily, including both parents.As in any study, our results should be considered withrespect to the limitations. First, the sample size wassmall which may have had an impact on our ability todetect significant associations other than large effects.The sample was also fairly homogenous, especially inregard to parental education and weight status, whichlimits the degree to which our findings generalize to thebroader population. Notably, however, the sample didinclude a larger percentage of non-white participantsthan is usually present in these types of studies. Popula-tion based studies of parent-child correlations usingdirectly measured physical activity seem warranted.Another limitation is that the study design was cross-sectional. Longitudinal studies of parent-child physicalactivity correlations are needed especially in light of thewell established finding that physical activity levelsdecline with age [24,51-54]. It remains to be determinedhow parent-child correlations change through develop-ment. Another methodological factor to consider relatesthe use of accelerometers. Although we believe thataccelerometer derived measures of physical activity areuseful, their sole use can present some limitations. Ingeneral, accelerometers like the type used in this study,have been shown to be a valid method for assessingphysical activity in children [55]. However, these devicesdo not capture certain types of activity well, such ascycling, climbing stairs, or swimming [56]. Further,future studies should be designed to provide more con-textual information on what parents and children aredoing during time periods of interest (e.g., time usedata), which would increase understanding of the typesof activities in which parents and children engage and ifthey are active or inactive together. Lastly, although notnecessarily a limitation, the total amount of MVPA washigher in children than in parents. This is in line withthe noted observation of declining physical activity withincreasing age. This difference in the total amount likelyreflects the different ways in which children and adultsFuemmeler et al. International Journal of Behavioral Nutrition and Physical Activity 2011, 8:17http://www.ijbnpa.org/content/8/1/17Page 7 of 9accumulate MVPA (e.g., spontaneous play vs. structuredor planned activities). We were not able to capture thatlevel of detailed information in this study. Future studiesare needed to look at how parents and children of differ-ent ages accumulate their MVPA throughout the dayand the mechanisms that help explain the parent-childcorrelations we observed in this study. Understandingthese explanatory factors will be beneficial to interven-tions that aim to increase children’s MVPA throughincreasing parental MVPA.ConclusionsThe present study is important and unique in its con-tribution to the literature examining parent-child cor-relations in accelerometer derived measures ofphysical activity. Although it is commonly assertedthat parents have a significant influence on theirchild’s adoption of a physically active lifestyle, manyprevious studies have not used accelerometer derivedmeasures of physical activity nor have they examinedhow correlations vary with respect to time of day orday of week [9]. As a result, equivocal findings havebeen reported. The present findings on patterns ofactivity among parents and children suggest that par-ents’ MVPA is related to their children’s and suchfindings are useful for justifying family-based interven-tions. Future studies are needed to confirm our find-ings as well as extend this literature examiningparent-child correlations in physical activity.AcknowledgementsThis research and the preparation of this article was supported by a grant fromNational Cancer Institute (1K07CA124905-01A1) awarded to Dr. Bernard F.Fuemmeler as well as from grants from the Cancer Research Foundation ofAmerica, American Cancer Society (IRG-9303406), Curtis Hankamer BasicResearch Fund at Baylor College of Medicine, and the National Cancer Institute(R03-CA90185 and R01-CA98662) awarded to Dr. Cheryl B. Anderson. Children’sNutrition Research Center from the USDA/ARS under Cooperative AgreementNo. 58-6250-6001 also provided support to Dr. Anderson for preparation of thiswork. The contents of this publication do not necessarily reflect the views orpolicies of the USDA, nor does mention of trade names, commercial products,or organizations imply endorsement by the US Government.Author details1Duke University Medical Center, Department of Community and FamilyMedicine, Durham, NC, USA. 2Baylor College of Medicine, Department ofPediatrics, Children’s Nutrition Research Center, Houston, TX, USA. 3Universityof British Columbia, Department of Pediatrics and School of Population andPublic Health, Vancouver, BC, Canada.Authors’ contributionsAll authors (BFF, CBA, LCM) contributed to the design, analysis, and draftingof the manuscript. Data for this study were originally collected as part of alarger measurement validation study conducted by CBA. All analyses wereconducted by BFF in consultation with LCM and CBA. All authors read andapproved the final manuscript.Competing interestsThe authors declare that they have no competing interests. The authorshave no financial arrangement or affiliation with any products or servicesused or discussed in this paper.Received: 3 March 2010 Accepted: 8 March 2011Published: 8 March 2011References1. World Health Organization; Obesity and overweight, fact sheet. [http://www.who.int/mediacentre/factsheets/fs311/en/index.html].2. Wang G, Dietz WH: Economic burden of obesity in youths aged 6 to 17years: 1979-1999. Pediatrics 2002, 109(5):E81-81.3. Hills AP, King NA, Armstrong TP: The contribution of physical activity andsedentary behaviours to the growth and development of children andadolescents: implications for overweight and obesity. Sports Med 2007,37(6):533-545.4. Lavizzo-Mourey R: Childhood obesity: what it means for physicians. JAMA2007, 298(8):920-922.5. Pate RR, Freedson PS, Sallis JF, Taylor WC, Sirard J, Trost SG, Dowda M:Compliance with physical activity guidelines: prevalence in a populationof children and youth. Ann Epidemiol 2002, 12(5):303-308.6. Bouchard C, Malina RM: Genetics of physiological fitness and motorperformance. Exerc Sport Sci Rev 1983, 11:306-339.7. Heitzler CD, Martin SL, Duke J, Huhman M: Correlates of physical activityin a national sample of children aged 9-13 years. Prev Med 2006,42(4):254-260.8. Sallis JF, Alcaraz JE, McKenzie TL, Hovell MF, Kolody B, Nader PR: Parentalbehavior in relation to physical activity and fitness in 9-year-oldchildren. Am J Dis Child 1992, 146(11):1383-1388.9. Gustafson SL, Rhodes RE: Parental correlates of physical activity inchildren and early adolescents. Sports Med 2006, 36:79-97.10. Baranowski T: Families and health actions. In Handbook of health behaviorresearch I: personal and social determinants. Edited by: Gochman DS. NewYork: Plenum Press; 1997:179-206.11. Fogelholm M, Nuutinen O, Pasanen M, Myohanen E, Saatela T: Parent-childrelationship of physical activity patterns and obesity. Int J Obes RelatMetab Disord 1999, 23(12):1262-1268.12. Freedson PS, Evenson S: Familial aggregation in physical activity. Res QExerc Sport 1991, 62(4):384-389.13. Simonen RL, Perusse L, Rankinen T, Rice T, Rao DC, Bouchard C: Familialaggregation of physical activity levels in the Quebec Family Study. MedSci Sports Exerc 2002, 34(7):1137-1142.14. Anderssen N, Wold B, Torsheim T: Are parental health habits transmittedto their children? An eight year longitudinal study of physical activity inadolescents and their parents. J Adolesc 2006, 29(4):513-524.15. Jago R, Fox KR, Page AS, Brockman R, Thompson JL: Parent and childphysical activity and sedentary time: do active parents foster activechildren? BMC Public Health 2010, 10:194.16. Oliver M, Schofield GM, Schluter PJ: Parent influences on preschoolers’objectively assessed physical activity. J Sci Med Sport 2010, 13(4):403-409.17. Moore LL, Lombardi DA, White MJ, Campbell JL, Oliveria SA, Ellison C:Influence of parents’ physical activity levels on activity levels of youngchildren. J Pediatr 1991, 118:215-219.18. Sallis JF, Prochaska JJ, Taylor WC: A review of correlates of physicalactivity of children and adolescents. Med Sci Sport Exer 2000, 32:963-975.19. Pfeiffer KA, Dowda M, McIver KL, Pate RR: Factors Related to ObjectivelyMeasured Physical Activity in Preschool Children. Pediatr Exerc Sci 2009,21:196-208.20. Center for Disease Control and Prevention: Youth Risk Behavior Surveillance:United States, 2003. MMWR Morb Mortal Wkly Rep 2004, 53:1-96.21. Sallis JF, Nader PR, Broyles SL, Berry CC, Elder JP, McKenzie TL,Nelson JA: Correlates of physical activity at home in Mexican-American and Anglo-American preschool children. Health Psychology1993, 12:390-398.22. Bungum TJ, Vincent ML: Determinants of physical activity among femaleadolescents. Am J Prev Med 1997, 13:115-122.23. Pate RR, Trost SG, Felton GM, Ward DS, Dowda M, Saunders R: Correlatesof physical activity behavior in rural youth. Research Quarterly for Exerciseand Sport 1997, 68:241-248.24. Troiano RP, Berrigan D, Dodd KW, Mâsse LC, Tilert T, McDowell M: Physicalactivity in the United States measured by accelerometer. Med Sci SportsExerc 2008, 40:181-188.25. Ferreira I, van der Horst K, Wendel-Vos W, Kremers S, van Lenthe FJ, Brug J:Environmental correlates of physical activity in youth - a review andupdate. Obes Rev 2007, 8:129-154.Fuemmeler et al. International Journal of Behavioral Nutrition and Physical Activity 2011, 8:17http://www.ijbnpa.org/content/8/1/17Page 8 of 926. Drenowatz C, Eisenmann JC, Pfeiffer KA, Welk G, Heelan K, Gentile D,Walsh D: Influence of socio-economic status on habitual physical activityand sedentary behavior in 8- to 11-year old children. BMC Public Health2010, 10:214.27. Trost SG, Kerr LM, Ward DS, Pate RR: Physical activity and determinants ofphysical activity in obese and non-obese children. Int J Obesity 2001,25:822-829.28. Ward DS, Dowda M, Trost SG, Felton GM, Dishman RK, Pate RR: PhysicalActivity Correlates in Adolescent Girls Who Differ by Weight Status[ast].Obesity 2006, 14:97-105.29. Steele RM, van Sluijs EM, Cassidy A, Griffin SJ, Ekelund U: Targetingsedentary time or moderate- and vigorous-intensity activity:independent relations with adiposity in a population-based sample of10-y-old British children. American Journal of Clinical Nutrition 2009,90:1185-1192.30. Morgan PJ, Okely AD, Cliff DP, Jones RA, Baur LA: Correlates of objectivelymeasured physical activity in obese children. Obesity (Silver Spring) 2008,16:2634-2641.31. Jago R, Anderson CB, Baranowski T, Watson K: Adolescent patterns ofphysical activity differences by gender, day, and time of day. Am J PrevMed 2005, 28(5):447-452.32. Steele R, van Sluijs E, Sharp S, Landsbaugh J, Ekelund U, Griffin S: Aninvestigation of patterns of children’s sedentary and vigorous physicalactivity throughout the week. International Journal of Behavioral Nutritionand Physical Activity 2010, 7:88.33. Matthews CE, Ainsworth BE, Thompson RW, Bassett DR Jr: Sources ofvariance in daily physical activity levels as measured by anaccelerometer. Med Sci Sports Exerc 2002, 34(8):1376-1381.34. Tudor-Locke C, Lee SM, Morgan CF, Beighle A, Pangrazi RP: Children’spedometer-determined physical activity during the segmented schoolday. Med Sci Sports Exerc 2006, 38(10):1732-1738.35. Sisson SB, Church TS, Martin CK, Tudor-Locke C, Smith SR, Bouchard C,Earnest CP, Rankinen T, Newton RL, Katzmarzyk PT: Profiles of sedentarybehavior in children and adolescents: the US National Health andNutrition Examination Survey, 2001-2006. Int J Pediatr Obes 2009,4(4):353-359.36. Anderson CB, Hagstromer M, Yngve A: Validation of the PDPAR as anadolescent diary: effect of accelerometer cut points. Med Sci Sports Exerc2005, 37(7):1224-1230.37. Brage S, Wedderkopp N, Franks PW, Andersen LB, Froberg K:Reexamination of validity and reliability of the CSA monitor in walkingand running. Med Sci Sports Exerc 2003, 35(8):1447-1454.38. Freedson PS, Melanson E, Sirard J: Calibration of the Computer Scienceand Applications, Inc. accelerometer. Med Sci Sports Exerc 1998,30(5):777-781.39. Janz KF: Validation of the CSA accelerometer for assessing children’sphysical activity. Med Sci Sports Exerc 1994, 26(3):369-375.40. Puyau MR, Adolph AL, Vohra FA, Butte NF: Validation and calibration ofphysical activity monitors in children. Obes Res 2002, 10(3):150-157.41. Trost SG, Ward DS, Moorehead SM, Watson PD, Riner W, Burke JR: Validityof the computer science and applications (CSA) activity monitor inchildren. Med Sci Sports Exerc 1998, 30(4):629-633.42. Masse LC, Fuemmeler BF, Anderson CB, Matthews CE, Trost SG, Catellier DJ,Treuth M: Accelerometer data reduction: a comparison of four reductionalgorithms on select outcome variables. Med Sci Sports Exerc 2005, 37(11Suppl):S544-554.43. Trost SG, Saunders R, Ward DS: Determinants of physical activity inmiddle school children. Am J Health Behav 2002, 26(2):95-102.44. Cohen J: A power primer. Psychol Bull 1992, 112(1):155-159.45. Pugliese J, Tinsley B: Parental socialization of child and adolescentphysical activity: A meta-analysis. Journal of Family Psychology 2007,21:331-343.46. Ara I, Vincente-Rodriguez G, Perez-Gomez J, Jimenez-Ramirez J, Serrano-Sanchez JA, Dorado C: Influence of extracurricular sport activities onbody composition and physical fitness in boys: A 3-year longitudinalstudy. Int J Obes 2006, 30:1062-1071.47. Elkins WL, Cohen DA, Koralewicz LM: After-school activities, overweight,and obesity among inner city youth. J Adolesc Health 2004, 27:181-189.48. Weintraub DL, Tirumalai EC, Haydel F, Fujimoto M, Fulton JE, Robinson TN:Team sports for overweight children: The Stanford Sports to PreventObesity Randomized Trial (SPORT). Arch Pediatr Adolesc Med 2008,162:232-237.49. Marshall SJ, Biddle S, Gorely T, Cameron N, Murdey I: Relationshipsbetween media use, body fatness and physical activity in children andyouth: A meta-analysis. Int J Obes 2004, 28:1238-1246.50. Biddle S: Sedentary Behavior. Am J Prev Med 2007, 33:502-503.51. Caspersen CJ, Kriska AM, Dearwater SR: Physical activity epidemiology asapplied to elderly populations. Baillieres Clin Rheumatol 1994, 8(1):7-27.52. Caspersen CJ, Pereira MA, Curran KM: Changes in physical activity patternsin the United States, by sex and cross-sectional age. Med Sci Sports Exerc2000, 32(9):1601-1609.53. Telama R, Yang X: Decline of physical activity from youth to youngadulthood in Finland. Med Sci Sports Exerc 2000, 32(9):1617-1622.54. van Mechelen W, Twisk JW, Post GB, Snel J, Kemper HC: Physical activity ofyoung people: the Amsterdam Longitudinal Growth and Health Study.Med Sci Sports Exerc 2000, 32(9):1610-1616.55. De Vries SI, Hopman-Rock M, Bakker I, Van Mechelen W: Meeting the 60-min physical activity guideline: effect of operationalization. Med SciSports Exerc 2009, 41(1):81-86.56. Strath SJ, Bassett DR, Swartz AM: Comparison of MTI accelerometercutpoints for predicting time spent in physical activity. Int J Sports Med2003, 24:298-303.doi:10.1186/1479-5868-8-17Cite this article as: Fuemmeler et al.: Parent-child relationship of directlymeasured physical activity. International Journal of Behavioral Nutritionand Physical Activity 2011 8:17.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitFuemmeler et al. International Journal of Behavioral Nutrition and Physical Activity 2011, 8:17http://www.ijbnpa.org/content/8/1/17Page 9 of 9


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items