Graduate Teses and Dissertations Graduate College 2010 Parenting styles, child BMI, and ratings of obesigenic environments in families of children age 5-11 Rachel Rae Johnson Iowa State University, graygirl@iastate.edu Follow this and additional works at: htp://lib.dr.iastate.edu/etd Part of the Kinesiology Commons Tis Tesis is brought to you for free and open access by the Graduate College at Digital Repository @ Iowa State University. It has been accepted for inclusion in Graduate Teses and Dissertations by an authorized administrator of Digital Repository @ Iowa State University. For more information, please contact hinefuku@iastate.edu. Recommended Citation Johnson, Rachel Rae, "Parenting styles, child BMI, and ratings of obesigenic environments in families of children age 5-11" (2010). Graduate Teses and Dissertations. Paper 11623.
Parenting styles, child BMI, and ratings of obesigenic environments in families of children age 5-11
by
Rachel Rae J ohnson
A thesis submitted to the graduate faculty
in partial fulfillment of the requirements for the degree of
MASTER OF SCIENCE
Major: Kinesiology (Behavioral Basis of Physical Activity) Program of Study Committee: Gregory J . Welk, Major Professor Douglas A. Gentile Amy S. Welch
Iowa State University
Ames, Iowa
2010
Copyright Rachel Rae Johnson, 2010. All rights reserved. ii
TABLE OF CONTENTS
ABSTRACT iii
INTRODUCTION 1
LITERATURE REVIEW 3
METHODS 18
RESULTS 28
DISCUSSION 37
CONCLUSION 42
APPENDIX A: ADDITIONAL RESULTS 43
APPENDIX B: FAMILY NUTRITION AND PHYSICAL ACTIVITY ASSESSMENT ADULT VERSION 46
APPENDIX C: FAMILY NUTRITION AND PHYSICAL ACTIVITY ASSESSMENT: 21 ITEM FORMAT 48
APPENDIX D: FAMILY NUTRITION AND PHYSICAL ACTIVITY ASSESSMENT CHILD VERSION 49
APPENDIX E: PARENT STYLE AND DIMENSIONS QUESTIONNAIRE (PSDQ) 51
REFERENCES 53 iii
ABSTRACT Background: The epidemic of childhood obesity is a multi-factorial problem but the childs home environment and parenting practices clearly play a role. This study evaluates the utility of a behaviorally based screening tool for evaluating practices and home environments. This study also seeks to determine if parenting styles influence parent and child environmental ratings and child BMI. Methods: 313 elementary students and 75 of their parents completed separate versions of the Family Nutrition and Physical Activity (FNPA) instrument. Parents also completed the Parenting Styles and Dimension Questionnaire (PSDQ), a 58 item survey that categorizes parenting practices into three styles: authoritative, authoritarian, and permissive. Body Mass Index (BMI) data was obtained by trained staff. Cronbachs alpha was run to check reliability of parent and child FNPA reports. Pearson product moment correlations among the parent and child FNPA scores were used to determine overall associations and parent-child agreement. Regression analyses were used to determine if parenting styles were related to FNPA and child BMI. Cluster analysis was also used to identify patterns in the PSDQ classifications that may be associated with particularly high or low FNPA scores. Results: Correlations between childrens FNPA scores and parent scores were low (r =.188). Correlations between the Parent FNPA score and child BMI-z score was low (r =-.31) but statistically significant. Parents were more internally consistent in evaluating home environments. Older children were more consistent reporters than younger children. Cluster analysis revealed clear differences in associated FNPA scores. Less authoritative parenting was associated with more obesigenic environments. Less obesigenic environments were associated with authoritative parenting styles. Regression analysis shows iv
that authoritative parenting was the best predictor of FNPA and no significant predictor was determined for child BMI. Conclusion: Parents and children differ in perceptions of their home environments and parenting style can alter parent ratings of the home environment. 1 INTRODUCTION The epidemic of obesity is one of the most pressing public health problems facing our country (NCHS, 2007). Obesity has become even more prevalent in Americas youth in recent years (Ogden et al., 2006; Freedman et al., 2002). There are specific concerns about the increasing prevalence in youth because overweight youth are more likely to become overweight adolescents and adults (Dietz & Gortmaker, 2001; Ebbeling et al., 2002). Being overweight can also increase risk for other negative and sometimes life threatening health outcomes (Must & Strauss, 1999; Reilly et al., 2005). Obesity is a multi-factorial condition that can be a result of many lifestyle habits and environments. A number of behaviors influence risk for overweight in children including but not limited to, television viewing time (Laurson et al., 2008; Deitz & Strasburger, 1991; Gortmaker et al., 1990; Muller et al., 2002; Vanhala et al., 2009; Cheng, 2005), video game play (Laurson et al, 2008; Skinner et al, 2004), physical inactivity (Harrison et al., 2006; Stephens & Wentz 1998; Lazzar et al., 2007; Vanhala et al., 2009), energy intake (Gordana & Levitsky, 2003; Rodriguez- Artalejo, 2002; Chang & Nayga, 2009), sleep deficiency (Chen et al., 2008; Ozturk et al. 2009; Landhuis et al, 2008; Padez et al., 2009; Touchette et al., 2008), sweetened drinks (Bowman, 2004; Must et al., 2009; Mrdjenovic & Levitsky, 2003) and parental influence (Barradas et al., 2007; Francis et al. 2003; Padez et al., 2009). Parenting styles are also of importance. Baumrinds (1991) parenting typologies/ styles include authoritative, authoritarian, and permissive. Golan & Crow (2004) report that authoritative parenting lead to healthier food choices and higher activity levels. Other studies have found similar results (Patrick et al., 2004; Rutledge et al., 2007; Rhee et al., 2006). 2
Studies evaluating parent-child agreement of behaviors that increase child weight suggest that parents and children do not often agree on child behaviors. Some of these behaviors are fruit and vegetable consumption (Tak et al., 2006), eating styles (Braet et al., 2007), television viewing time (Rossiter & Robertson, 1975), sleep time (Gruber et al., 1997), and parental behaviors (Tein et al., 1994). In all behaviors children reports are not similar to parent reports. The factors above demonstrate the complex social, environmental, and behavioral characteristics that influence a childs risk of becoming overweight. Genetics can play a role as well but the recent obesity epidemic is more likely attributed to changes in environmental interactions rather than inherited risk. Although studies have made connections with eating habits, activity levels, and parenting styles, none to date have connected childrens home environment and compared them with parenting styles and child BMI. The purpose of this study is to examine factors that may explain variability in childrens potential risk for obesity. The study will be conducted using a behaviorally based screening tool called the Family Nutrition and Physical Activity (FNPA) Assessment (Ihmels et al., 2009a). This tool consists of ten items that capture a diverse set of constructs that have been empirically shown to predict risk of overweight. The tool is designed to evaluate the obesigenic nature of home environments and practices and has potential utility for primary prevention. Determining factors that influence FNPA scores will provide valuable information for future research with the FNPA tool. This study will address two specific research questions: 1. Do parents and children provide similar evaluations of the FNPA constructs? 2. Are parenting styles associated with FNPA scores and child BMI? 3
LITERATURE REVIEW Obesity has been a growing epidemic in America for many years. Thirty-four percent of Americas population is obese, compared to the twenty-three percent in 1994 (NCHS, 2007), the prevalence of class 3 obesity (Body Mass Index >40) has risen to 2.2% in 2000 (Freedman et al, 2002) compared to .78% in 1990. Considerable attention has focused on the progression of obesity in Americas youth (YRBS, 2008, NCHS, 2007, Ogden et al, 2006). Risk factors that have been recently researched include: unhealthy eating habits, low physical activity, sleep time, parental influence, video game play and television viewing. Studies have proven that these risks are directly related to overweight children. The rising obesity rates in children highlight the continued need to develop prevention strategies that can be used in family and school environments. Childhood obesity rates have increased consistently each year since 1999 in the United States (Ogden et al., 2006). The ongoing battle between policy makers and snack food companies continue to leave the large amount of unhealthy foods and snacks available to grade school students (Finkelstein et al., 2008). As grade level increases, more unhealthy options are available in schools (Demory-Luse et al., 2004). Other countries report limited healthy options for youth enrolled in grade school as well (Maddah, 2008; Drewnowski & Popkin, 1997). There are many consequences of overweight including negative self-esteem (McCullough et al, 2008), orthopedic problems (bowing of legs) (Must & Strauss, 1999), hypertension (Lauer et al., 1975), and asthma (Velazquez et al., 2008 & Tai et al., 2009). More severe health consequences occur over time if obesity persists. Childhood obesity increases risk of type 2 diabetes (Shaw, 2007), cardiovascular disease (Maddah, 2008), and persistence of obesity into adolescence and adulthood (Freedman et al., 1999). 4
As the BMI of US children increase, so do the risks that were once considered adult diseases. Lifestyle practices such as sedentary behavior, dietary patters, and even parental obesity in early life can determine the risk for obesity later in life (Reilly et al., 2005). Other risk factors contributing to child obesity will be discussed later in this review. A new screening tool, the Family Nutrition and Physical Activity (FNPA) (Appendix B) screening tool, works to assess family environmental and behavioral factors that predict the future risk of childhood obesity (Ihmels et al., 2009b). The FNPA tool includes 21 questions about sedentary activities (video games, television viewing, etc.), eating behaviors, physical activity, as well as family environments. The goal of the FNPA tool is to provide an indicator of modifiable environments that may predispose youth to becoming overweight using the FNPA tool in a new 10 question format. This tool may prove useful in advancing public health efforts but additional research is needed to investigate the uses of the tool. This literature review will review the development of the FNPA tool, factors that have been shown to influence risk of obesity, and patterns of parent-child agreement on these risk factors. Development and Validation of the FNPA
The FNPA tool was developed from constructs in the American Dietetic Association (ADA) Evidence Analyses shown to be predictive of childhood obesity (http://www.adaevidencelibrary.com/category.cfm?cid=7&cat=0). This Evidence Analysis reports the strength of association for various identifiable risks, using a grading system to represent strengths of evidence (1: good, 2: fair, 3: limited, 4: expert opinion only, 5: studies presently lacking). The ADA EA identified ten factors with grades of 2 or 3 and these were used as the basis for the FNPA tool (see Appendix B). These ten factors are: breakfast 5
patterns, family eating, food choices, beverage choices, restriction and reward patterns, screen time, television usage, family activity, child activity, and family bedtime routine. This study provides new information about the utility of the FNPA tool for characterizing home environments and behaviors related to risk of child overweight. The FNPA total score was previously shown to predict risk of children becoming overweight even after controlling for baseline BMI and parent BMI (Ihmels et al., 2009b). The original FNPA tool used a Likert type scale but this is the first study to use a modified version of the FNPA with behaviorally anchored rating (BAR) scales. The BAR items are designed to capture each of the 10 FNPA constructs in 10 single items (See Appendix B) rather than 21 separate items (See Appendix C). An advantage of the BAR format is that it is also more suitable for use in counseling applications. The present study replicated some aspects of the original FNPA design but also addressed several new questions. Other instruments have been developed to measure home environments (Golan & Weizman, 1998, Bryant et al., 2008). The FNPA is unique in capturing diverse aspects of the home environment that may predispose youth to become overweight. Responses from each item are summed to create a score for the participating family. Higher FNPA scores suggest a less obesigenic environment. Lower scores suggest a more obesigenic environment. A recent study (Ihmels et al., 2009a) evaluated the utility of the FNPA for predicting a childs risk for becoming overweight. There were a total of 854 observed families residing in Midwestern U.S. and they were of various ethnicities (White, Black, Latino, and Other), education levels, and income levels. The tool was made accessible online and in paper versions in English and Spanish. Strong correlations between child BMI and parent BMI were reported, showing that children and parents share similar weight status. 6
Of more interest was the association between child BMI and the overall FNPA score. This supported the construct validity of the FNPA tool in assessing obesigenic environments when completed by parents. A follow-up paper by Ihmels et al. (2009b) demonstrated that scores on the FNPA tool predicted BMI change even after controlling for baseline BMI and parent BMI. The table below (Table 1) shows the risk factors addressed in this review, the ADA EA grade, and current conclusions (evidence statements) drawn by the ADA EA. Additional detail of these risk factors will be discussed. 7
Table 1. Derived from the ADA Evidence Analysis Library (except Sleep Deprivation) (http://www.adaevidencelibrary.com/topic.cfm?cat=2792) American Dietetics Association Evidence Analysis Risk Factor & Grade Evidence Statement Energy Intake Grade 2 Total energy (caloric) intake measured using current dietary assessment tools, which may not accurately assess total energy intake does not appear to have a strong association with overweight in children. Sweetened Beverages Grade 2 Intake of calorically-sweetened beverages is positively related to adiposity in children.
Physical Inactivity Grade 2 Participation in regular physical activity is associated with lower adiposity in youth. This association is stronger in boys than in girls. Television Viewing Grade 2 Excessive television viewing is associated with increased adiposity in youth. Video Gaming Grade 3 Excessive use of video games may be associated with increased adiposity in youth. Sleep Deprivation Grade 3 Lower levels of sleep have been associated with increased risk of overweight in several studies. Parental Influence Grade 2 Using family-based counseling including parent training or modeling as part of a clinical intervention treatment program results in significant reductions in weight status/adiposity in children 13 years and younger. The results of studies in older children and adolescents are inconclusive.
8
Risk Factors for Childhood Obesity Obesity is a multi-factorial condition that is influenced by many lifestyle factors. Because genetic risk cannot be altered, it is important to better understand modifiable risks. The following is a review of studies that have made connections between child BMI and these lifestyle factors. Energy Intake Excess Energy intake has been consistently shown to increase risk for overweight. Energy intake can be a choice (e.g., vending machines) or a reflection of a childs environment. Fresh produce available to the child may be dependent upon the neighborhood they reside in or family income (Dammann, 2009). Those communities that do offer healthy options can be located too far from the childs place of residence. Factors such as culture, race and socioeconomic status can play a role in the eating habits of the child (Caprio et al., 2008). A study of four cities with very high child obesity rates assessed the energy intake of fifty children from each school within the four cities, and found that these children had diets that were high in saturated fats, sugar and cholesterol (Rodriguez-Artalejo et al., 2002). Children with poor nutrition also had higher BMI. Foods high in sugar, fats, and oils, that are often found in fast-food options, increases the total energy intake and lowers the nutrient intake of children who eat them (Bowman et al., 2004; Chang & Nayga, 2009). Access to healthier food options at home and in the school can promote consumption. Increased consumption of fruits and vegetables is particularly important as these foods have a protective effect on the body and are important in prepubescent stages (Rockett et al., 2001). Sweetened Beverages 9
Increased consumption of high calorie beverages such as soda and other sweetened beverages, instead of water and milk is thought to contribute to child weight gain. Some studies have linked the intake of sugary drinks and weight gain in youth (Mrdjenovic & Levitsky, 2003). Others have stated that the connection is near zero (Forshee et al., 2008). Stronger evidence for the importance of sweetened beverages can be found from intervention research that has demonstrated reductions in weight of children with high BMI by replacing sugar-sweetened beverages with water or diet beverages (Ebbeling et al., 2006). Physical Inactivity Physical inactivity has been consistently shown to increase risk for overweight. Children who have a sedentary lifestyle are more likely to gain weight (Laurson et al., 2008) and children and adolescents have been consistently less active each year (Healthy People 2000; Powell et al. 2009). Of the nations adolescents, 65% did not meet recommended activity levels and 46% of students did not participate in a physical education class (YRBS, 2007). In particular, the young female population has become less active each year, particularly of Black ethnic background (Pate et al., 2005). School is where children and adolescents spend most of their time; therefore this setting has been commonly recommended as a promising target for intervention. Some school-based interventions have been successful in their attempts to increase physical activity levels of adolescent females (Pate et al., 2005; Flores, 1995) and grade school children (Harrison et al., 2006; Stephens & Wentz 1998; Lazzar et al., 2007), lowering BMI in doing so. Benefits go far beyond a reduction in weight when increased physical activity is achieved as active children are at decreased risk for developing other health risks mentioned above (Obarzanek et al., 1994; Rowlands et al., 1999). 10
Television viewing Studies have been conducted that relate television watching to childrens unhealthy eating or overweight (Cheng, 2005). Most studies have demonstrated positive associations between television and BMI in youth (Laurson et al., 2008; Muller et al., 2002; He et al., 2009). Intervention studies aimed at reducing TV viewing have demonstrated higher physical activity rates and in some cases lower BMI and skinfold thickness (Berkey, 2000; Robinson, 1999). Youth become targets for promotional ads on television, marketing products that are high in sugar during programs intended for children (Byrd-Bredbenner, 1999). The greater the TV time the greater the exposure to unhealthy food options. Television receives a stronger relationship to food intake when the family begins to eat and snack in front of the television (Francis et al., 2003). Macfarlane et al. (2009) conducted a longitudinal study and found that children who ate dinner while watching television were more likely to have a higher BMI than children who did not watch television during dinner. Not only does limiting television lower the absorption of unhealthy ads in youth, it also increases available time to be used for physical activity. Video Games Computer and video game play, like television viewing, is a sedentary activity that has consistently increased in popularity among grade school children. Studies have shown that there is a connection between video game play and weight gain in youth (Skinner et al, 2004). Some have shown that boys are more susceptible to the risk of weight gain, because girls were less likely to play video games, or played for shorter times (Laurson et al, 2008). 11
The introduction of new interactive video games (ie: Dance Dance Revolution, Nintendo Wii, Sony Eye Toy) may create a change in the aforementioned results (Maddison et al., 2009), due to the player using physical activity to interact with screen images. Graf et al. (2009) found that childrens energy expenditure, heart rate, and perceived exertion were higher during walking, playing DDR, and playing Wii games than during television viewing. Energy expenditure was highest during DDR-2 and Wii Boxing, when compared to walking at 5.7 km/hour. McDougall & Duncan (2008) found that not only did children enjoy playing the Sony Eye Toy interactive video games but understood and attained the positive health results from the activity. Increases in physical activity were also identified with the Nintendo Wii in adults and children, children having higher activity levels than adults (Lanningham- Foster et al, 2009). Further research is needed to determine if there can be direct benefits of using other varieties of interactive video games rather than non-interactive television viewing. Activity gained through gaming is not meant to replace normal physical activity. Sleep deprivation Sleep recommendations for children ages 5-12 years are 9-11 hours each night (CDC: http://www.cdc.gov/sleep/how_much_sleep.htm). A rise in the risk of obesity occurs when children get less sleep. A longitudinal study by Landhuis et al. (2008) found that shorter amounts of sleep during childhood are associated with higher BMI in childhood and into adulthood. Other studies have found similar results and noted that at least ten hours of sleep is needed for children (Chen et al., 2008, Ozturk et al. 2009, Landhuis et al, 2008, Padez et al., 2009; Touchette et al., 2008). Padez et al. (2009) found that children of parents who have low education levels sleep less. Further, children who are less physically active and watch more television sleep less. The connection between shorter sleep time and increase in weight 12
is hypothesized to be due to lack of hormonal development time provided for the child in relation to energy expenditure and eating habits prepared in the brain (Taheri, 2006; Sekine et al., 2002). Even stronger relationships have been found for younger children; sleep deprivation in early childhood creates a greater risk of obesity (Reilly et al., 2005; Sekine et al., 2002). Parental influence
Parenting involves a great responsibility in limiting and promoting behaviors in children. Behaviors studied include activity habits, diet, sleep and television viewing time. Parents not only can control these practices, but they are also role models. Most children mimic or repeat the activities of their parents (Francis et al., 2003), making the parent a direct influence on the childs adoption of lifestyle habits. Parental influence over eating in children is complex. Although parents are the gatekeepers to the meals provided to the child, there can be changes in the childs preference for meals depending on the rigidity of the parents routines involving food (Costanzo, 1985). Robinson et al. (2001) alternatively found that girls whose parents have greater control over their childs eating regimen had daughters with lower weight. Parental feeding restriction has been shown to be directly associated with child BMI (Joyce & Zimmer-Gembeck, 2009).This topic needs to be researched further to attain more recent and specific results in relation to culture, race, and SES. A longitudinal study associated parents BMI with their childs BMI, snacking habits, and TV viewing (Francis et al., 2003; Whitaker et al., 1998). Both overweight parents and their overweight children had an increase in BMI over the span of the study, along with increases in snacking and television viewing time. This comparison between parent and child 13
weight in relation to television viewing time was looked at further by Barradas, et al. (2007) showed that parental rules limiting childrens TV viewing time was more common in children whose parents were not overweight. Parental sedentary levels are also a risk for a childs TV viewing time, an inverse relationship when controlling for income level. A child having a TV in the bedroom is at increased risk of being overweight (He et al., 2009). Other characteristics such as parent education level, income level, and perceived neighborhood safety are inversely related to child BMI, with television viewing identified as the primary mediator (Cecil-Karb & Grogan-Kaylor, 2009; Morgenstern et al., 2009). Parenting Style Parenting styles are also of importance in the lifestyle factors and BMI of children. Baumrind (1966) created three original parenting typologies: authoritative (provide clear direction and warm discipline), authoritarian (have clear authority over the child and expects unquestioning obedience), and permissive (allow child to make their own decisions and provide minimal punishment). The introduction of the fourth typology was done by Maccoby & Martin (1983). The fourth typology, uninvolved, characterized parents who had traits of permissiveness and provided the child little warmth. Parents who were uninvolved did not demand much from the child and had little control over the childs actions. Figure 1 exhibits the two-by-two representation of the parenting styles and characteristics.
14
High expectations for self-control Low expectations for self control High Sensitivity Authoritative: respectful of childs opinions, but maintains clear boundaries Permissive: indulgent, without discipline Low Sensitivity Authoritarian: strict disciplinarian
Neglectful/ Uninvolved: emotionally uninvolved and does not set rules Figure1. Parenting Styles (Rhee et al., 2006) Golan & Crow (2004) report that authoritative parenting lead to healthier food choices and higher activity levels in children, whereas children of permissive parents had poorer diets and less restriction to unhealthy food intake. Other studies have found matching results (Patrick et al., 2004; Rutledge et al., 2007; Rhee et al., 2006). Parenting styles have been compared to demographic information. Most studies, like those mentioned above, comparing parenting style is on predominantly White, middle class samples. Studies like these have results that show White parents have more authoritative parenting styles and Latino have more authoritarian parenting styles (Chaundhuri et al., 2009). Studies particularly studying parenting styles of low-income samples have found that these parents have more authoritarian typologies, regardless of ethnicity (Pinderhughes et al., 2000). This supports the strength of influence income has on parent stress, control, and sensitivity. Other factors that are determined by income such as number of hours parents worked (Morawska & Sanders, 2007) and low SES neighborhoods (Roche et al., 2007) are related to parenting styles. Chaundhuri at al. (2009) explain further that parents of low SES use authoritarian parenting styles to prepare their children for blue-collar or industrial work. Parents of high SES use authoritative parenting styles to prepare their children for white- collar positions where negotiation and creativity is needed. 15
The Parenting Styles and Dimensions Questionnaire (PSDQ) (Robinson et al., 1995) utilize Baumrinds (1966) original three typologies, not including Maccoby & Martins (1983) typology of uninvolved parenting style. Studies have used the PDSQ in various ways; some include relating parenting style to childrens attachment, temperament, and school adjustment (Coplan et al., 2008; Coplan et al., 2009; Karavasilis et al., 2003). Blissett & Haycraft (2008) found no correlation between parenting styles and child BMI, using the PSDQ, but did find relations between parental pressure to eat and parental drive for thinness and lower child BMI. Permissive parenting and childs temperament increased the odds of the child being obese (Zeller et al., 2008). Hubbs-Tait et al., (2008) conducted a similar study and discovered that parental feeding practices can predict parenting styles. Parent and Child Agreement Studies have indicated that certain issues can lower parent-child agreement (Grills & Ollendick, 2002) but few have studied agreements in factors that are related to child weight gain. Tak et al. (2006) studied parent-child agreement on fruit and vegetable consumption. Children in this study reported higher values than did parents at baseline and one year follow-up. Low consumers of fruits and vegetables had higher parent-child agreement than those who reported high consumption of fruits and vegetables. Tak et al. (2006) suggest that child participants (elementary grade 4) may have overestimated their fruit and vegetable intake. Reported eating styles of overweight children and their parents are studied by Braet et al. (2007), who finds that parents rated the emotional and external eating of their overweight children higher than the child. Parents produced lower ratings of restrained eating than their child counterpart. Agreement was lowest for children under the age of ten. The low parent- 16
child agreements propose that parents and overweight children attribute their reasons to eating differently. Television viewing times reported by parents have shown to underestimate their childrens reports (Rossiter & Robertson, 1975). In this study parents rating of household rules, television co-viewing with child, parent-child interactions were all also higher than their child counterparts. Parent and child report differences increased with socioeconomic status. These results suggest a higher bias of social desirability for more affluent families. Significant differences in parent and child reports were found in relation to estimated child sleep time (Gruber et el., 1997). Children estimated sleep time more accurately than did parents. Goodwin et al. (2007) compared parent reports of child sleep time to polysomnography output to determine accuracy of parent reports. Parents were found to overestimate the habitual total sleep time of their child. Hispanic parents significantly less than Caucasian parents. Parental behaviors (control, discipline, acceptance, and rejection) have presented low parent-child agreement (Tein et al., 1994) and children rate parental behaviors of mothers and father similarly. Suggesting that if either of the parents practices a behavior of warmth or control, children perceive the behavior is directed from both parents. In this study children older than age 10 disagree with mothers on degree of discipline. The aforementioned studies all support that children and their parents differ on reports of behaviors that have all been related to child weight gain. Grills and Ollendick (2002) suggest possible issues with parent-child agreement including miscommunication between parties, family conflict, child age, child and parent social desirability, etc. 17
Regardless of the reasons behind the difference in parent and child reports the consistent result is that they do not usually correspond.
Conclusion
The risk factors for obesity in children include high energy intake, physical inactivity, television viewing, sleep time, and parental influence and parenting style. Each of the risk factors has their own influence on potential risk of weight gain. Parent-child agreement is also very low in reporting of these risk factors. The FNPA tool is a potentially useful screening instrument that can evaluate the likelihood of a child being overweight prior to weight gain This project will advance research on the FNPA tool by evaluating whether child ratings (scores) of the home environment are similar to parents ratings. If there is an agreement between parent and child FNPA scores, it can help identify ways to promote healthier environments in the home. Differences in FNPA scores can shed light on the environmental and behavioral factors that could be perceived differently by parents and children. Another objective of this study is to identify parenting styles that may be associated with ratings of home environment, identified by the FNPA tool. Comparing these two factors can help determine a childs risk for obesity increases according to their parents typology. BMI is also of interest in relationship with parenting styles. This finding would support the predicted risk determined by the FNPA tool. These questions will be examined in this research study. 18
METHODS
Participants The sample of this study includes a total of 313 elementary grade students (155 males, 158 females) and 77 of their parents (83%, Mothers) from a low socioeconomic status Midwest elementary school in the state of Iowa. The surveys were administered to the students during their physical education class. All students grades K-5 were welcome to participate, 27% of students opted out of completing the survey. BMI data were collected from all students (average age of the elementary students was 8.56 +1.73 (5-11years). Descriptive statistics for BMI percentile are provided in Table 2 to characterize the prevalence of overweight in this sample. The mean BMI for students was 19 +4 kg/m 2 and this corresponded to an average BMI percentile of 68.3% +28.3. Based on the standard CDC definitions, approximately 61% of participants were normal weight, 18% were at risk for overweight (85 th -95 th percentile) and 21% were classified as overweight (>95 th percentile). The distributions for males (48%, 22%, 26%) were slightly different than females (64%, 15%, 17%). The average BMI percentile for youth with parent FNPA data was 65.4 % (males: 68.1%; females: 64.1%). Free and reduced lunches are provided to 80.1% of students. Data on FNPA were obtained from 312 children involved in the project. The goal of this study was to obtain corresponding FNPA scores from each parent but surveys were returned from only 73 parents (25%). Most parents (48%) of students were between 30-40 years of age; 55% of the parent sample was White followed by 17% Hispanic, 10% Black, 8% Asian or Pacific Islander, and 9% Multi-racial. The majority of parents (50%) had income levels below $25,000 per year, 35% had income between $25,000-50,000 per year, 19
7% had income between $51,000 75,000 per year, and 8% had income greater than $75,000 per year. The 85% of families with income below $50,000 per year and 80% of children on free and reduced lunch support this sample being classified as a low SES. Parent education level varied; 14.3% completed some high school, 26% completed high school, 37.7% completed some college, 22% completed college. Table 2. Descriptive data for children by gender and grade Male Female N Mean BMI % SD N Mean BMI% SD All Grades 155 72.9 27.2 158 63.6 28.5 Grade 0 24 71.7 29.4 20 67.6 23.6 Grade 1 35 70.9 25.0 19 54.7 32.3 Grade 2 20 72.8 24.6 34 71.8 28.7 Grade 3 31 64.5 30.4 28 66.2 27.1 Grade 4 21 82.8 21.3 26 58.6 30.2 Grade 5 24 74.5 32.6 25 62.6 28.9
Procedures The sensitive nature of the study necessitated the use of procedures to ensure confidentiality of the participants. Specifically, the study required merging childrens self- report data (and BMI) with parents self-report data in a confidential manner. To accomplish this, student ID numbers from class lists were pre-printed on both a student survey and a separate parent survey form (with informed consent form attached). Directions on how to complete the FNPA survey were provided to the students and each question was read aloud 20
by trained testers to assure understanding for the elementary students. Students had their height and weight measured by trained testers and then completed the FNPA assessment during a separate class period. This took most students about thirty minutes to complete. A TANITA BF-681W (Tokyo, Japan) scale was used to measure weight and a SECA Road Rod stadiometer (Hanover, MD) was used to measure height. The height and weight data were recorded on a form along with student ID to facilitate tracking and computation of BMI. After students completed the testing they took parent packets home for parents to complete and return to school. Two weeks after the first distribution of parent packets (parent FNPA and PSDQ), there was a second distribution for those yet to return the first. This study was approved by the Iowa State University Institutional Review Board.
Instruments Family Nutrition and Physical Activity (FNPA) assessment - Adult Version The FNPA was used to collect detailed information on home environments and behavior related to overweight youth. The FNPA measures ten risk factors (constructs) associated with overweight/obesity in children: (1) breakfast patterns, (2) family eating, (3) food choices, (4)beverage choices, (5) parental restriction and reward, (6) TV/ video game/ computer screen time, (7) TV usage, (8) family activity, (9) child activity, and (10) family bedtime routine. One question in a BAR format is assigned to each of the aforementioned constructs, creating the 10 items. Each item is asked in terms of frequency (usually, often, etc.). Responses have a minimum of 1 point (most obesigenic) and a maximum of 3 points (least obesigenic). Responses to the 10 items are summed for a total score. The maximum total score of 30 would be the healthiest home environment and the minimum score of 10 21
would be the home environment with a high risk for obesity. A Spanish version of the FNPA Adult Version was created for the large proportion of parents whose first language was Spanish. The constructs are based on factors shown in a comprehensive evidence analyses to be predictive of child overweight. The total score reflects the overall obesigenic nature of the home environment; low scores indicate a more obesigenic (high risk) environment while high scores are indicative of healthy (low risk) environments. Previous research has supported the construct validity (Ihmels et al., 2009a) and predictive validity (Ihmels et al., 2009b) of the FNPA tool. FNPA - Child Version The phrasing on the Adult Version of the FNPA tool was modified to fit an elementary reading level. A pilot testing of the child version was conducted on children ages 5-14 from varying school districts. Students were able to understand each of the items when read aloud to them. The same basic structure of the parent version was retained in the child version. The FNPA Youth Version uses the same constructs as the aforementioned Adult Version with the same scoring concept (Appendix D). Parenting Styles and Dimensions Questionnaire (PSDQ) The PSDQ is designed to characterize parenting styles of preschool and school-age children (Robinson et al., 1995). The tool was designed around Baumrinds (1966, 1991) three main typologies (authoritative, authoritarian, and permissive) but there is no specific classification scheme available to determine a predominant style. The instrument includes 58 questions scored on a 1-5 scale and the items are clustered into different stylistic dimensions which are then aggregated to create separate scores for each of the three typologies. 22
Reliability of the individual PSDQ scales ranged from .91-.75 (Robinson et al., 1995). Questions such as I ignore our childs misbehaviors help to determine parents permissiveness; other questions such as I demand for our child to do things help to determine parents authoritarian styles. Different stylistic dimensions are used to characterize the three parenting typologies. Stylistic dimensions of permissiveness include lack of follow through, ignoring misbehavior, and self-confidence. Authoritative stylistic dimensions include warmth and involvement, reasoning/ induction, democratic participation, good natured/ easy going. Authoritarian stylistic dimensions are verbal hostility, corporal punishment, non-reasoning/ punitive strategies, and directiveness. The number of items in each stylistic dimension varied to mean scores. These were first computed for each dimension. The total composite score for each parenting typology was determined by computing an average of each of the associated stylistic dimensions. This weights each stylistic dimension equally rather than basing the overall typology on the mean of all associated items. Four questions were removed from the original questionnaire to avoid reporting types of corporal punishment. Data Analysis The data analyses for the present study were conducted using SAS. The student BMI and FNPA data were first merged with the parent FNPA data by school ID. Demographic data obtained from the school district were then merged into the dataset (by school ID) to create the final dataset. Student BMI, an outcome measure presents some challenges when used to summarize data across different age groups of children. To provide a common indicator it was necessary to convert the BMI data into standardized scores. Specialized SAS programs available from the CDC were used to convert the student BMI data into BMI 23
percentiles (BMIpct) and BMI z-scores (BMIz). These were used to categorize students into weight categories and for all subsequent analyses. Data analyses were conducted to answer the two primary research questions in the study.
Research Question 1: (Agreement between parent and child FNPA scores) One key research question was to determine whether a child version of the FNPA scores yielded similar results. The revised format of the FNPA uses a different format (behaviorally anchored rating scores) than the original (Likert). Therefore, it was important to first evaluate the reliability of this new format (for both children and parents). Cronbachs Alpha was calculated to measure internal consistency of the 10 item format of the parent and child FNPA. Because age and gender may influence the childrens responses the alpha reliability of child responses were examined separately by age group (younger: K-2; older: 3- 5). Several analyses were conducted to evaluate agreement between parent and child scores. Pearson product moment correlations were computed among the parent and child FNPA scores to examine overall agreement between the two measures. Correlations were examined separately by age and gender to test for any differential relationships. Differences between the parent and child total FNPA scores were examined with a simple t-test. Differences in individual items were also examined to determine whether there was variability in parent and child perceptions of specific home environments or practices. These relationships were also examined separately by age and gender. The primary determinant of agreement was whether they possessed similar correlations with measured BMI. Comparing the associations between the parent and child versions make it possible to see if associations 24
were similar for the parent and child version. These correlations were also examined separately by age group and gender. Collectively, these analyses are designed to determine whether parental and child evaluations of home obesigenic environments are similar and whether they are equally effective at predicting risk of overweight.
Research Question 2: (Relationship between Parenting Style, FNPA scores, and child BMI) The second goal of these analyses was to determine whether aspects of parenting style (based on the PSDQ scales and Baumrinds typologies) were associated with child BMI and parent FNPA scores. The computed scores for each typology provide an indicator of the overall score for each person but the goal in this study was to determine if certain parenting styles are associated with high or low scores on the FNPA. Therefore, it was important to try to characterize typologies on a more relative basis. To accomplish this parenting typologies were converted into Z-scores. This allowed the value for each typology to be interpretable relative to others in the sample. Parents can then be classified as more or less authoritative, more or less permissive, or more or less authoritarian. It is possible for parents to score high on some typologies but low on others. Figure 2 shows a two-by-two representation of possible styles that Baumrinds typologies do not include, represented by the shaded areas. Amounts of control and warmth can vary in a parent and dimensions are not mutually exclusive. In other words, it is possible to score high (or low) on each of the typologies. To examine this, a total of 8 distinct groups were created based on whether parents scored above the mean (+z) or below the mean (- z) on the individual scales. The 8 groups represent all possible combinations of these three variables. The characterizations of the 8 groups are summarized in Table 3. Frequencies of 25
classification were computed to determine the relative distribution based on this classification scheme. Mean FNPA scores and BMI percent scores for each group were then compared using a simple ANOVA to examine possible differences in groupings.
A limitation of this approach is that it forces categorization using arbitrary cutpoints. Therefore, additional analyses were conducted to examine the patterns in more detail. Cluster analysis was employed to determine if there were any natural distinctions or patterns in the data. This empirically derived approach may be useful for identifying patterns in the PSDQ classifications that may be associated with particularly high or low FNPA scores. The cluster analysis used Wards method with squared Euclidian distance (maximize cases further apart and therefore increasing the differences between clusters, viewed as the most popular method), and the cluster solution was defined to identify the most appropriate number of clusters based on the visual illustration of the resulting dendogram (see Appendix A: Figure 13). 27
Regression analysis was also used to determine if parenting styles were related to FNPA and child BMI. The value of the regression approach is that it allows the relative weight of each of the individual factors to be empirically determined. Separate regression analyses were conducted for FNPA and BMI percent outcome measures. 28
RESULTS Descriptive Results To characterize the sample population, the outcome variables (BMI percent and FNPA score) were stratified by demographic variables. Figure 3 shows the difference in child BMI percent by Ethnicity Group, Figure 4 shows the difference in FNPA score by Ethnicity Group, Figure 5 and 6 shows the same variables stratified by Income Level. Due to inconsistent results, parents who classified as Multi-Racial werent included in this analysis. The results show no significant relationship between child BMI percent and Ethnicity. White parents reported the highest FNPA score. No association was found between parents income level and the childs BMI percent. No association was found when parent FNPA scores were stratified by income level. The sample sizes for income levels higher than $50,000 were very small. This prevents generalizations about these patterns to higher income levels so results are intended to be used only for descriptive purposes. BMI percent by Ethni ci ty 30 40 50 60 70 80 90 100 110 1 2 3 4 Ethnicity Gr oup B M I
-
p e r c e n t
( m e a n )
Figure 3. Child BMI percent stratified by Ethnicity Group (1: Asian or Pacific Islander, 2: Black, 3: White, 4: Hispanic) 29
Parent FNPA by Ethni ci ty 22 23 24 25 26 27 28 29 1 2 3 4 Et hnicit y Gr oup P a r e n t
F N P A
S c o r e
( m e a n )
Figure 4. Parent FNPA scores stratified by Ethnicity Group (1: Asian or Pacific Islander, 2: Black, 3: White, 4: Hispanic)
BMI per cent by Income 50 60 70 80 90 100 110 1 2 3 4 Income Level B M I
p e r c e n t
( m e a n )
Figure 5. Child BMI percent stratified by Income Level (1) <$25,000, (2) $25,000- 50,000, (3) $51,000 75,000, (4) > $75,000
30
Parent FNPA by Income 20 21 22 23 24 25 26 27 28 1 2 3 4 Income Level P a r e n t
F N P A
( m e a n )
Figure 6. Parent FNPA scores according to income level: (1) <$25,000, (2) $25,000- 50,000, (3) $51,000 75,000, (4) > $75,000
Results for Research Question 1: Agreement between Parent and Child Versions. Cronbachs alpha was first used to examine the internal consistency of the 10 parent and child FNPA dimensions. The alpha reliability score for the parent FNPA was a =.68 while the value for the child FNPA was a =.71. According to Nunnaly (1978), values above 0.70 are indicative of acceptable internal consistency. While the values for parents were a bit lower it is possible that this is due to the smaller sample. The results indicated that the deletion of any single variable would not improve the alpha coefficient. This indicates that the scale is most internally consistent with all items included. The reliability test was taken a step further by stratifying child FNPA responses to age groups. Results showed low reliability for the younger sample (a =.58) than for the older sample (a =.76). This suggests that younger children may not be able to provide internally consistent responses on the FNPA. 31
Correlations were used to examine the relationships among the key main outcome measures. The correlations between parent and child FNPA scores were low (r =.19) and not significant (p =.13), indicating relatively poor agreement between these indicators. The correlation between parent FNPA scores and the students BMIpct score was analyzed. The correlations were variable but consistently negative for most grades (ranging from r =-.16 to r =-.66), with the exception of a somewhat spurious finding for fourth grade youth (r =.14). The data for this age group were examined in more detail and three outliers were detected that may have caused the discrepant value. When these were removed from the group of fourth graders being analyzed, the correlation for this age group increased (r =-.17), making the composite correlation across all grades strengthen to statistical significance (r =-.31). Correlations between the child FNPA score and child BMIpct scores were considerably lower. The correlations ranged from r =.03 to r =-.26 across grades with the composite correlation (based on all students) of r =-.08. Correlations were also examined among the various items in the FNPA tool (See Table 4). The correlations with BMIpct for individual items ranged from r =.03 to r =-.28. Interestingly, the highest correlations with BMI were found for breakfast eating patterns and sleep patterns items not typically thought to have a direct impact on overweight. The correlations of the individual items with the parent FNPA total score ranged from r =.44 to r =.57. In all cases, each item was more strongly correlated with the total parent FNPA score than with any other item (see Table 4). Child FNPA correlations were mostly non-significant r =-.06 to r =.19, other than TV usage (r =.26). 32
Table 4. Correlations between FNPA factors, parent, and child FNPA totals *p<.05, **p<.01
Brkfst Family Eating Food Choice Beverage Restr & Rewrd Screen Time TV Usage Family Act. Child Act. Bed time Pttl Cttl BMIperct 0.28* -0.01 -0.10 -0.01 0.03 -0.10 -0.17 -0.06 -0.06 -0.26* -0.27* -0.08 Brkfst .28* 0.21 -0.01 .25* 0.12 .29* 0.08 0.06 -0.25* 0.47** 0.19 Family Eating .33** 0.19 0.22 .28* 0.21 0.09 0.06 0.23 0.56** 0.17 Food Choice 0.04 .35** 0.00 0.07 .288* 0.02 0.34** 0.51** 0.01 Beverage 0.12 0.14 0.22 .256* .26* 0.14 0.48** -0.06 Restr & Rewrd 0.22 .39** 0.08 .27* 0.18 0.57** 0.08 Screen Time .42** 0.04 0.04 0.21 0.49** 0.10 TV Usage 0.16 .24* 0.03 0.59** 0.26* Family Act. 0.20 0.21 0.51** 0.06 Child Act. 0.12 0.44** 0.19 Bed time 0.51** 0.01 Pttl 0.19
Further investigation was conducted according to age group: 1 (K-2 nd grade) and 2 (3 rd -5 th grade) and gender (male, female). Group 1 child FNPA correlations with BMI percent were not significant (r =-.12). Group 1 parent FNPA had a significant correlation with BMI percent (r =-.54). Group 2 child FNPA correlations with BMI percent were even lower than group 1 (r =-.04), parent FNPA shows a similar trend (r =-.15). Parents of males (r =-.32) and females (r =-.23) had correlations with BMI. Parents FNPA of group 1 females had a significant correlation with BMI percent (r =-.55). Parents of group 2 females did not have a significant correlation (r =-.07). Both groups 1 and 2 females had low correlations with BMI percent (r =-.10, r =-.13). Parents FNPA of group 1 males had a significant correlation with BMI percent (r =-.62), parents of group 2 males did not (r =-.22). Both groups 1 and 2 33
males had low correlation with BMIpct (r =-.17, r =-.07). Parent FNPA correlations were stronger for male children, especially those in group 1. Direct comparisons were also made to determine if the parent and child FNPA scores were significantly different from each other. The overall score was not significantly different (t =1.56 (64), p =.13). Additional tests were made for individual items and it was clear that there were large disparities for individual items. Difference scores were computed for each item (parent - child) so items with positive differentials reflect items that parents rated as more healthy (less obesigenic) than children (see Figure 7). Significant positive differentials were found for items 2 (t =2.82 (68), p =.006), 4 (t =2.12 (68), p =.038), 5 (t =2.28 (67), p =.026), and 7 (t =6.2 (68), p <.001). A significant negative differential was found for item 10 (t =-3.07 (68), p =.003).
Figure 7. Difference scores for parent and child FNPA reports for each factor (parent score child score) *p <.05, **p <.01, p<.001
Parent and Child FNPA Difference Bed Time** Child Act. Family Act. TV Usage Screen Time Rest. And Reward* Brkfst Patterns Family Eating** Food Choices
Beverage Choices* -0.4 -0.2 0 0.2 0.4 0.6 1 2 3 4 5 6 7 8 9 10 F N P A
F a c t o r s
Difference Score 34
Stratifying differentials by age and gender produced more extreme differences on several variables. The difference between parent and young child ratings of screen time was significant (t =2.7 (23), p =.013), as was parent and females ratings of screen time (t =3.92 (11), p =.002). Older childrens television usage (t =5.38 (44), p <.0001) was also found to be significant. Results for Research Question 2: Parenting Style, FNPA scores, and child BMI The PSDQ was used to characterize parents into 8 different groups. There was considerable variability in responses to the PSDQ and this led to clear distinctions in the parenting style groups (See Table 5). Table 5. Description and Frequencies of Parenting Style Groups Parenting Style Groups Group Description n = 1 Parents with low responses to all parenting styles 12 2 Parents who are overall more permissive 7 3 Parent who are overall more authoritarian 5 4 Parents with stylistic dimensions in both authoritarian and permissive 14 5 Parents who are overall more authoritative 17 6 Parents with stylistic dimensions in both authoritative and permissive 3 7 Parents with stylistic dimensions in both authoritarian and authoritative 5 8 Parents with stylistic dimensions in all parenting styles 14
A cluster analysis was used to group parent FNPA z scores according to parenting styles. Based on the dendogram (see Appendix A: Figure 13), three clusters were found to 35
categorize parent responses according to reported stylistic dimensions of authoritative, authoritarian, or permissive parenting style (see Figures 9 & 10). One of the clusters (cluster 2) was not used because parents within it rated high in all three parenting styles; this report of equal typologies is uncommon with both the PSDQ (Robinson et al. 2001, Coolahan et al., 2002) and the adapted version of the PSDQ (PBQ: Hart et al., 1998). The other two clusters were clearly interpretable so these were focused on in the analyses. Authoritarian and permissive parenting styles are considered to be less effective parenting styles than authoritative. Therefore a low amount of permissive or authoritarian ratings would be expected to be associated with better home environments (higher FNPA scores). High scores on these variables would be expected to have the opposite effect (lower FNPA scores). Consistent with this expectation, cluster 1(Figure 9) shows that parents who were less authoritative (and more permissive / authoritarian) had lower scores on the FNPA (indicative of more obesigenic environments). In contrast, cluster 3 (Figure 10) shows that parents who were more authoritative had higher FNPA scores (less obesigenic environment). The clusters exhibit the expected relationship but it is not possible to determine if the presence of a favorable trait (authoritative) is more important than the absence of a more negative one (e.g. permissiveness or authoritarian) 36
Cluster 1- Low Parent FNPA -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 Authoritarian Authoritative Permissive P a r e n t i n g
S t y l e
z - s c o r e
Figure 9. Parenting style trends of less obesigenic parent FNPA scores Cluster 3- High Parent FNPA -1.5 -1 -0.5 0 0.5 1 Authoritarian Authoritative Permissive P a r e n t i n g
S t y l e
z - s c o r e
Figure 10. Parenting style trends of more obesigenic parent FNPA scores
Regression analyses were conducted to determine the relative importance of the three parenting typologies on parent FNPA scores and child BMI percent. Regressions illustrated that authoritative parenting (t =3.66 (30), p =.001) was the only significant predictor of obesigenic environments. Authoritarian (t =-.02 (30), p =.985) and Permissive (t =-.20 (30), p =.841) were not. Parenting styles were not significant predictors of child BMI percent. 37
DISCUSSION This study provides new information about the use of the FNPA tool. One goal was to determine whether parents and their children provide similar ratings of their home/obesigenic environments as assessed by the FNPA. Both the child and parent versions were found to have reasonable internal consistency but clear age related differences were noted for the child version. Reliability was good for the older sample but not for younger samples. These results show that older children are more reliable in rating their home environments than younger children. The results support the utility of the FNPA for adults and older children but not younger children. It may be that as children transition into preadolescence they are more aware of their behaviors and have greater control in making some decisions about lifestyle habits. Correlations were low between the child and parent versions of the FNPA tool suggesting poor agreement. The actual test of differences in mean FNPA values was not statistically significant (suggesting parent-child agreement) but there is wide variability in parent and child responses on individual items. The score differentials in evaluations between parent and child were greatest when determining TV usage, family eating, and bedtime routines. Parents underestimated the child in TV use and frequency of fast food consumption in front of a television. This result confirms Rossiter & Robertsons (1975) study that suggested parents reported significantly lower television viewing time than their children. Social desirability is a determining factor in parent reports of their childs television viewing time and this may explain the results. Parents provided lower reports of sleep time than the child but this does not agree with findings from other studies. Goodwin et al., (2007) compared child and parent reports of sleep time to actual sleep time and found that parents 38
reported longer sleep times than the child reported. A possible explanation to parents underestimation of child sleep times could be that when completing the survey/questionnaire the parent is considering just the bedtime of the child and assumed hours of rest. There may also have been a misinterpretation of the recommended time span of sleep for children due to the BAR format of the questionnaire. If parents are less aware of this, social desirability may have influenced parents responses. Unlike other studies (Goodwin et al., 2007; Gruber et al., 1997), there is no comparison of reports to actual measured sleep times in children to determine who is a more reliable source on the subject. Taken together, results suggest that children and parents do not provide similar ratings of their home environment. Age and gender of the child was looked at as a possible reason to the difference and what was found was that parents of males in K-2 nd grade had the highest correlation with BMI percent. Younger children were more predictive of BMI using the child FNPA. These results could support the idea that older children have more autonomy than younger children. Parents may not know all meals that the child intakes and television that is watched. Parents of older children may also use the television as the babysitter and assumes the childs behaviors. Regardless of the cause, the results here suggest that children do not provide similar evaluations to parents. Differences in the accuracy of FNPA reporting can be inferred by examining the correlations with child BMI. The correlations between parent FNPA and BMI percentile were considerably higher (r =-.31) than values between the child FNPA and BMI percentile (r =.02). While the values are low in absolute terms, the values for the parent FNPA are larger than the values reported by Ihmels et al. (2009a) using the original Likert scale (r =- .173). The Ihmels study demonstrated predictive utility of the FNPA for detecting risk of 39
overweight so these relatively low correlations are still noteworthy. The correlations with the BAR format suggests that this format may work as well or better for parents characterizing home environments but the small sample makes it hard to draw too many conclusions. A second goal of the study was to determine whether parenting styles relate to parent and child FNPA scores. An interesting outcome from the study was the finding that FNPA scores were related to parenting styles. Parents who were more authoritative rated their homes as less obesigenic. Parents who were less authoritative rated their homes as more obesigenic. Previous studies (Rhee et al., 2006; Rutledge et al., 2007) have reported that permissive parenting was associated with an increased risk of overweight, relative to authoritative parenting. Authoritative was the single parenting style predictive of parent FNPA scores in the regression analyses. Several studies (Golan & Crow, 2004; Kremers et al., 2003; Schmitz et al., 2002) have found that authoritative parenting styles are associated with positive health behaviors such as child fruit and vegetable consumption and child physical activity levels. The results from the cluster analyses confirmed that parenting styles had an impact on the FNPA scores as clear in differences in cluster 1 and 3. Interestingly, regression analyses revealed that none of the parenting styles were significant predictors of child BMI. Other studies have found similar results (Blissett & Haycraft, 2008; Hennessey et al., 2010, Hughes et al., 2008). A number of other variables have shown to be significant predictors of child BMI. Parent BMI for example has been found to be a strong predictor of child weight gain (Francis et al., 2003). This suggests a genetic component may have a stronger (over-riding) impact on BMI than the environment although this study cant determine this. 40
This study differs from previous research by using parenting styles and finding relationships between them and obesigenic environments in children by using the FNPA tool. Blissett & Haycraft (2008) has used the PSDQ and found no relationship between parenting styles and child BMI. The null findings in the previous study may be due to the use of individual items such as restricting feeding styles. An advantage of the FNPA tool is that it captures multiple dimensions of the home environment. As previously described each item exhibited the highest correlation with the parent FNPA total score findings comparable to the results found by Ihmels et al. (2009a) see Table 4 and Appendix A: Table 6). This similarity suggests that the parent responses to the FNPA collectively capture the overall obesigenic nature of the home environment. The associations with parenting style are interesting but further work is needed to determine the best methodology to categorize parents into a single parenting style group. In the validation literature, (Robinson et al., 1995) detail was not provided in how the stylistic dimension responses translate into definitive parenting styles, resulting in eight categories instead of three. The cluster analysis was essential in determining typologies of parenting according to home environment ratings. The PSDQ has been used to measure parenting styles in relation to attachment (Karavasilis et al., 2003), shyness (Coplan et al., 2008), temperament (Coplan et al., 2009), and feeding practices (Hubbs-Tait et al., 2008). This study is the first to compare 10 factors shown to be significant in child weight gain with parenting styles, finding that parents who are less authoritative have children with a higher risk of obesity. Limitations to this study include a low response rate from parents of children who participated. This may be due to the large percent of low income families; there may be less 41
time available or less perceived incentive to complete the survey. This low response rate would limit the generalizability of the results. Further research would be needed to find if these results are plausible across all SES levels and minority groups. Another limitation would be the possible bias reporting from parents, some questions may seem probing or personal and parents may answer falsely, if at all. Future questions to be answered are whether parenting styles are correlated with demographic information such as age, education level, etc. Another question may be whether child FNPA factors such as consumption of sweetened drinks and physical activity level are related to parenting styles. Longitudinal data would also be of value in extension to this cross-sectional analysis; patterns of control may vary over time in parents of young children. Answers to these questions would be beneficial to programs working to lower childhood obesity rates by incorporating of the family and home environment. 42
CONCLUSION This study provides information on the differences between parent and child perceptions of obesigenic environments and can be helpful in a variety of intervention settings. In this low-income sample parent-child agreement is low. Parents were better at rating home environments for younger children. Older children were better at rating home environments than younger children. Parents who are more authoritative rate their environments as less obesigenic. Authoritative parenting was the best predictor of FNPA and no significant predictor was determined for child BMI. Few studies incorporate the child, parent, and home environment as all part of the transition into a healthy lifestyle; this study may help promote change in those practices. 43
APPENDIX A: ADDITIONAL RESULTS FNPA vs. BMIz -2 -1 0 1 2 3 17 19 21 23 25 27 29 31 Parent FNPA Score C h i l d
B M I
Z - s c o r e
Figure 11. Child BMI z-scores by parent FNPA scores
Figure 12. Bland Altman Plot- FNPA differences compared to mean FNPA difference
44
Table 6. Acquired from Ihmels et al., (2009a)
45
Figure 13. Dendogram grouped distribution by ID numbers
46
APPENDIX B: FAMILY NUTRITION AND PHYSICAL ACTIVITY ASSESSMENT ADULT VERSION Family Nutrition and Physical Activity Assessment (Adult Version) Instructions: For each category, circle the description that best fits your child or your family. It is important to indicate the most common or typical pattern and not what you would like to happen. 1. Breakfast Patterns My child rarely eats breakfast and we dont typically eat together as a family. My child does not regularly eat breakfast but we eat together as a family on most days of the week. My child eats breakfast on most days but we dont typically eat together as a family. My child eats breakfast on most days and we typically eat together as a family. 2. Family Eating Our family regularly eats fast food and we eat while watching TV. Our family regularly eats fast food but we rarely eat while watching TV. Our family rarely eats fast food but we eat while watching TV. Our family rarely eats fast food and we rarely eat while watching TV. 3. Food Choices Our family uses prepackaged foods frequently and we usually do not eat fruits and vegetables with meals (or as snacks). Our family uses prepackaged foods frequently but we regularly consume fruits and vegetables with meals (or as snacks). Our family eats mostly freshly prepared meals but we usually do not eat fruits or vegetables with meals (or as snacks). Our family eats mostly freshly prepared meals and we regularly consume fruits and vegetables with meals (or as snacks). 4. Beverage Choices Our child frequently drinks soda pop or other sweetened drinks, and rarely drinks low fat milk with meals or at snacks. Our child frequently drinks soda pop or other sweetened drinks but frequently drinks low fat milk with meals or at snacks. Our child rarely drinks soda pop or other sweetened drinks, but rarely drinks low fat milk with meals or at snacks. Our child rarely drinks soda pop or other sweetened drinks, and frequently drinks low fat milk with meals or at snacks. 5. Restriction and Reward I dont monitor my childs snack food consumption and snack foods such as candy are frequently used as a reward for good behavior. I dont monitor my childs snack food consumption but snack foods such as candy are not used as a reward for good behavior. I monitor my childs snack food consumption but snack foods such as candy are used as a reward for good behavior. I monitor my childs snack food consumption and snack foods such as candy are not used as a reward for good behavior. 6. Screen Time My child watches television or plays on the computer (or with video games) for more than 4 hours each day. My child watches little television but plays on the computer or with video games for 2-4 hours each day. My child doesnt play on the computer (or with video games) but watches television for 2-4 hours each day. My child watches television or plays on the computer (or with video games) less than 2 hours each day. 7. Television Usage I rarely monitor the amount of TV my child watches and my child has access to a TV in I monitor the amount of TV my child watches but my child has access to a TV in I rarely monitor the amount of TV my child watches but my child does not I monitor the amount of TV my child watches and my child does not 47
his/her bedroom. his/her bedroom. have access to a TV in his/her bedroom. have access to a TV in his/her bedroom.
8. Family Activity I rarely participate in physical activity (e.g. walking) and our family does not play games outside, ride bikes, or walk together very often. I participate regularly in physical activity (e.g. walking) but our family does not play games outside, ride bikes, or walk together very often. I rarely participate in physical activity (e.g. walking) but our family plays games outside, ride bikes, or walks together fairly frequently. I participate regularly in physical activity (e.g. walking) and our family plays games outside, ride bikes, or walks together fairly frequently. 9. Child Activity My child participates in almost no physical activity during his/her free time and is not enrolled in any organized sports or activities with a coach or leader. My child participates in some physical activity a few days a week (2-3 days) in his/her free time but does not typically participate in any organized sports or activities with a coach or leader. My child does not participate in physical activity in his/her free time but does participate in some organized sports or activities with a coach or leader a few days a week (2-3 days). My child regularly participates (i.e. on most days) in physical activity in his/her free time and also participates in sports or activities with a coach or leader. 10. Family Routine Our family does not have a daily routine or schedule for our childs bedtime and our child gets less than 12 hours of sleep each night. Our family does not have a daily routine or schedule for our childs bedtime but our child typically gets at least 12 hours of sleep. Our family follows a daily routine or schedule for our childs bedtime but our child tends to get less than 12 hours of sleep a night. Our family follows a daily routine or schedule for our childs bedtime and our child typically gets at least 12 hours of sleep a night. 48
APPENDIX C: FAMILY NUTRITION AND PHYSICAL ACTIVITY ASSESSMENT: 21 ITEM FORMAT
49
PART 1 Instructions: Circle the square of each row thats most like you and your family. 1. Family Meals I dont eat breakfast often and I dont usually eat meals with my family. I dont eat breakfast often but I eat meals with my family on most days of the week. I eat breakfast most days but I dont usually eat meals with my family. I eat breakfast most days and I eat meals with my family on most days of the week. 2. Eating with Family My family eats fast food often and we eat while watching TV. My family eats fast food often but we dont usually eat while watching TV. My family does not eat a lot of fast food but we eat while watching TV. My family does not eat a lot of fast food and we dont eat while watching TV. 3. Food Choices My family uses a lot of packaged foods, like TV dinners or frozen pizzas, and we do not eat many fruits and vegetables. My family uses a lot of packaged foods, like TV dinners or frozen pizzas but we eat a lot fruits and vegetables. My family eats freshly made meals but we dont eat many fruits and vegetables. My family eats freshly made meals and we eat a lot of fruits and vegetables. 4. Drink Choices I drink a lot of soda pop or kool-aid and I dont drink milk often. I drink a lot of soda pop or kool-aid but I drink a lot of milk. I dont drink soda pop or kool-aid often but I dont drink milk often. I dont drink soda pop or kool-aid often and I drink a lot of milk. 5. Restriction and Reward My parents do not mind how many snacks I eat and I get candy for being good. My parents do not mind how many snacks I eat but I do not get candy for being good. My parents watch how many snacks I eat but I get candy for being good. My parents watch how many snacks I eat and I do not get candy for being good. Family Nutrition and Physical Activity Screening Tool Child Version APPENDIX D: FAMILY NUTRITION AND PHYSICAL ACTIVITY ASSESSMENT CHILD VERSION 50
6. Screen Time I watch TV, play video games, or play on the computer for more than 4 hours each day. I dont watch a lot of TV but I play video games, or play on the computer for 2-4 hours each day. I dont play video games, or play on the computer often but I watch TV for 2-4 hours each day. I watch TV, play video games, or play on the computer for less than 2 hours each day. 7. T.V. Usage My parents do not mind how much TV I watch and I have a TV in my bedroom. My parents limit my TV time but I have a TV in my bedroom. My parents do not mind how much TV I watch but I do not have a TV in my bedroom. My parents limit my TV time and I do not have a TV in my bedroom. 8. Family Activity My parents do not get physical activity, like walking, often and my family does not play games outside, ride bikes, or walk together often. My parents get plenty of physical activity, like walking but my family does not play games outside, ride bikes, or walk together often. My parents do not get physical activity, like walking, often but my family plays games outside, ride bikes, or walk together often. My parents get plenty of physical activity, like walking and my family plays games outside, ride bikes, or walk together often. 9. My Activity I am not involved in physical activities (like tag, bike riding) in my free time and I am not in any sports or activities with a coach or leader. I am involved in physical activities (like tag, bike riding) a few days a week in my free time but I am not in any sports or activities with a coach or leader. I am not involved in physical activities (like tag, bike riding) in my free time but I am involved in sports or activities with a coach or leader a few days a week. I am involved in physical activities (like tag, bike riding) in my free time and I am involved in sports or activities with a coach or leader. 10. Family Routine My family does not have a routine or bedtime for me and I get less than 9 hours of sleep each night. My family does not have a routine or bedtime for me but I get at least 9 hours of sleep each night. My family follows a routine for my bedtime but I get less than 9 hours of sleep each night. My family follows a routine for my bedtime and I get at least 9 hours of sleep each night. 51
APPENDIX E: PARENT STYLE AND DIMENSIONS QUESTIONNAIRE(PSDQ)
52
53
REFERENCES Barradas DT, Fulton JE, Blanck HM, Huhman M. Parental influence on youth television viewing. The Journal of Pediatrics. 2007. 151. 369-373.
Baumrind D. Effects of authoritative parental control on child behavior. Child Development. 1966. 37. 4. 887 907.
Baumrind D. Parenting styles and adolescent development. Encyclopedia of Adolescence. 1991. Vol. 2, pp. 746-758) Garland, NY. 1991.
Berkey CS, Rockett HR, Field AE, Gillman MW, Fraizer L, Camargo CA, Colditz GA. Activity, dietary intake, and weight changes in a longitudinal study of preadolescent and adolescent boys and girls. Pediatrics. 2000. 105, 4. 1-9.
Blissett J, Haycraft E. Are parenting style and controlling feeding practices related? Appetite. 2008. 50. 477-485.
Bowman SA, Gortmaker SL, Ebbeling CB, Pereira MA, Ludwig DS. Effects of fast-food consumption on energy intake and diet quality among children in a national household survey. Pediatrics. 2004. 13. 1. 112-118.
Braet C, Soetens B, Moens E, Mels S, Goossens L, Vlierberghe L. Are tow informants better than one? Parent-child agreement on the eating styles of children who are overweight. European Eating Disorders Review. 2007. 15. 410 417.
Bryant MJ, Ward DS, Hales D, Vaughn A, Tabak RG, Stevens J. Reliability and validity of the healthy home survey: A tool to measure factors within the homes hypothesized to relate to overweight in children. International Journal of Behavioral Nutrition and Physical Activity. 2008. 5. 23. 1- 11.
Byrd-Bredbenner, C. and Grasso, D. Prime-time health: An analysis of television commercials broadcast during programs viewed heavily by children. International Electronic Journal of Health Education. 1999. 2. 159-169.
Caprio S, Daniels SR, Drewnowski A, Kaufman FR, Palinkas LA, Rosenbloom AL, Schwimmer, JB. Influence of race, ethnicity, and culture on childhood obesity: Implications for prevention and treatment. Obesity. 2008. 16. 2566-2577.
Cecil-Karb R, Grogan-Kaylor A. Childhood Body Mass Index in community context: Neighborhood safety, television viewing, and growth trajectories of BMI. Health & Social Work. 2009. 34. 3. 169-177.
Chang HH, Nayga RM. Television viewing, fast-food consumption, and childrens obesity. Contemporary Economic Policy. 2009. 27. 3. 293- 307.
54
Chaundhuri JH, Easterbrooks MA, Davis CR. The relation between emotional availability and parenting style: Cultural and economic factors in a diverse sample of young mothers. Parenting Science and Practice. 2009. 9. 277 299.
Chen X, Beydoun MA, Wang Y. Is sleep duration associated with childhood obesity? A systematic review and meta-analysis. Obesity. 2008. 16. 265- 274.
Cheng TO. Television viewing as a global risk factor for childhood obesity. International Journal of Cardiology. 2005. 103. 344.
Coolahan K, McWayne C, Fantuzzo J, Grim S. Validation of a multidimensional assessment of parenting styles for low-income African-American children. Early Childhood Research Quarterly. 2002. 356 373.
Coplan RJ, Arbeau KA, Armer M. Dont fret, be supportive! Maternal characteristics linking child shyness to psychosocial and school adjustment in kindergarten. Journal of Abnormal Child Psychology. 2008. 36. 3. 359-371.
Coplan RJ, Reichel M, Rowan K. Exploring the associations between maternal personality, child temperament, and parenting: A focus on emotions. Personality and Individual Differences. 2009. 46. 2. 241-246.
Costanzo PR, Woody EZ. Domain-specific parenting styles and their impact on the childs development of particular deviance: The example of obesity proneness. Journal of Social and Clinical Psychology. 1985. 3. 4. 425- 445.
Dietz WH, Gortmaker SL. Preventing obesity in children and adolescents. Annual Review of Public Health. 2001. 22. 337-353.
Drewnowski A, Popkin BM. 1997.The nutrition transition: New trends in the global diet. Nutrition Reviews. 1997. 55, 2. 31-43.
Ebbeling CB, Feldman HA, Osganian SK, Chomitz VR, Ellenbogen SJ, Ludwig DS. Effects of decreasing sugar-sweetened beverage consumption on body weight in adolescents: A randomized, controlled pilot study. Pediatrics. 2006. 117. 673-680.
Ebbeling CB, Pawlak DB, Ludwig DS. Childhood obesity: public-health crisis, common sense cure. The Lancet. 2002. 360. 9331. 473-482.
Finkelstein DM, Hill EL, Whitaker RC. School food environments and policies in US public schools. Pediatrics. 2008. 122:1. e251-259.
Flores R. Dance for health: improving fitness in AfricanAmerican and Hispanic adolescents. Public Health Reports. 1995. 110. 189193.
55
Forshee RA, Anderson PA, Storey ML. Sugar-sweetened beverages and body mass index in children and adolescents: A meta-analysis. American Journal of Clinical Nutrition. 2008. 87, 6. 1662- 1671.
Francis LA, Lee Y, Birch LL. Parental weight status and girls television viewing, snaking, and body mass index. Obesity Research. 2003.11:1. 143- 151. Freedman DS, Khan LK, Serdula MK, Galuska DA, Dietz WH. Trends and correlates of class 3 obesity in the United States from 1990 through 2000. JAMA. 2002. 288. 14. 1758-1761. Freedman DS, Dietz WH, Srinivasan SR, and Berenson GS. The Relation of Overweight to Cardiovascular Risk Factors Among Children and Adolescents: The Bogalusa Heart Study. Pediatrics. 1999. 103, 6. 1175-1182. Golan M, Crow S. Parents are key players in the prevention and treatment of weight-related problems. Nutrition Reviews. 2004. 62. 1. 39-50. Golan M, Weizman A. Reliability and validity of the family eating and activity habits questionnaire. European Journal of Clinical Nutrition. 1998. 52. 771- 777. Goodwin J L, Silva GE, Kaemingk KL, Sherrill DL, Morgan WJ, Quan SF. Comparison between reported and recorded total sleep time and sleep latency in 6-to 11-year-old children: The Tucson Childrens Assessment of Sleep Apnea Study (TuCASA). Sleep Breath. 2007. 11. 85 92. Gordana M, Levitsky DA. Nutritional and energetic consequences of sweetened drink consumption in 6 to 13 year-old children. Journal of Pediatrics. 2003. 142. 604-610.
Gortmaker SL, Dietz WH, Cheung LWY. Inactivity, diet, and the fattening of America. Journal of American Dieting Association. 1990. 90. 1247-1252.
Grills AE, Ollendick TH. Issues in parent-child agreement: The case of structured diagnostic interviews. Clinical Child and Family Psychology Review. 2002. 5. 1. 57 83.
Gruber R, Sadeh A, Raviv A. Sleep of school-age children: objective and subjective measures. Sleep Research. 1997. 26. 158. Harrison M, Burns CF, McGuinness M, Heslin J, Murphy NM. Influence of a health education intervention on physical activity and screen time in primary school children: Switch OffGet Active. Sports Medicine Australia. 2006. 9. 388-394.7. 1154-1161. Hart CH, Nelson DA, Robinson CC, Olsen SF, McNeilly-Choque MK. Overt and relational aggression in Russian nursery school-age children: Parenting style and marital linkages. Developmental Psychology. 1998. 34. 4. 687 697. 56
He MZ, Harris S, Piche L, et al. Understanding screen-related sedentary behavior and its contributing factors among school aged children: A social-ecologic exploration. American Journal of Health Promotion. 2009. 23. 5. 299- 308. Hennessey E, Hughes SO, Goldberg JP, Hyatt RR, Economos CD. Parent behavior and child weight status among a diverse group of underserved rural families. Appetite. 2010. 54. 369 377. Hubbs-Tait L, Kennedy TS, Page MC, Topham GL, Harrist AW. Parental feeding practices predict authoritative, authoritarian, and permissive parenting styles. Journal of the American Dietetic Association. 2008. 108. Hughes SO, Shewchuk RM, Baskin ML, Nicklas TA, Qu H. Indulgent feeding style and childrens weight status in preschool. Journal of Developmental and Behavioral Pediatrics. 2008. 29. 403 410. Ihmels MA, Welk GJ, Eisenmann JC, Nusser SM. Development and preliminary validation of a Family Nutrition and Physical Activity (FNPA) screening tool. International Journal of Behavioral Nutrition and Physical Activity. 2009. 6. 14.
Ihmels MA, Welk GJ, Eisenmann JC, Nusser SM, Myers EF. Prediction of BMI change in young children with the Family Nutrition and Physical Activity (FNPA) screening tool. Annual Behavioral Medicine. 2009. 38. 60-68.
Joyce JL, Zimmer-Gembeck MJ. Parent feeding restriction and child weight. The mediating role of child disinhibited eating and the moderating role of the parenting context. Appetite. 2009. 52. 726-734.
Karavasilis L, Doyle AB, Markiewicz D. Associations between parenting style and attachment to mother in middle childhood and adolescence. International Journal of Behavioral Development. 2003. 27. 2. 153-164.
Kremers SPJ, Brug J. de Vries H, Engels RCME. Parenting style and adolescent fruit consumption. Appetite. 2003. 41. 43 50.
Landhuis CE, Poulton R, Welch D, Hancox RJ. Childhood sleep time and long-term risk for obesity: A 32-year prospective birth cohort study. Pediatrics. 2008. 122:5. 955-960.
Lanningham-Foster L, Foster RC, McCrady SK, Jensen TB, Mitre N, Levine J A. Activity-promoting video games and increased energy expenditure. Journal of Pediatrics. 2009. 154. 819-823. Lauer RM, Connor WE, Leaverton PE. Coronary heart disease risk factors in school children: the Muscatine study. Journal of Pediatrics. 1975. 86. 697-706. 57
Laurson KR, Eisemann JC, Welk GJ, Wickel EE, Gentile DA, Walsh DA. Combined influence of physical activity and screen time recommendations on childhood overweight. Journal of Pediatrics. 2008. 153. 209-214.
Lazaar N, Aucouturier J, Ratel S, Rance M, Meyer M, Duche P. Effect of physical activity intervention on body composition in young children: Influence of body mass index status and gender. Acta Paediatrica International Journal Paediatrics. 2007. 96. 13151320.
Macfarlane A, Cleland V, Crawford D, Campbell K, Timperio A. Longitudinal examination of the family food environment and weight status among children. International Journal of Pediatric Obesity. 2009. 4. 343-352.
Maccoby EE, Martin J A. Socialization in the context of the family: Parent-child interaction. Handbook of Child Psychology. 1983. Vol. 4. Wiley, NY.
Maddah M. Childhood obesity and early prevention of cardiovascular disease: Iranian families act to late. International Journal of Cardiology. 2008. 126. 292- 294.
Maddison R, Foley L, Mhurchu C, Jull A, J iang Y, Prapavessis H, Rodgers A, Vander H, Hohepa M, Schaaf D. Feasibility, design and conduct of a pragmatic randomized controlled trial to reduce overweight and obesity in children: The electronic games to aid motivation to exercise (eGAME) study. BMC Public Health. 2009. 9. 146.
McCullough N, Muldoon O, Dempster M. Self-perception in overweight and obese children: a cross- sectional study. Childcare, Health, and Development. 2008. 35. 357-364.
McDougall J, Duncan MJ. Children, video games and physical activity: An exploratory study. International Journal of Disabilities and Human Development. 2008. 7. 1. 89-94.
Morawska A, Sanders MR. Concurrent predictors of dysfunctional parenting and maternal confidence: implications for parenting interventions. Childcare, Health, and Development. 2007. 33. 6. 757 767.
Morgenstern M, Sargent JD, Hanewinkel R. Relation between socioeconomic status and Body Mass Index evidence of an indirect path via television use. Archives of Pediatrics & Adolescent Medicine. 2009. 163. 8. 731- 738.
Mrdjenovic G, Levitsky DA. Nutritional and energetic consequences of sweetened drink consumption in 6 to 13 year old children. Journal of Pediatrics. 2003. 142. 6. 604- 610.
Muller MJ, Grund A, Krause H, Siewers M, Bosy-Westphal A, Rieckert H. Determinants of fat mass in prepubertal children. British Journal of Nutrition. 2002. 88. 545- 554.
Must A, Barish EE, Bandini LG. Modifiable risk factors in relation to changes in BMI 58
and fatness: What have we learned from prospective studies of school-aged children? International Journal of Obesity. 2009. 33. 705- 715.
Must A, Strauss RS. Risks and consequences of childhood and adolescent obesity. International Journal of Obesity. 1999. 23:2. S2- S11.
National Center for Health Statistics. Health, United States, 2007 with Chartbook on Trends in the Health of Americans. Hyattsville, MD 2007.
Nunnally J C. Psychometric theory (2 nd ed.). New York: McGraw-Hill. 1978.
Obarzanek E, Schreiber GB, Crawford PB, Goldman SR, Barrier PM, Frederick MM, Lakatos E. Energy intake and physical activity in relation to indexes of intake and physical activity in relation to indexes of body fat: The National Heart, Lung, and Blood Institute Growth and Health Study. American Journal of Clinical Nutrition. 1994. 60. 15 22.
Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 1994-2004. Journal of the American Medical Association. 2006. 29. 13. 1549-1555.
Ozturk A, Mazicioglu MM, Poyrazoglu S, Cicek B, Gunay O, Kurtolglu S. The relationship between duration and obesity in Turkish children and adolescents. ACTA Pediatrica. 2009. 98. 699- 702.
Padez C, Mourao I, Moreira P, Rosado V. Long sleep duration and childhood overweight/obesity and body fat. American Journal of Human Biology. 2009. 21. 371-376.
Pate RR, Ward DS, Saunders RP, Felton G, Dishman RK, Dowda M. Promotion of physical activity among high-school girls: a randomized controlled trial. American Journal of Public Health. 2005. 95. 1582-1587.
Patrick H, Nicklas TA, Hughes SO, Morales M. The benefit of authoritative feeding style: caregiver feeding styles and childrens food consumption patterns. Appetite. 2005. 44. 243-249.
Pinderhughes EE, Dodge KA, Bates JE, Petit GS, Zelli A. Discipline responses: Influences of parents socioeconomic status, ethnicity, beliefs about parenting, stress, and cognitive- emotional processes. Journal of Family Psychology. 2000. 14. 3. SI. 300 400.
Powell KE, Roberts AM, Ross JG, Phillips MC, Ujamaa DA, Zhou M. Low physical fitness among fifth- and seventh-grade students, Georgia, 2006. American Journal of Preventive Medicine. 2009. 36. 4. 304-310.
Reilly J J, Armstrong J, Dorotsty AR, Emmett PM, Ness A, Rogers I, Steer C, Sherrif A. Early life risk factors for obesity in childhood: cohort study. British Medical Journal. 2005. 330. 1357- 1359. 59
Rhee KE, Lumeng J C, Appugliese DP, Kaciroti N, Bradley RH. Parenting styles and overweight status in first grade. Pediatrics. 2006. 117. 6. 2047- 2054.
Robinson CC, Mandelco B, Olsen S F, Hart CH. Authoritative, authoritarian, and permissive parenting practices: development of a new measure. Psychological Reports. 1995. 77. 819- 830.
Robinson TN, Kiernan M, Matheson DM, Haydel KF. Is parental control over childrens eating associated with childhood obesity? Results from a Population-based sample of third graders. Obesity. 2001. 9. 306 312.
Robinson TN. Reducing children's television viewing to prevent obesity: a randomized controlled trial. Journal of the American Medical Association. 1999. 282. 15611567.
Roche KM, Ensminger ME, Cherlin AJ. Variations in parenting and adolescent outcomes among African-American and Latino families living in low-income, urban areas. Journal of Family Issues. 2007. 28. 7. 882 909.
Rockett HR, Berkey CS, Field AE, Colditz GA. Crossectional measurement of nutrient intake among adolescents in 1996. Preventive Medicine. 2001. 33. 27-37.
Rodriguez-Artalejo F, Garces C, Gorgojo L, Lopez Garcia E, Martin-Moreno JM et al. Dietary patterns among children aged 6-7y in four Spanish cities with widely differing cardiovascular mortality. European Journal of Clinical Nutrition. 2002. 56. 141- 148.
Rossiter JR, Robertson TS. Childrens television viewing: An examination of parent-child consensus. Sociometry. 1975. 38. 2. 308 - 326.
Rowlands AV, Eston RG, Ingledew DK. Relationship between activity levels, aerobic fitness, and body fat in 8 to 10 year old children. Journal of Applied Physiology. 1999. 86. 1428- 1435.
Rutledge J M, Topham GL, Kennedy TS, Page MC, Hubbs-Tait L, Harrist AW. Relation between parenting style and child weight. Federation of American Societies for Experimental Biology. 2007. 21. 353-354.
Schmitz KH, Lytle LA, Phillips GA, Murray DM, Brinbaum AS, Kubik MY. Psychosocial correlates of physical activity and sedentary leisure habits in young adolescents: The teens eating for energy and nutrition study. Preventive Medicine. 2002. 34. 266 278.
Sekine M, Yamagami T, Handa K et al. A dose-response relationship between short sleeping hours and childhood obesity: results of the Toyama Birth Cohort Study. Child Care Health and Development. 2002. 28. 163-170.
60
Shaw J. Epidemiology of childhood type 2 diabetes and obesity. Pediatrics Diabetes. 2007. 8. 9. 7- 15.
Skinner JD, Bounds W, Carruth BR, Morris M, Ziegler P. Predictors of children's body mass index: a longitudinal study of diet and growth in children aged 2-8 y. International Journal of Obesity and Related Metabolic Disorders. 2004. 28. 4. 476-82.
Stephens MB, Wentz SW. Supplemental fitness activities and fitness in urban elementary school classrooms. Family Medicine. 1998. 30. 220223.
Taheri S. The link between short sleep duration and obesity: We should recommend more sleep to prevent obesity. Archives of Disease in Childhood. 2006. 91. 881884.
Tai A, Volkmer R, Burton A. Association between asthma symptoms and obesity in preschool (4-5 year old) children. Journal of Asthma. 2009. 46. 362-365.
Tak NI, te Velde SJ, de Vries JHM, Brug J. Parent and child reports of fruit and vegetable intakes and related family environmental factors show low levels of agreement. Journal of Human Nutrition and Dietetics. 2006. 19. 275 285.
Tein J, Roosa MW, Michaels M. Agreement between parent and child reports on parental behaviors. Journal of Marriage and Family. 1994. 56.2. 341 355.
Touchette E, Petit D, Tremblay RE, Boivin M, Falissard B, Genolini C, Montplaisir JY. Associations between sleep duration patterns and overweight/obesity at age 6. Sleep. 2008. 31. 11. 1507-1514.
US Dept of Health and Human Services. Healthy People 2000. Washington, DC: US Government Printing Office; 1998.
Velazquez V, Silva M, Arbona I, et al. Asthma and obesity in children: Is there a link? Journal of Allergy and Clinical Immunology. 2008. 121. 2. S225.
Vanhala M, Korpelainan R, Tapanainen P, Kaikkonen K, Kaikkonen H, Saukkonen T, Keinnen- Kiukaanniemi S. Lifestyle risk factors for obesity in 7 year old children. Obesity Research & Clinical Practice. 2009. 3. 99-107 Whitaker RC, Wright JA, Pepe MS, Seidel KD, Deitz WH. Predicting obesity in young adulthood from childhood and parental obesity. New England Journal of Medicine. 1998. 337. 869 - 873. Zeller MH, Boles RE, Reiter-Purtill J. The additive and interactive effects of parenting style and temperament on obese youth seeking treatment. International Journal of Obesity. 2008. 32. 1474- 1480.