Parental Guidance Advised: Associations Between Parental Television Limits and Health Behaviors Among Obese Children Jennifer K. Cheng, MD, MPH; Renata L. Koziol, MPH; Elsie M. Taveras, MD, MPH From the Division of General Pediatrics, Department of Medicine, Boston Children’s Hospital, Boston, Mass (Dr Cheng); Obesity Prevention Program, Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, Mass (Ms Koziol); and Division of General Pediatrics, Department of Pediatrics, Massachusetts General Hospital for Children, Boston, Mass Dr (Dr Taveras) The authors declare that they have no conflict of interest. Address correspondence to Jennifer K. Cheng, MD, MPH, Division of General Pediatrics, Department of Medicine, Children’s Hospital Boston, Hunnewell Ground Building, 300 Longwood Ave, Boston, MA 02115 (e-mail: [email protected]). Received for publication May 18, 2014; accepted August 24, 2014.
ABSTRACT OBJECTIVE: To examine associations between parental limits
watching TV (0.38 hours/day; 95% CI 0.54, 0.21), playing video or computer games (0.15 hours/day; 95% CI 0.25, 0.04), and using the Internet (0.08 hours/day; 95% CI 0.15, 0.006). In addition, children with limits on TV time were less likely to eat breakfast (OR 0.73; 95% CI 0.53, 0.99) or dinner (OR 0.53; 95% CI 0.39, 0.73) with the TV on, and they slept longer during weeknights: 0.21 hours/day (95% CI 0.05, 0.36). Moreover, children whose parents set rules on TV programming content were also less likely to have a TV in their bedroom (OR 0.37; 95% CI 0.18, 0.79). CONCLUSIONS: The relationship between parental limits on TV viewing and obesogenic child behaviors deserves further investigation.
on TV viewing and child health behaviors. METHODS: We surveyed 816 parents of children 6 to 12 years of age who were participating in a primary care–based obesity intervention. The main exposures were parental limits placed on child TV viewing time and TV content. Outcomes included screen-related behaviors, sleep habits, eating routines, and physical activity. We performed bivariate and multivariable analyses to examine independent associations of parental TV viewing rules with our outcomes. RESULTS: In multivariable analyses adjusted for child age, sex, race/ethnicity; parental education and US-born status; income and primary language, children whose parents set limits on TV time were less likely to have a TV in their bedroom (odds ratio [OR] 0.45; 95% confidence interval [CI] 0.31, 0.64) or to fall asleep while watching TV (OR 0.47; 95% CI 0.34, 0.67). Children with parental limits spent fewer weekday hours
KEYWORDS: diet; obesity; parent; physical activity; sleep; television
ACADEMIC PEDIATRICS 2015;15:204–209
WHAT’S NEW
performance among children in the 4th grade.10 A recent longitudinal study suggests a link between excessive television viewing in childhood and adolescence and increased risk for criminal convictions, antisocial personality disorder, and more aggressive personality traits in early adulthood.11 Parental guidance has shown a powerful moderating effect on a range of child behaviors, from the development of prosocial behaviors among young children12 to the reduction of risk behaviors including substance use, gang affiliation, and sexual activity among youth.13 However, little is known about the role of parental guidance regarding TV viewing on child health behaviors; to our knowledge, no previous studies have described this relationship among obese children. Our objective was to examine associations between parental limits on TV viewing and child lifestyle behaviors relevant to obesity including screen media use, sleep, physical activity, and eating habits. We hypothesized that children whose parents set limits on child screen time or content would have more favorable screen viewing, sleep, physical activity, and eating routines compared with children without such rules.
Television use is widespread and has been linked to obesity. Parental limits on television use may influence several key health behaviors including screen time, sleeping, and eating routines among obese children.
THE USE OF television and other screen media has proliferated over the past several decades. Television penetration of US households rose from 9.0% in 1950 to 64.5% in 1955, 92.6% in 1965, and 97.1% in 2012.1 The average US child watches an average of 80 minutes of TV per day and is exposed to nearly 4 hours of background television on a typical day.1,2 Eight- to 18-year-olds “spend more time with media than in any other activity besides (maybe) sleeping— an average of more than 7½ hours a day, seven days a week.”3 Considerable evidence supports the relationship between time spent watching television and a host of adverse outcomes in terms of physical activity,4 sleep,5 diet,6 weight,7 and cardiovascular health.8 High levels of television exposure in early childhood also appear to undermine attention9 and have been associated with poorer academic ACADEMIC PEDIATRICS Copyright ª 2015 by Academic Pediatric Association
204
Volume 15, Number 2 March–April 2015
ACADEMIC PEDIATRICS
PARENTAL GUIDANCE ADVISED
METHODS PARTICIPANTS AND SETTING We administered a baseline survey to English-speaking parents or caregivers (hereafter, parents) of children ages 6 to 12 years with a body mass index (BMI) in the >90th percentile seen for a well-child care visit within the past 12 months at one of 14 Harvard Vanguard Medical Associates (HVMA) primary care offices participating in The Study of Technology to Accelerate Research (STAR) project, a large cluster-randomized controlled trial. Eligibility for the STAR study was a BMI in the >95th percentile, while our recruitment pool included children with a BMI in the >90th percentile because our preliminary studies showed that these children had the highest likelihood of onset of obesity (vs those with a BMI in the 85th to 90th percentile). Additional eligibility criteria included residence in permanent housing, and plan to continue receiving primary care within HVMA over the next 12 months. Children with chronic health conditions that could interfere with growth or anthropometric measurements or that could contraindicate diet or physical activity modifications were excluded. Details of the STAR intervention are described elsewhere.14 Briefly, STAR is a 3-arm, cluster-randomized controlled trial designed to examine the extent to which computerized decision support tools in the electronic health record delivered to primary care providers at the point of care, with or without direct-to-parent support and coaching, will increase adoption of comparative effectiveness research evidence for management of obese children.14 SURVEY A research assistant verified eligibility to participate in STAR, obtained verbal informed consent, and administered the 58-item, approximately 30-minute telephone survey using a structured script. Demographic information collected included questions about the family composition, respondent national origin, education, race/ ethnicity, age, primary language, and marital status. Participants received a $20 gift card as a token of appreciation for their time. The institutional review boards of Harvard Pilgrim Health Care and HVMA approved the study protocol. EXPOSURES To assess the main exposures of parental limits on 1) TV viewing time and 2) content of the programming that was viewed, we asked parents: “Do you limit how much time (your child) is allowed to watch television?” and “Do you put limits on what (your child) is and is not allowed to watch on television?” The response categories were “Yes” or “No.” OUTCOMES We assessed multiple child behaviors as outcomes using questions adapted from validated screen,15,16
205
physical activity,17 sleep,18,19 and eating20 instruments. For screen-related behaviors, we asked parents to report: 1) whether there was a television in the room where the child slept and 2) how much time on average (in hours and minutes per day) their child had spent during the past month: a) watching TV shows, b) viewing DVDs, videos, or movies, c) playing video or computer games, and d) using Internet-related activities such as email, iPhone Apps, Facebook or YouTube, not including homework and games. We assessed eating routines by asking parents to estimate how often their child ate 1) breakfast and 2) dinner in a room with the television turned on for both a typical weekday and a weekend day over the past month; response categories were: every day of the week, 4 to 6 days a week, 1 to 3 days a week, or never. To assess sleep behaviors, we asked parents how often their child had fallen asleep while watching television over the last month (never; less than once per week; 1–2 times per week; 3–4 times per week; 5–7 times per week; don’t know) and also how many hours per day their child slept in a usual 24-hour period on a weekday as well as on a weekend day. Finally, to assess physical activity, we asked parents how many hours per week, in the past month, their child had spent 1) walking (for example, to/from school, a friend’s house or to the store), 2) in light or moderate recreational activities or sports (such as biking, skateboarding, dancing, gymnastics, baseball, playing outdoors, or other similar activities, not including walking) and in 3) vigorous recreational activities or sports (such as swimming, running, or basketball). STATISTICAL ANALYSIS We first examined descriptive statistics and bivariate analyses to assess differences in sample characteristics and associations between our exposures and outcomes. We used the Chi-square test statistic for categorical variables and mean, standard deviation (SD), and the t-test statistic for continuous variables. We then performed multivariable analyses and adjusted for covariates and potential confounders determined a priori such as child age, sex, race/ ethnicity, parental education, US-born status, household income and primary language spoken at home to examine independent associations of parental TV viewing rules with our outcomes of interest. We conducted all analyses using SAS version 9.2 (SAS Institute, Cary, NC).
RESULTS Participant characteristics are shown in Table 1. Of the 816 children enrolled, 53% were white, 20% black, 15% Hispanic, 13% other. Mean (SD) child age was 9.7 years (1.9) with BMI z score of 1.9 (0.4). 57% of parents had a college or postgraduate education and 74% were overweight or obese. Parents who set time limits for their child’s TV viewing tended to have higher educational attainment and household incomes and were less likely to be US born (72.4% vs 83.2%, P ¼.001) or speak English as the primary language (89.5% vs 95.9%, P ¼ .003). In terms of parental guidance regarding TV programming
206
CHENG ET AL
ACADEMIC PEDIATRICS
Table 1. Characteristics of 816 Study Participants Has Time Limits Characteristic Child Age, mean (SD) BMI z score, mean (SD) Boy, n (%) Race/ethnicity, n (%) White Black Hispanic Other Parent/Caregiver Mother, n (%) Age, mean (SD) BMI, kg/m2, mean (SD) BMI status Normal Overweight Obese US birth, n (%) Marital status Married Other English language, n (%) Education High school graduate or less College graduate Postgraduate training Household Income $20K or less $20K to $70K More than $70K Size, mean (SD)
Has Content Limits
Total Sample
Yes (n ¼ 572)
No (n ¼ 244)
P*
Yes (n ¼ 770)
No (n ¼ 46)
P*
9.7 (1.9) 1.9 (0.4) 442 (54.2)
9.6 (1.9) 1.9 (0.4) 305 (53.3)
9.9 (2.0) 1.9 (0.5) 137 (56.2)
.12 .72 .46 .07
9.7 (1.9) 1.9 (0.4) 413 (53.6)
10.4 (1.8) 1.9 (0.4) 29 (63.0)
.01 .21 .21 .57
427 (52.5) 163 (20.0) 120 (14.7) 104 (12.8)
290 (50.8) 123 (21.5) 78 (13.7) 80 (14.0)
137 (56.4) 40 (16.5) 42 (17.3) 24 (9.9)
401 (52.2) 156 (20.3) 115 (15.0) 96 (12.5)
26 (56.5) 7 (15.2) 5 (10.9) 8 (17.4)
724 (88.7) 41.0 (6.7) 29.5 (6.6)
501 (87.6) 41.2 (6.6) 29.4 (6.4)
223 (91.4) 40.6 (6.9) 29.9 (7.0)
688 (89.4) 40.9 (6.6) 29.5 (6.7)
36 (78.3) 42.9 (7.7) 29.4 (5.2)
204 (26.0) 278 (35.4) 303 (38.6) 617 (75.6)
145 (26.3) 199 (36.1) 208 (37.7) 414 (72.4)
59 (25.3) 79 (33.9) 95 (40.8) 203 (83.2)
194 (26.3) 259 (35.1) 286 (38.7) 584 (75.8)
10 (21.7) 19 (41.3) 17 (37.0) 33 (71.7)
617 (75.8) 197 (24.2) 746 (91.4)
440 (77.2) 130 (22.8) 512 (89.5)
177 (72.5) 67 (27.5) 234 (95.9)
583 (75.9) 185 (24.1) 707 (91.8)
34 (73.9) 12 (26.1) 39 (84.8)
351 (43.1) 269 (33.0) 195 (23.9)
227 (39.8) 191 (33.5) 153 (26.8)
124 (50.8) 78 (32.0) 42 (17.2)
328 (42.7) 257 (33.4) 184 (23.9)
23 (50.0) 12 (26.1) 11 (23.9)
74 (9.4) 269 (34.1) 446 (56.5) 4.3 (1.3)
49 (8.8) 175 (31.6) 330 (59.6) 4.3 (1.2)
25 (10.6) 94 (40.0) 116 (49.4) 4.3 (1.4)
65 (8.7) 254 (34.1) 426 (57.2) 4.3 (1.3)
9 (20.5) 15 (34.1) 20 (45.5) 4.2 (1.2)
.12 .29 .27 .71
.001 .16
.003 .003
.03
.58
.02 .05 .90 .65
.53 .76
.05 .53
.03
.87
BMI indicates body mass index. *P significant at .05.
content, higher-income-earning mothers and those who spoke English as their primary language (91.8% vs 84.8%, P ¼ .05) were more likely to have rules about what their children were allowed to watch on TV. Table 2 shows bivariate analyses of child behaviors by parental limits on TV time as well as parental limits on TV programming content. Parental limits on TV viewing time was associated with lower rates of TV presence in the child’s bedroom (35% vs 54.5%, P
ABSTRACT OBJECTIVE: To examine associations between parental limits
watching TV (0.38 hours/day; 95% CI 0.54, 0.21), playing video or computer games (0.15 hours/day; 95% CI 0.25, 0.04), and using the Internet (0.08 hours/day; 95% CI 0.15, 0.006). In addition, children with limits on TV time were less likely to eat breakfast (OR 0.73; 95% CI 0.53, 0.99) or dinner (OR 0.53; 95% CI 0.39, 0.73) with the TV on, and they slept longer during weeknights: 0.21 hours/day (95% CI 0.05, 0.36). Moreover, children whose parents set rules on TV programming content were also less likely to have a TV in their bedroom (OR 0.37; 95% CI 0.18, 0.79). CONCLUSIONS: The relationship between parental limits on TV viewing and obesogenic child behaviors deserves further investigation.
on TV viewing and child health behaviors. METHODS: We surveyed 816 parents of children 6 to 12 years of age who were participating in a primary care–based obesity intervention. The main exposures were parental limits placed on child TV viewing time and TV content. Outcomes included screen-related behaviors, sleep habits, eating routines, and physical activity. We performed bivariate and multivariable analyses to examine independent associations of parental TV viewing rules with our outcomes. RESULTS: In multivariable analyses adjusted for child age, sex, race/ethnicity; parental education and US-born status; income and primary language, children whose parents set limits on TV time were less likely to have a TV in their bedroom (odds ratio [OR] 0.45; 95% confidence interval [CI] 0.31, 0.64) or to fall asleep while watching TV (OR 0.47; 95% CI 0.34, 0.67). Children with parental limits spent fewer weekday hours
KEYWORDS: diet; obesity; parent; physical activity; sleep; television
ACADEMIC PEDIATRICS 2015;15:204–209
WHAT’S NEW
performance among children in the 4th grade.10 A recent longitudinal study suggests a link between excessive television viewing in childhood and adolescence and increased risk for criminal convictions, antisocial personality disorder, and more aggressive personality traits in early adulthood.11 Parental guidance has shown a powerful moderating effect on a range of child behaviors, from the development of prosocial behaviors among young children12 to the reduction of risk behaviors including substance use, gang affiliation, and sexual activity among youth.13 However, little is known about the role of parental guidance regarding TV viewing on child health behaviors; to our knowledge, no previous studies have described this relationship among obese children. Our objective was to examine associations between parental limits on TV viewing and child lifestyle behaviors relevant to obesity including screen media use, sleep, physical activity, and eating habits. We hypothesized that children whose parents set limits on child screen time or content would have more favorable screen viewing, sleep, physical activity, and eating routines compared with children without such rules.
Television use is widespread and has been linked to obesity. Parental limits on television use may influence several key health behaviors including screen time, sleeping, and eating routines among obese children.
THE USE OF television and other screen media has proliferated over the past several decades. Television penetration of US households rose from 9.0% in 1950 to 64.5% in 1955, 92.6% in 1965, and 97.1% in 2012.1 The average US child watches an average of 80 minutes of TV per day and is exposed to nearly 4 hours of background television on a typical day.1,2 Eight- to 18-year-olds “spend more time with media than in any other activity besides (maybe) sleeping— an average of more than 7½ hours a day, seven days a week.”3 Considerable evidence supports the relationship between time spent watching television and a host of adverse outcomes in terms of physical activity,4 sleep,5 diet,6 weight,7 and cardiovascular health.8 High levels of television exposure in early childhood also appear to undermine attention9 and have been associated with poorer academic ACADEMIC PEDIATRICS Copyright ª 2015 by Academic Pediatric Association
204
Volume 15, Number 2 March–April 2015
ACADEMIC PEDIATRICS
PARENTAL GUIDANCE ADVISED
METHODS PARTICIPANTS AND SETTING We administered a baseline survey to English-speaking parents or caregivers (hereafter, parents) of children ages 6 to 12 years with a body mass index (BMI) in the >90th percentile seen for a well-child care visit within the past 12 months at one of 14 Harvard Vanguard Medical Associates (HVMA) primary care offices participating in The Study of Technology to Accelerate Research (STAR) project, a large cluster-randomized controlled trial. Eligibility for the STAR study was a BMI in the >95th percentile, while our recruitment pool included children with a BMI in the >90th percentile because our preliminary studies showed that these children had the highest likelihood of onset of obesity (vs those with a BMI in the 85th to 90th percentile). Additional eligibility criteria included residence in permanent housing, and plan to continue receiving primary care within HVMA over the next 12 months. Children with chronic health conditions that could interfere with growth or anthropometric measurements or that could contraindicate diet or physical activity modifications were excluded. Details of the STAR intervention are described elsewhere.14 Briefly, STAR is a 3-arm, cluster-randomized controlled trial designed to examine the extent to which computerized decision support tools in the electronic health record delivered to primary care providers at the point of care, with or without direct-to-parent support and coaching, will increase adoption of comparative effectiveness research evidence for management of obese children.14 SURVEY A research assistant verified eligibility to participate in STAR, obtained verbal informed consent, and administered the 58-item, approximately 30-minute telephone survey using a structured script. Demographic information collected included questions about the family composition, respondent national origin, education, race/ ethnicity, age, primary language, and marital status. Participants received a $20 gift card as a token of appreciation for their time. The institutional review boards of Harvard Pilgrim Health Care and HVMA approved the study protocol. EXPOSURES To assess the main exposures of parental limits on 1) TV viewing time and 2) content of the programming that was viewed, we asked parents: “Do you limit how much time (your child) is allowed to watch television?” and “Do you put limits on what (your child) is and is not allowed to watch on television?” The response categories were “Yes” or “No.” OUTCOMES We assessed multiple child behaviors as outcomes using questions adapted from validated screen,15,16
205
physical activity,17 sleep,18,19 and eating20 instruments. For screen-related behaviors, we asked parents to report: 1) whether there was a television in the room where the child slept and 2) how much time on average (in hours and minutes per day) their child had spent during the past month: a) watching TV shows, b) viewing DVDs, videos, or movies, c) playing video or computer games, and d) using Internet-related activities such as email, iPhone Apps, Facebook or YouTube, not including homework and games. We assessed eating routines by asking parents to estimate how often their child ate 1) breakfast and 2) dinner in a room with the television turned on for both a typical weekday and a weekend day over the past month; response categories were: every day of the week, 4 to 6 days a week, 1 to 3 days a week, or never. To assess sleep behaviors, we asked parents how often their child had fallen asleep while watching television over the last month (never; less than once per week; 1–2 times per week; 3–4 times per week; 5–7 times per week; don’t know) and also how many hours per day their child slept in a usual 24-hour period on a weekday as well as on a weekend day. Finally, to assess physical activity, we asked parents how many hours per week, in the past month, their child had spent 1) walking (for example, to/from school, a friend’s house or to the store), 2) in light or moderate recreational activities or sports (such as biking, skateboarding, dancing, gymnastics, baseball, playing outdoors, or other similar activities, not including walking) and in 3) vigorous recreational activities or sports (such as swimming, running, or basketball). STATISTICAL ANALYSIS We first examined descriptive statistics and bivariate analyses to assess differences in sample characteristics and associations between our exposures and outcomes. We used the Chi-square test statistic for categorical variables and mean, standard deviation (SD), and the t-test statistic for continuous variables. We then performed multivariable analyses and adjusted for covariates and potential confounders determined a priori such as child age, sex, race/ ethnicity, parental education, US-born status, household income and primary language spoken at home to examine independent associations of parental TV viewing rules with our outcomes of interest. We conducted all analyses using SAS version 9.2 (SAS Institute, Cary, NC).
RESULTS Participant characteristics are shown in Table 1. Of the 816 children enrolled, 53% were white, 20% black, 15% Hispanic, 13% other. Mean (SD) child age was 9.7 years (1.9) with BMI z score of 1.9 (0.4). 57% of parents had a college or postgraduate education and 74% were overweight or obese. Parents who set time limits for their child’s TV viewing tended to have higher educational attainment and household incomes and were less likely to be US born (72.4% vs 83.2%, P ¼.001) or speak English as the primary language (89.5% vs 95.9%, P ¼ .003). In terms of parental guidance regarding TV programming
206
CHENG ET AL
ACADEMIC PEDIATRICS
Table 1. Characteristics of 816 Study Participants Has Time Limits Characteristic Child Age, mean (SD) BMI z score, mean (SD) Boy, n (%) Race/ethnicity, n (%) White Black Hispanic Other Parent/Caregiver Mother, n (%) Age, mean (SD) BMI, kg/m2, mean (SD) BMI status Normal Overweight Obese US birth, n (%) Marital status Married Other English language, n (%) Education High school graduate or less College graduate Postgraduate training Household Income $20K or less $20K to $70K More than $70K Size, mean (SD)
Has Content Limits
Total Sample
Yes (n ¼ 572)
No (n ¼ 244)
P*
Yes (n ¼ 770)
No (n ¼ 46)
P*
9.7 (1.9) 1.9 (0.4) 442 (54.2)
9.6 (1.9) 1.9 (0.4) 305 (53.3)
9.9 (2.0) 1.9 (0.5) 137 (56.2)
.12 .72 .46 .07
9.7 (1.9) 1.9 (0.4) 413 (53.6)
10.4 (1.8) 1.9 (0.4) 29 (63.0)
.01 .21 .21 .57
427 (52.5) 163 (20.0) 120 (14.7) 104 (12.8)
290 (50.8) 123 (21.5) 78 (13.7) 80 (14.0)
137 (56.4) 40 (16.5) 42 (17.3) 24 (9.9)
401 (52.2) 156 (20.3) 115 (15.0) 96 (12.5)
26 (56.5) 7 (15.2) 5 (10.9) 8 (17.4)
724 (88.7) 41.0 (6.7) 29.5 (6.6)
501 (87.6) 41.2 (6.6) 29.4 (6.4)
223 (91.4) 40.6 (6.9) 29.9 (7.0)
688 (89.4) 40.9 (6.6) 29.5 (6.7)
36 (78.3) 42.9 (7.7) 29.4 (5.2)
204 (26.0) 278 (35.4) 303 (38.6) 617 (75.6)
145 (26.3) 199 (36.1) 208 (37.7) 414 (72.4)
59 (25.3) 79 (33.9) 95 (40.8) 203 (83.2)
194 (26.3) 259 (35.1) 286 (38.7) 584 (75.8)
10 (21.7) 19 (41.3) 17 (37.0) 33 (71.7)
617 (75.8) 197 (24.2) 746 (91.4)
440 (77.2) 130 (22.8) 512 (89.5)
177 (72.5) 67 (27.5) 234 (95.9)
583 (75.9) 185 (24.1) 707 (91.8)
34 (73.9) 12 (26.1) 39 (84.8)
351 (43.1) 269 (33.0) 195 (23.9)
227 (39.8) 191 (33.5) 153 (26.8)
124 (50.8) 78 (32.0) 42 (17.2)
328 (42.7) 257 (33.4) 184 (23.9)
23 (50.0) 12 (26.1) 11 (23.9)
74 (9.4) 269 (34.1) 446 (56.5) 4.3 (1.3)
49 (8.8) 175 (31.6) 330 (59.6) 4.3 (1.2)
25 (10.6) 94 (40.0) 116 (49.4) 4.3 (1.4)
65 (8.7) 254 (34.1) 426 (57.2) 4.3 (1.3)
9 (20.5) 15 (34.1) 20 (45.5) 4.2 (1.2)
.12 .29 .27 .71
.001 .16
.003 .003
.03
.58
.02 .05 .90 .65
.53 .76
.05 .53
.03
.87
BMI indicates body mass index. *P significant at .05.
content, higher-income-earning mothers and those who spoke English as their primary language (91.8% vs 84.8%, P ¼ .05) were more likely to have rules about what their children were allowed to watch on TV. Table 2 shows bivariate analyses of child behaviors by parental limits on TV time as well as parental limits on TV programming content. Parental limits on TV viewing time was associated with lower rates of TV presence in the child’s bedroom (35% vs 54.5%, P
