UBC Faculty Research and Publications

Home environment relationships with children’s physical activity, sedentary time, and screen time by… Tandon, Pooja S; Zhou, Chuan; Sallis, James F; Cain, Kelli L; Frank, Lawrence D; Saelens, Brian E Jul 26, 2012

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

Item Metadata


52383-12966_2012_Article_583.pdf [ 210.78kB ]
JSON: 52383-1.0223553.json
JSON-LD: 52383-1.0223553-ld.json
RDF/XML (Pretty): 52383-1.0223553-rdf.xml
RDF/JSON: 52383-1.0223553-rdf.json
Turtle: 52383-1.0223553-turtle.txt
N-Triples: 52383-1.0223553-rdf-ntriples.txt
Original Record: 52383-1.0223553-source.json
Full Text

Full Text

RESEARCH Open AccessHome environment relationships with children’sphysical activity, sedentary time, and screen timeby socioeconomic statusPooja S Tandon1,2*, Chuan Zhou1,2, James F Sallis3,4, Kelli L Cain3,4, Lawrence D Frank5 and Brian E Saelens1,2AbstractBackground: Children in households of lower socioeconomic status (SES) are more likely to be overweight/obese.We aimed to determine if home physical activity (PA) environments differed by SES and to explore homeenvironment mediators of the relation of family SES to children’s PA and sedentary behavior.Methods: Participants were 715 children aged 6 to 11 from the Neighborhood Impact on Kids (NIK) Study.Household SES was examined using highest educational attainment and income. Home environment wasmeasured by parent report on a survey. Outcomes were child’s accelerometer-measured PA and parent-reportedscreen time. Mediation analyses were conducted for home environment factors that varied by SES.Results: Children from lower income households had greater media access in their bedrooms (TV 52% vs. 14%,DVD player 39% vs. 14%, video games 21% vs. 9%) but lower access to portable play equipment (bikes 85% vs.98%, jump ropes 69% vs. 83%) compared to higher income children. Lower SES families had more restrictive rulesabout PA (2.5 vs. 2.0). Across SES, children watched TV/DVDs with parents/siblings more often than they engaged inPA with them. Parents of lower SES watched TV/DVDs with their children more often (3.1 vs. 2.5 days/week).Neither total daily and home-based MVPA nor sedentary time differed by SES. Children’s daily screen time variedfrom 1.7 hours/day in high SES to 2.4 in low SES families. Media in the bedroom was related to screen time, andscreen time with parents was a mediator of the SES--screen time relationship.Conclusions: Lower SES home environments provided more opportunities for sedentary behavior and fewer forPA. Removing electronic media from children’s bedrooms has the potential to reduce disparities in chronic diseaserisk.Keywords: Social epidemiology, Childhood obesity, Children, Ecological models, FamilyBackgroundLower socioeconomic status (SES) has been consistentlyassociated with poorer health in childhood [1,2]. Child-hood socioeconomic circumstances also shape adult dis-ease risks and explain in part the origins of adult healthdisparities. As with many other aspects of health, chil-dren in lower SES households in the U.S. and otherdeveloped countries are more likely to be overweight orobese [3,4].There are likely multiple factors that mediate theSES--weight status relationship, [5] and it is importantto identify modifiable factors that would improve chil-dren’s weight-related health behaviors. Children’s healthbehaviors develop within an ecological niche, [6] withthe family environment being a critical influence. Factorssuch as access to media, parenting practices (e.g., rulesabout media), sibling influences, and family habits, maybe important influences on children’s sedentary andactive behaviors [7]. These home environment charac-teristics may be influenced by the parents’ educationalattainment or income and in turn contribute to differ-ences in children’s sedentary behavior, physical activityand ultimately, weight status.* Correspondence: pooja@uw.edu1Seattle Children’s Research Institute, M/S CW8-6, P.O. Box 5371, Seattle, WA98145-5005, USA2University of Washington, Seattle, WA, USAFull list of author information is available at the end of the article© 2012 Tandon 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.Tandon et al. International Journal of Behavioral Nutrition and Physical Activity 2012, 9:88http://www.ijbnpa.org/content/9/1/88Except for children’s screen time, which is negativelyassociated with SES, [8] family SES has generally notbeen found to be related to children’s activity levels, butfurther investigation examining parent education and in-come levels as separate measures is needed [9]. Thepresent study aimed: 1) to determine if home activityenvironments differed by parental education and incomelevels and 2) to explore the processes by which the homeactivity environment mediates the association of familySES on children’s physical activity and sedentary behav-ior. It was hypothesized that the home physical environ-ment would be less healthful in families of lower SES,and these families would have less active and more sed-entary children.MethodsParticipantsParticipants were part of the Neighborhood Impact onKids (NIK) Study, an NIH funded longitudinal, observa-tional cohort study of children aged 6 to 11 and a parentin Seattle/King County, WA and San Diego County, CA[10,11]. NIK was designed to evaluate the association ofneighborhood and home environmental factors withchildren and parent’s weight status and weight-relatedbehaviors. This study was approved by the InstitutionalReview Boards at Seattle Children’s Hospital and SanDiego State University.ProtocolParticipants were recruited September 2007 - September2008 in San Diego and November 2007- January 2009 inKing County. We attempted to contact a total of 8,616households, of which 4,975 were screened for interestand eligibility, and 944 agreed to participate. Amongfamilies agreeing to participate, 730 consented and wereenrolled. The final sample consisted of 713 child–parentpairs who completed the survey and had valid acceler-ometer data. Additional details regarding recruitmentand inclusion/exclusion criteria have been previouslypublished [10].At a home or clinic visit, parents provided consentand children provided assent. The parent completed asurvey (online or paper) that assessed, among otherthings, access to media and physical activity equipmentat home, children’s sedentary behaviors, household rulesand practices about physical activity and sedentary be-havior, and sociodemographic information. The completeNIK survey is available at: http://www.seattlechildrens.org/research/child-health-behavior-and-development/saelens-lab/measures-and-protocols. Children and parentswere instructed on having the child wear an Actigraphaccelerometer for 7 days and were provided a log forrecording when the accelerometer was worn. Study staffcalled participants several times within the week toanswer questions and encourage daily wearing of theaccelerometer.MeasuresThe highest level of reported education of the parent(s)in the household and the household income were bothused as SES indicators. The original 7 categories for edu-cational attainment (ranging from< 7th grade to com-pleted graduate/professional degree) and 11 categoriesfor income (ranging from< $10,000 to> $100,000) onthe survey were combined into 3 categories each foranalyses according to the following a priori criteria:Education- low (≤completed high school), medium(completed college), high (completed graduate degree);income - low (≤$39,000), medium ($40,000-$89,000),high (≥$90,000). The Spearman’s rank correlation be-tween household income and highest education in thehousehold was 0.39. The 2008 median family incomewas $87,903 in Seattle/King County and $74,593 in SanDiego County [12].The physical home environment was assessed usingsurvey items on the presence of electronic media in thechild’s bedroom and access to fixed and portable equip-ment in and around the home that could be used forphysical activity [13]. A Bedroom Media Score was gen-erated using 5 items from a reliable scale which asks if aTV, DVD/VCR, computer, video game system and/orhand held video game player are present in the child’sbedroom (prior test-retest reliability ICC= .51 - .96) [14].A Fixed Play Equipment Score was generated by summingyes/no items regarding presence of a basketball hoop, aswimming pool and/or a fixed swing set (prior test-retestreliability ICC= .53-.80) [14]. A Portable Play EquipmentScore was generated based on access to a bike, jump rope,sports equipment (balls, racquets) and/or roller skates(prior test-retest reliability ICC= .60 - .82) [14].Other home environment measures included the pres-ence of parental rules on outdoor play and on mediause, and parent, sibling and friend participation in sed-entary and physical activities with the child. The SafetyRules Score was the sum of “yes” responses by parentsabout whether they had the following rules: “Stay close/within sight of house/parent,” “do not go into street,”“do not ride bike on street.”(prior test-retest reliabilityICC= .61-.74) [14]. A Media Rules Score was generatedby summing “yes” responses to the following 2 rules: “noTV before homework” and “<2 hours of TV per day”(prior test-retest ICCs of .57 and .73, respectively).Screen time, sedentary time and physical activityParents reported their children’s “typical weekday time”spent watching TV/DVDS, playing video games andusing the internet/other electronic media with responseoptions of none, 15 min, 30 min, 1 hr, 2 hrs, 3 hrs, ≥4Tandon et al. International Journal of Behavioral Nutrition and Physical Activity 2012, 9:88 Page 2 of 9http://www.ijbnpa.org/content/9/1/88hours per day (prior test-tetest reliability ICC = .66,.73, .72 respectively). Responses were summed to cre-ate a parent-reported child screen time value in aver-age hours/day.Child overall physical activity and sedentary behaviorwere measured by the GT1M Actigraph accelerometer(Pensacola, FL). The Actigraph has been validated andcalibrated for use among children [15]. Accelerometerdata were collected in 30 second epochs. Participantswere asked to wear the accelerometer for seven daysduring all waking hours. Upon return, the Actigraph wasimmediately downloaded and screened for completenessand irregularities/malfunction. A valid day was definedas having at least 10 valid hours of wearing time; and avalid hour contained no more than 20 minutes of con-secutive zero counts. Data were included for childrenwith at least 3 valid days. Data were converted to min-utes engaged in sedentary behavior (≤ 100 counts perminute) and moderate-to-vigorous intensity physical ac-tivity (MVPA; ≥3 metabolic equivalents (METs)) usingFreedson age-specific cut-points with the participant’sage rounded to half a year [16]. Data were also examinedusing Evenson (4 METs) cut-points given a recent studywhich found that these gave the best classification accur-acy for all four levels of physical activity intensity andperformed well among children of all ages [17,18]. Accel-erometer data were cleaned and scored using MeterPlusversion 4.0 (Santech, Inc., www.meterplussoftware.com).As participants had been instructed to remove theaccelerometers overnight, all data files were screened fornon-zero counts between the hours of 11 pm and 6 am.In all, 93 participants were identified as having overnightactivity during valid days. Participants' accelerometer logdata were triangulated with the activity counts to deter-mine sleep hours. In 92% of cases, the log-reported sleepstart time corresponded with a significant drop in activ-ity counts (below 1000), and the reported wake time cor-responded to an increase in activity counts (above 1000)exactly or within 1/2-1 hour. In cases with the slight dis-crepancies between the log-reported times and activitycounts, we relied more on the meter data and assignedthe exact sleep and wake up times based on changes inactivity counts. In 7 cases (including 3 sleepovers), therewere discrepancies> 1 hour between the log and activitycounts in the accelerometer data. In these cases, the datawere reviewed by at least 2 individuals to arrive at anagreement about sleep hours based on activity counts(increase above 1000 or decrease below 1000). In 6 casesthere was no log available and the following criteria wereused: “asleep time” when counts dropped below 1000 forat least 3 consecutive 30 min blocks (1.5 hours) and“awake time” when counts increased above 1000 for atleast 3 consecutive 30 min blocks (1.5 hours). Estimatedsleep hours were converted to “non-wearing time” toprevent an overestimation of sedentary time due to theinclusion of overnight time.Average accelerometer wear time for the whole samplewas 5688 minutes. The differences in wear time acrossSES groups were not statistically significant. Wear times(in minutes) across income groups: Low: 5787, Med:5750, High: 5640; across educational attainment groups:Low: 5772, Med:5675, High: 5622. Forty-seven percentof children’s accelerometer wearing time was spent athome.Parents completed a place log of where their childwent while wearing the accelerometer. Place categorieswere created to assess where children were while wear-ing the accelerometer. Accelerometer data were matchedby day and time to the place log. From this, non-wear,sedentary, light, moderate, hard, and very hard acceler-ometer wear times were aggregated within the giventimeframe of each location. For the purpose of thecurrent study, the “Home” category included one singlelocation for each participant (i.e. each child had only oneaddress designated as home). If parent listed ‘front yard’or ‘backyard’ in the place log, this was also consideredhome. “Home” did not include other parent/guardians’homes or homes of relatives, friends or neighbors.AnalysisChildren’s home physical activity and sedentary environ-ments and their total and home-based activity levelswere compared across different education and house-hold income groups using chi-square test for categoricalvariables and linear regression for continuous variables.Parent’s age, marital status and ethnicity were includedas covariates in the regression models as they differedacross categories of income and education.Home environment variables, which were found tovary in a statistically significant manner across SES, wereselected for further analysis using the Sobel-Goodmantest which tests if the indirect relationship between theindependent and dependent variable through the medi-ator is significantly different from zero. Mediator ana-lyses were conducted to examine the role of media inthe bedroom, access to portable play equipment, rulesaround media and safety and parent screen time withtheir child as potential mediators in the statistically sig-nificant relationship between SES and screen time. Forthe mediation analyses, the original 11 categories of in-come and 7 categories of educational attainment wereused (instead of the tertiles) in an effort to retain themost information. All analyses were conducted usingSTATA software version 10.1.ResultsParents were predominantly white, married, motherswith approximately half working ≥15 hours/week outsideTandon et al. International Journal of Behavioral Nutrition and Physical Activity 2012, 9:88 Page 3 of 9http://www.ijbnpa.org/content/9/1/88the home. Parents in higher SES households (both edu-cation and income) were older and more were marriedcompared to lower SES households. The racial and eth-nic composition of parents across SES tertiles was simi-lar except for significantly fewer Hispanic householdswith higher educational attainment or income levels.(Table 1).A higher percentage of children in lower SES house-holds had a TV, a DVD/VCR player, and a video gamesystem in their bedrooms compared to children ofhigher SES (Table 2). There was no significant differenceby SES in whether children had a hand held video gameplayer or computer in their bedroom. Most children hadaccess to portable active play equipment at or aroundtheir homes, but this access was higher in families ofhigher SES. Fewer children had access to fixed playequipment around their home, but this did not differacross SES. Approximately half of the children acrossSES categories reported having active video games athome, with no SES-based differences.Though outdoor play rules were common across SEScategories (Table 2), “do not go into street” and “do notride bike on street” rules were significantly more com-mon in low SES families. Parents in the middle incomecategory had fewer rules regarding media use comparedto those in the lowest and highest income categories.Parents of lower SES tended to watch TV/DVDs withtheir children more often than in families of higher SES.Across SES levels, children watched TV/DVDs withtheir parents and siblings more days per week, on aver-age, than they did physical activities with them. Therewere no statistically significant differences in parentalsupport of physical activities and sports between SESgroups.Accelerometer-measured total daily MVPA, MVPA athome, total daily sedentary time and sedentary time athome did not differ significantly between levels ofhousehold education or income using 3 METs Freedsoncritera. The total daily MVPA results for the Evensencut-points were lower but also were not significantly dif-ferent across SES. Parents’ reported average daily screentime for their children varied significantly by SES, ran-ging from 1.7 in the high SES to 2.4 hours/day in thelow SES families. (Table 3).The mediation analysis found that some of the rela-tionship between SES and children’s screen time wasmediated by media in the child’s bedroom, access toportable play equipment and screen time with parents.(Table 4) For example, the estimated direct effect of in-come on screen time was -.08, so for each unit increasein income (on the 11-point scale), there was a .08 hoursdecrease in daily screen time. With media in theTable 1 Participant characteristicsCharacteristic Highest Education in Household a Household Income bN Low Med High p Low Med High pN= 165 N= 279 N=261 N= 67 N= 218 N= 428Child’s age (years), mean (SD) 9.3 (1.6) 9.1 (1.6) 9.1 (1.6) NS 9.2 (1.4) 9.1(1.6) 9.1 (1.6) NSChild’s gender, N (% Female) 81 (49%) 143 (51%) 123 (47%) NS 35 (52%) 115 (53%) 202 (47%) NSParent’s age (years), mean (SD) 39.6 (6.4) 41.4 (5.2) 42.9 (9.7) <.001 38.3 (7.1) 41.2 (6.2) 42.2 (5.2) <.001Parent’s gender, N (% Female) 146 (89%) 241 (85%) 221 (84%) NS 64 (96%) 184 (84%) 363 (86%) NSParent’s race/ethnicity, N (%)White 144 (89%) 246 (90%) 229 (89%) NS 54 (87%) 187 (86%) 382(91) NSBlack 4 (2%) 6(2%) 6(2%) 4 (6%) 6 (3%) 6 (1%)Asian/Pacific Is 7 (4%) 12 (4%) 13 (5%) 4 (6%) 13 (6%) 15 (6%)More than one 4 (2%) 5(2%) 7 (3%) 0(0%) 6 (3%) 10 (2%)Other 3 (2%) 5 (2%) 3 (1%) 0 (0%) 6 (3%) 5 (1%)Hispanic 54 (33%) 25 (9%) 14 (5%) <.001 31 (46%) 38 (17%) 25 (6%) <.001Marital status, N (%)Married 132 (80%) 264 (95%) 251 (96%) <.001 35 (52%) 195 (90%) 417 (97%) <.001Hours worked outside home per week, N (%)< 15 75 (46%) 150 (54%) 113 (44%) NS 40 (60%) 102 (47%) 197 (47%) NS15–35 39 (23%) 61 (22%) 58 (22%) 14 (21%) 53 (24%) 93 (22%)36+ 51 (31%) 68 (24%) 90 (34%) 13 (19%) 63 (29%) 133 (31%)*missing data on 0–16 participants for some variables.a Defined by highest educated adult in household.b Defined by annual household income.Tandon et al. International Journal of Behavioral Nutrition and Physical Activity 2012, 9:88 Page 4 of 9http://www.ijbnpa.org/content/9/1/88Table 2 Characteristics of Home physical activity environment by education and incomeCharacteristic Education IncomeItems in child’s bedroom Low Med High p Low Med High p(% yes) N=165 N=279 N=261 N=67 N=218 N=428TV 45% 16% 11% <.001 52% 25% 14% <.001DVD/VCR 34% 16% 12% <.001 39% 21% 14% <.001Computer 16% 12% 13% NS 10% 13% 14% NSVideo game system 23% 9% 8% <.001 21% 15% 9% .004Hand held video game player 57% 59% 52% NS 46% 55% 58% NS* MEDIA SCORE(0-5) 1.7(1.4) 1.1(1.1) 1.0 (1.1) <.001 1.7 (1.4) 1.3 (1.3) 1.1 (1.2) <.001Play equipment available at or around home (% yes)PORTABLEBike 93% 97% 98% .025 85% 95% 98% <.001Jump rope 75% 81% 83% NS 69% 78% 83% .02Sports equipment (balls/bats/etc) 94% 97% 97% NS 91% 95% 98% .002Roller skates, skateboard, scooter 82% 89% 83% NS 78% 83% 87% NS*PORTABLE EQUIP SCORE (0-4) 3.4(0.8) 3.6(0.6) 3.6(0.6) .01 3.2(1.0) 3.5(0.7) 3.7(0.6) <.001FIXEDBasketball hoop 54% 60% 60% NS 48% 56% 62% NSSwimming pool 52% 55% 47% NS 55% 46% 53% NSFixed play equipment (court, pool) 56% 64% 61% NS 55% 60% 63% NS* FIXED EQUIP SCORE (0-3) 1.6 (1.0) 1.8(1.0) 1.7(1.0 NS 1.6(1.0) 1.6(0.9) 1.8(1.0) NSACTIVE VIDEO GAMES 56% 46% 48% NS 51% 47% 50% NSParent rules enforced (% yes)SAFETY RULESStay close to house/parent 86% 81% 82% NS 90% 87% 80% .02Do not go into street 79% 68% 74% .029 84% 77% 69% .01Do not ride bike on street 69% 57% 55% .008 73% 61% 55% .02Do not cross busy streets 96% 91% 89% NS 96% 90% 92% NS*SAFETY RULES SCORE (0-4) 2.4(0.9) 2.1(1.0) 2.1(1.0) .01 2.5(0.8) 2.3(1.0) 2.0(1.0) <.001MEDIA RULESNo TV/computer before homework 83% 72% 79% .035 87% 71% 78% .02<2 hours TV/computer per day 70% 70% 76% NS 78% 63% 76% .002*MEDIA RULES SCORE (0-2) 1.5(0.6) 1.4(0.7) 1.5(0.7) NS 1.6(0.5) 1.3(0.8) 1.5(0.6) <.001Screen time with (mean days/week, SD)Siblings 3.5 (2.6) 3.5(2.6) 3.1(2.5) NS 3.3(2.7) 3.5(2.7) 3.3(2.6) NSParent/caregiver 2.9 (2.5) 2.8(2.5) 2.3(2.2) .02 3.1(2.7) 3.0(2.5) 2.5(2.2) .02How often does an adult in the household(mean days/week, SD)Watch child playing sports/PA* 3.1 (2.4) 2.8(2.2) 2.6(2.0) NS 2.9(2.5) 2.8(2.2) 2.8(2.1) NSEncourage sports/PA* 4.5(2.4) 4.6(2.3) 4.6(2.3) NS 4.3(2.6) 4.4(2.4) 4.7(2.3) NSProvide transport to sports/PA* 2.7 (2.3) 2.8(2.1) 3.0(2.0) NS 2.3(2.0) 2.8(2.2) 2.9(2.0) NSDo sports/PA with child 2.0(2.1) 2.0(1.8) 1.9(1.6) NS 2.4(2.4) 2.0(1.8) 1.9(1.7) .05*PARENT SUPPORT 3.4(1.8) 3.4(1.8) 3.4(1.7) NS 3.2(1.9) 3.3(1.8) 3.5(1.7) NSHow often do your child’s siblings/ friends dosports/PA with child (mean days/week, SD)3.2(2.4) 3.6(2.2) 3.3(2.2) NS 3.7(2.6) 3.3(2.3) 3.4(2.2) NSTandon et al. International Journal of Behavioral Nutrition and Physical Activity 2012, 9:88 Page 5 of 9http://www.ijbnpa.org/content/9/1/88bedroom as a potential mediator, the indirect effect ofincome on screen time was -.02, which is a significantmediation effect (p value< .001). The size of the total in-direct path suggests that approximately 23% of the totalassociation between income and screen time wasmediated through the amount of electronic mediapresent in the child’s bedroom. Household rules onsafety and media use were not found to be mediators inour analysis.DiscussionChildren’s home environments for physical activity andsedentary behavior varied by socioeconomic status.Children in lower SES households had significantlygreater access to electronic media devices in their bed-rooms but lower access to portable play equipment.Household rules around outdoor play were more re-strictive in lower SES households. These differenceswere found across both household income and highestlevel of educational attainment in the household. Chil-dren’s screen time was higher in low-SES householdsbut there were no SES differences in children’s overallor home-based MVPA or sedentary time.The SES disparities in screen time are similar to previ-ous studies that found inverse associations between SESand screen-based media use [19,20]. Approximately halfof the children from low SES families in this sample hada television in their bedroom and a quarter had a videoTable 3 Child’s daily screen time, sedentary time and physical activity by education and incomeOutcome Educationa IncomebLow Med High pc Low Med High pcN =165 N= 279 N= 261 N= 67 N= 218 N= 428Mean(SD) Mean(SD) Mean(SD) Mean(SD) Mean(SD)Screen time (parent-reported, hours/day) 2.4(1.4) 1.9(1.3) 1.7(1.4) <.001 2.4(1.6) 2.2(1.4) 1.7(1.3) .004Child’s total sedentary time(accelerometer-measured, min/day)397 (68) 393 (71) 397 (69) NS 394 (68) 402(69) 393 (71) NSChild’s sedentary time at home(accelerometer-measured, min/day)189 (69) 192 (72) 188 (69) NS 186 (74) 200 (76) 184 (65) NSChild’s total Evenson (4-MET) MVPA(accelerometer-measured, min/day)44 (20) 48 (23) 48 (20) NS 42 (18) 43 (21) 49 (22) NSChild’s total Freedson (3-MET) MVPA(accelerometer-measured, min/day)142 (51) 149 (57) 147 (51) NS 138 (51) 143 (56) 150 (52) NSChild’s total 3-MET MVPA at home(accelerometer-measured, min/day)61 (37) 64 (38) 61 (35) NS 62 (40) 64 (40) 61 (34) NSa Defined by highest educated adult in household.b Defined by annual household income.c Adjusted for parent’s age, marital status and ethnicity.Table 4 Potential mediation effect of various home environment variables on the relationship between SES andchildren’s screen timeIndependent variable Potential mediator Direct effectestimateaIndirect effectestimateb% of totaleffect that ismediatedcp-valuedIncome Media in bedroom -.08 -.02 23% <.001Access to portable play equipment -.09 -.01 8% .04Household rules about media -.09 -.01 7% .36Household rules about safety -.10 .01 −1% .50Screen time with parents -.09 -.01 13% .01Education Media in bedroom -.18 -.08 31% <.001Access to portable play equipment -.24 -.01 6% .05Household rules about media -.25 .01 -.1% .99Household rules about safety -.25 .01 -.6% .73Screen time with parents -.23 -.03 10% .03a Refers to the estimate of the direct effect of income or education on screen time.b Refers to the estimate of income or education on screen time through the pathway with the potential mediator.c Refers to the percentage of the total effect of income or education on screen time that is mediated by the potential mediator.d Is the p-value from the Sobel mediation test which tests the significance of the indirect effect of the potential mediator.Tandon et al. International Journal of Behavioral Nutrition and Physical Activity 2012, 9:88 Page 6 of 9http://www.ijbnpa.org/content/9/1/88game system, significantly higher than for children fromhigh SES families. This paradox between low SES andhigh access to often expensive equipment has beenexplained by findings suggesting that parents in low SESfamilies have greater concerns about their neighbor-hood’s safety [21], may lack time to supervise children intheir neighborhoods, [22] and have less access to alter-native activities, [23,24] which makes indoor screen-based entertainment an appealing alternative to outdoorplay. Extensive marketing of electronic entertainmentdevices may be another contributing factor. Further-more, higher parental SES may be related to greaterawareness of and ability to adopt screen time recom-mendations; supporting the theory that many initiativesintended to improve population health also may increasedisparities since social position determines how well onecan adopt preventive health knowledge [25]. Our screentime results (mean of 1.9 hours/day), however, werelower than national estimates that suggest children thisage are exposed to over 3–4 hours of screen time perday [20].The physical activity results in this study are consist-ent with those of a review which found that variousestimates of family SES were generally unrelated to chil-dren’s physical activity [9]. Ferreria et al. hypothesizedthat since physical activity in younger children is mostlyinformal, it may not involve much extra financial cost.As activity levels generally decline with age and asso-ciated costs for athletic participation potentially increase,perhaps such disparities in income affect physical activ-ity more in adolescents [26] and adults [27]. Of note,family support for physical activities (watching, encour-aging and providing transport to sports/physical activ-ities) did not differ by SES in this sample.Though sedentary behavior may displace some phys-ical activity, it is not simply the inverse of active behav-ior, and sedentary time is also independently associatedwith poor health outcomes [8,28]. Thus, focusing effortson modifiable factors to both decrease sedentary behav-ior and increase physical activity in high risk groups iscritical. Analyses identified some potentially modifiablefactors in the home environment that were found to me-diate the relationship between SES and screen time, acommon sedentary behavior. Media in the bedroom, ac-cess to portable play equipment and joint screen timewith parents are all potentially modifiable in interven-tions. Portable play equipment may stimulate activebehaviors that are incompatible with screen time, thoughaffordability of some equipment could be a challenge forlow income families. More marketing of play equipmentor counter-marketing to electronic entertainment tar-geted to low SES families may be required.Joint media use has been recommended so parentscan monitor their children’s television exposure, helpchildren interpret what they see, and moderate the im-pact of media exposure by reducing adverse effects andincreasing the possibility of benefit [29]. However, a pre-vious study found that co-viewing was not motivated byparental determination to mediate children’s televisionexperiences, and it occurred less often with youngerchildren who need it most [30]. That study found par-ents co-view with children when their viewing prefer-ences coincide, and co-viewing is associated withpositive parental attitudes towards television. Thus, ex-cessive parent–child joint screen time appears to be arisk factor for child screen time and is an under-studiedcorrelate of child sedentary time that could be targetedin an intervention. A better understanding of how fam-ilies spend time together and interventions that promotejoint physical activities could be helpful.Media in the bedroom, especially TV, may be the mostimportant mediator identified here because it has beenassociated with overweight, likely for several reasons, in-cluding greater screen time, [31] interference with sleep,[32,33] and increased exposure to advertising for un-healthy foods. Previous research has found that media inthe bedroom mediates the relationship between SES andBMI in adolescents [34]. Our study highlights the needto target media in the bedroom at even younger ages.There are some study limitations that warrant consid-eration. First, our screen time outcome was by parent-report, which has been shown to correlate with actualviewing time, [35] but is still subject to social-desirabilitybiases that may differ by SES. Second, we did not exam-ine school and neighborhood level factors in this study,which likely vary by SES and contribute to overall phys-ical activity and sedentary behaviors, as well as physicalactivity opportunities in the school and neighborhood.Third, given the cross-sectional data for this study, wewere unable to evaluate causality. Fourth, there are mul-tiple scoring decisions and sets of cut-points for acceler-ometer data in children (e.g. 3 METs vs. 4 METs formoderate activity), and results can change significantly ifdifferent criteria are used, making comparisons betweenstudies difficult [36]. We focused on the 3 MET cut-point for moderate intensity physical activity becausethis is the level specified in the US physical activityguidelines, [37] but we did analyses using 4 MET Even-son criteria as well for comparison. Fifth, we developed anovel method (using all the information available to us)for handling accelerometer wear time during sleep hoursin order to minimize inflation of sedentary time. How-ever this approach has not been validated and we mayhave inadvertently eliminated some wear time. Sixth, asmany complex factors influence children’s activity levels,unmeasured factors that are related to both SES and ac-tivity levels likely exist. Outdoor time, in particular, hasbeen found to be correlated with physical activity inTandon et al. International Journal of Behavioral Nutrition and Physical Activity 2012, 9:88 Page 7 of 9http://www.ijbnpa.org/content/9/1/88children and was not specifically measured in our study[9,38]. Seventh, our sample had relatively small numbersof families in the lower SES groups, and generally highlevels of physical activity across SES, potentially limitingthe generalizability of our findings.Present findings are a step in understanding SESdisparities in childhood obesity. The finding that lowSES home environments have more electronic devices inbedrooms and fewer pieces of play equipment than inhigh SES homes is cause for concern. Sedentary-promoting devices in the bedroom emerged as an import-ant mediator of the SES-sedentary behavior association.Additional research is recommended that can informinterventions to improve the healthfulness of homeenvironments of low SES families.AbbreviationsBMI: Body mass index; NIK: Neighborhood Impact on Kids Study;SES: Socioeconomic status; PA: Physical activity; METs: Metabolic equivalents;MVPA: Moderate-to-vigorous physical activity.Competing interestsThe authors declare that they have no competing interests.AcknowledgementsWe are grateful to Dr. Trina Colburn for her assistance and supportthroughout the project. We thank Stephanie Kneeshaw-Price for her work onthe categorization of accelerometer data by place logs. We also thank theNIK staff and study participants. This project was supported by grants fromNIH (ES014240 and UL1RR025014) and a Hearst Family Foundation Grant tothe Seattle Children’s Center for Child Health Behavior and Development.Author details1Seattle Children’s Research Institute, M/S CW8-6, P.O. Box 5371, Seattle, WA98145-5005, USA. 2University of Washington, Seattle, WA, USA. 3SDSU, SanDiego, CA, USA. 4University of California, San Diego, CA, USA. 5University ofBritish Columbia, Vancouver, Canada.Authors’ contributionsPT and BS developed the study design for this manuscript. BS is the pi and jsand lf are the co-investigators of the NIK study, from which data for thismanuscript were obtained; BS, JS and LF designed the NIK study. Dataanalyses were conducted BY PT and CZ. PT wrote the initial draft of themanuscript and all authors critically reviewed and revised versions. Allauthors read and approved the final manuscript.Received: 20 January 2012 Accepted: 26 July 2012Published: 26 July 2012References1. Newacheck PW: Poverty and childhood chronic illness. Archives ofpediatrics & adolescent medicine 1994, 148(11):1143–1149.2. Starfield B, Robertson J, Riley AW: Social class gradients and health inchildhood. Ambulatory pediatrics: the official journal of the AmbulatoryPediatric Association 2002, 2(4):238–246.3. Goodman E, Slap GB, Huang B: The Public Health Impact ofSocioeconomic Status on Adolescent Depression and Obesity.Am J Public Health 2003, 93(11):1844–1850.4. Delva J, O’Malley PM, Johnston LD: Racial/Ethnic and SocioeconomicStatus Differences in Overweight and Health-Related Behaviors amongAmerican Students: National Trends 1986–2003. J Adolesc Heal 2006,39(4):536–545.5. Baranowski T, Lin L, Wetter D, Resnicow K, Hearn M: Theory as mediatingvariables: Why aren’t community interventions working as desired?Ann Epidemiol 1997, 7:S89–S95.6. Davison KK, Birch LL: Childhood overweight: a contextual model andrecommendations for future research. Obesity reviews: an official journal ofthe International Association for the Study of Obesity 2001, 2(3):159–171.7. Jago R, Davison KK, Thompson JL, Page AS, Brockman R, Fox KR: ParentalSedentary Restriction, Maternal Parenting Style, and Television ViewingAmong 10- to 11-Year-Olds. Pediatrics 2011, 128(3):e572–e578.8. Fairclough SJ, Boddy LM, Hackett AF, Stratton G: Associations betweenchildren’s socioeconomic status, weight status, and sex, with screen-based sedentary behaviours and sport participation. Int J Pediatr Obes2009, 4(4):299–305.9. 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(2):129–154.10. Saelens BE, Sallis JF, Frank LD, Couch SC, Zhou C, Colburn T, Cain KL,Chapman J, Glanz K: Obesogenic Neighborhood Environments, Child andParent Obesity: The Neighborhood Impact on Kids Study. AmericanJournal of Preventive Medicine 2012, 42(5):e57–e64.11. Frank LD, Saelens BE, Chapman J, Sallis JF, Kerr J, Glanz K, Couch SC,Learnihan V, Zhou C, Colburn T, et al: Objective Assessment ofObesogenic Environments in Youth: Geographic Information SystemMethods and Spatial Findings from the Neighborhood Impact on KidsStudy. American Journal of Preventive Medicine 2012, 42(5):e47–e55.12. U.S. Census Bureau: State and County QuickFacts. Data derived fromPopulation Estimates, Census of Population and Housing, Small Area Incomeand Poverty Estimates, State and County Housing Unit Estimates, CountyBusiness Patterns, Nonemployer Statistics, Economic Census, Survey ofBusiness Owners, Building Permits, Consolidated Federal Funds Report.http://quickfacts.census.gov/qfd/index.html.13. Grow HM, Saelens BE, Kerr J, Sallis JF, Durant NH, Norman GJ: Where areyouth active? Roles of proximity, active transport, and built environment.Med Sci Sports Exerc Medicine and Science in Sports and Exercise 2008,40(12):2071–2079.14. Joe L, Carlson J: Active Where? Individual Item Reliability Report.: Active LivingResearch; 2010. http://www.drjamessallis.sdsu.edu/measures.html.15. Pate RR, Almeida MJ, McIver KL, Pfeiffer KA, Dowda M: Validation andCalibration of an Accelerometer in Preschool Children[ast]. Obesity 2006,14(11):2000–2006.16. Freedson P, Sirard J, Debold E, Pate R, Dowda M, Trost S, Sallis J: Calibrationof the Computer Science and Applications, Inc. (CSA) accelerometer.Medicine & Science in Sports & Exercise 1997, 29(Suppl):45.17. Evenson KR, Gill K, Catellier DJ, Ondrak KS, McMurray RG: Calibration of twoobjective measures of physical activity for children. J Sports Sci Journal ofSports Sciences 2008, 26(14):1557–1565.18. Trost SG, Loprinzi PD, Moore R, Pfeiffer KA: Comparison of accelerometercut points for predicting activity intensity in youth. Med Sci Sports ExercMedicine and Science in Sports and Exercise 2011, 43(7):1360–1368.19. Hoyos Cillero I, Jago R: Systematic review of correlates of screen-viewingamong young children. Preventive Medicine 2010, 51(1):3–10.20. Rideout V: Generation M2: Media in the lives of 8 to 18 year olds. Menlo Park,CA: Kaiser Family Foundation; 2010.21. Weir LA, Etelson D, Brand DA: Parents’ perceptions of neighborhoodsafety and children’s physical activity. Preventive Medicine 2006,43(3):212–217.22. Stenhammar C, Sarkadi A, Edlund B: The role of parents’ educationalbackground in healthy lifestyle practices and attitudes of their6-year-old children. Public Health Nutrition 2007, 10(11):1305–1313.23. Chowhan J, Stewart JM: Television and the behaviour of adolescents:Does socio-economic status moderate the link? Social Science & Medicine2007, 65(7):1324–1336.24. Fairclough SJ, Boddy LM, Hackett AF, Stratton G: Associations betweenchildren’s socioeconomic status, weight status, and sex, with screen-based sedentary behaviours and sport participation. Int J Pediatr ObesInternational Journal of Pediatric Obesity 2009, 4(4):299–305.25. Link BG, Phelan J: Social Conditions As Fundamental Causes of Disease.J Heal Soc Behav 1995, 35:80–94.26. Ferreira I, Horst K, Wendel-Vos S, Kremers S, van Lenthe F, Brug J:Environmental correlates of physical activity in youth - a review andupdate. Obes Rev 2006, 8:129–154.27. Centers for Disease C: Prevention: Prevalence of regular physical activityamong adults--United States, 2001 and 2005. MMWR Morbidity andmortality weekly report 2007, 56(46):1209–1212.Tandon et al. International Journal of Behavioral Nutrition and Physical Activity 2012, 9:88 Page 8 of 9http://www.ijbnpa.org/content/9/1/8828. Marshall SJ, Biddle SJH, Sallis JF, McKenzie TL, Conway TL: Clustering ofSedentary Behaviors and Physical Activity Among Youth: ACross-National Study. Pediatr Exerc Sci 2002, 14:401–417.29. Committee on Public Education: Children, Adolescents, and Television:American Academy of Pediatrics Policy Statement. Pediatrics 2001,107(2):423–426.30. Dorr A, Kovaric P, Doubleday C: Parent–child Coviewing of Television.Journal of Broadcasting & Electronic Media 1989, 33(1):35–51.31. Barr-Anderson DJ, Van Den Berg P, Neumark-Sztainer D, Story M:Characteristics Associated With Older Adolescents Who Have aTelevision in Their Bedrooms. Pediatrics 2008, 121(4):718–724.32. Li S, Jin X, Wu S, Jiang F, Yan C, Shen X: The impact of media use on sleeppatterns and sleep disorders among school-aged children in China.Sleep 2007, 30(3):361–367.33. Garrison MM, Liekweg K, Christakis DA: Media use and child sleep: Theimpact of content, timing, and environment. Pediatrics Pediatrics 2011,128(1):29–35.34. Morgenstern M, Sargent JD, Hanewinkel R: Relation BetweenSocioeconomic Status and Body Mass Index: Evidence of an IndirectPath via Television Use. Arch Pediatr Adolesc Med 2009, 163(8):731–738.35. Anderson DR, Field DE, Collins PA, Lorch EP, Nathan JG: Estimates of YoungChildren’s Time with Television: A Methodological Comparison of ParentReports with Time-Lapse Video Home Observation. Child Development1985, 56(5):1345–1357.36. Guinhouya CB, Hubert H, Soubrier S, Vilhelm C, Lemdani M, Durocher A:Moderate-to-vigorous physical activity among children: discrepancies inaccelerometry-based cut-off points. Obesity (Silver Spring, Md) 2006,14(5):774–777.37. Physical Activity Guidelines for Americans. http://www.health.gov/paguidelines/.38. Sallis J, Prochaska J, Taylor W: A review of correlates of physical activity ofchildren and adolescents. Med Sci Sports Exerc 2000, 32:963–975.doi:10.1186/1479-5868-9-88Cite this article as: Tandon et al.: Home environment relationships withchildren’s physical activity, sedentary time, and screen time bysocioeconomic status. International Journal of Behavioral Nutrition andPhysical Activity 2012 9:88.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/submitTandon et al. International Journal of Behavioral Nutrition and Physical Activity 2012, 9:88 Page 9 of 9http://www.ijbnpa.org/content/9/1/88


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