Optimal Defaults and Parent Empowerment in the Prevention of Early Childhood Obesity: A Community Center-Based Pilot Study

<p>NON-TECHNICAL SUMMARY:<br/> Obesity is dangerous, costly, and threatens the future of our children. Solutions are far from straightforward as the 300% rise in pediatric obesity to 16.9% during the past 30 years stands to tremendously burden the healthcare system and curb the productivity of our youth. The call for a shift in emphasis from treatment of already obese children to prevention is timely. Obesity prevention symbolizes a core philosophical tension between individual freedom and government actions designed to promote public welfare. On the one hand, a model emphasizing personal responsibility identifies the causes of obesity as lying within the individual including both biological factors and personal choices; to date, the efforts extending from this model have yielded insufficient success relative to the scope of the childhood obesity epidemic. Conversely, a
public health model places more emphasis on environmental causes. Understandably, the models have different conceptions on how each problem should be addressed with education, self-control and ultimately treatment being the logical interventions under the personal responsibility model, and prevention and policy changes representing the public health model. Improving default options represents a compromise between the two; public policies can determine what the optimal default positions are, yet the choice remains with the individual to opt out. This community center-based project examines how optimizing the health value of the default mode combined with parent empowerment may serve as a critical strategy in the prevention of pediatric obesity. Specifically, the two experiments in this proposal examine how hard parents work to remain in or obtain healthy choices for their young children.
These choices influence child energy intake and expenditure, which in turn influence obesity. In a food experiment, children in the optimal default condition will be automatically registered for a healthy breakfast; parents must actively opt out of this breakfast to gain access to a less healthy breakfast if desired. In the suboptimal default condition, this is reversed. Similarly, in an exercise experiment, children in the optimal default condition are automatically registered for a physical activity; parents must actively opt out and choose a sedentary activity (a video) if desired. In the suboptimal default condition, the video is the default choice. Across these experiments, we are also testing the effects of parent education regarding healthy options for their children - with or without a message empowering parents to implement optimal defaults in the home - on these health-related
choices. We expect that making the default option more optimal will lead to more frequent healthy choices, and that this effect will be more pronounced when parents are not just educated about the nature and benefits of healthy foods and physical activity, but are also specifically empowered to implement optimal defaults for their children. We will use our finding to educate the community (parents and schools) on the best methods for making healthy choices the automatic and easy option for young children.
<p>APPROACH:<br/> Design: This pilot study involves two tests of optimal defaults to generate new knowledge of the behavioral factors that influence energy intake and expenditure, which in turn influence childhood obesity. Specifically, we will run two randomized between-subjects 2x2 experiments to test the effects of (a) shifts toward optimal defaults in the food and exercise domains and (b) parent education that teaches the concept of optimal defaults and empowers parents to facilitate healthy choices for their children. We will stratify by age (3-5 and 6-8). Sample: 216 parent-child dyads will be recruited from a community-based family center. Inclusion criteria: child ages 3-8 with at least one parent/guardian willing to participate. Exclusion criteria: any medical condition in the child that would preclude fasting before the experimental breakfast or eating the foods
that will be served, or preclude participating in moderate exercise; a diagnosis of autism, mental retardation, or other developmental disability that would make participation in group activities challenging. Power analysis: N=27 subjects per cell will provide at least 80% power to detect a large difference in proportions of .35. Data Analysis: Logistic regressions will test the two main effects (default condition and education condition) and their interaction term, with age as a covariate; the outcome variable for each experiment is binary (healthy choice opted yes/no). Secondary analyses with a 2x2 ANOVA will be conducted measuring actual quantity of healthy food consumed in the food experiment. Food Optimal Default Experiment: Parent participants will begin by watching an educational video. The content of the video will either focus on nutrition education only or nutritional education
plus parent empowerment and will test of the effects of priming parents about the concept of optimal defaults. The test meal will follow the video presentation. In the optimal default condition, children are automatically registered for a healthy breakfast and parents must opt out and actively choose a less healthy breakfast if desired; in the suboptimal condition this is reversed. Exercise Optimal Default Experiment: Parents in this experiment will also begin by viewing an educational video on the importance of exercise (with or without parent empowerment). The exercise default experiment will follow; in the optimal default condition, children are automatically registered for physical activity and parents must opt out and actively choose an alternative sedentary activity (a video) if desired; in the sub-optimal default condition, this is reversed. Expected Outcomes: We expect that
making the default option more optimal will lead to more frequent healthy choices, and that this effect will be more pronounced when parents are not just educated about the nature and benefits of healthy foods and physical activity, but are also specifically empowered to implement optimal defaults for their children. Dissemination efforts based on the findings will be conducted with our collaborating community outreach organization as part of their Obesity Prevention Initiative.
<p>PROGRESS: 2011/02 TO 2013/02<br/>OUTPUTS: This study applies behavioral economics to optimize child food and activity choices via parent-driven decisions. Specifically, this project tests an optimal defaults paradigm, representing new and innovative research aimed at co-opting the default-influenced response bias to stay with an existing, pre-selected, or easier choice (rather than seeking out an available alternative) as a critical strategy in the prevention of pediatric obesity. We ran two randomized 2x2 experiments to test the effects of (a) shifts toward optimal defaults in the food and exercise domains, and (b) parent education that teaches the concept of optimal defaults and empowers parents to facilitate healthy choices for their children. We hypothesized that making the default option more optimal (less obesogenic) would yield more frequent choice of healthier
foods and behaviors, and that parent empowerment education would potentiate these effects, in parent-child (age 3-8) dyads. Outputs of the project to date are as follows: Activities have included conducting the food and exercise experiments with a collective total of 120 families and administering an assessment battery to participants, the variables from which will be tested as predictors and moderators in planned secondary data analyses; and conducting preliminary data analysis on the main effects of default condition and parent education condition, as well as their interaction (see Outcomes below). Events have included presenting the theoretical model of optimal defaults as applied to the prevention of childhood obesity, and the proof-of-concept experiments incorporated in this seed grant, within invited talks at New York Universitys Child Study Center, Columbia Universitys New York
Obesity Nutrition Research Center, the City University of New York, and Boston University, as well as at the AFRI annual obesity project directors meeting in November 2011. We also presented data from the food experiment at the 2012 Society for Nutrition Education and Behavior. We have presented our experimental model and preliminary data as part of a peer reviewed symposium at the Association for Behavioral and Cognitive Therapies conference in November 2012, and have an accepted peer reviewed paper presentation planned for May 2013 at the International Conference on Eating Disorders. In addition, we have been invited to present our findings at the Rudd Center for Obesity and Food Policy at Yale University and invited back to Columbia University to present our data as part of their Appetitive Behavior seminar series. Products from the project have included 8 recorded educational videos
being used in the experiments (nutrition and exercise health, with and without specific messages to empower parents to implement optimal defaults on their childrens behalf, in English and Spanish); and a database of our ongoing results. Dissemination of findings beyond the scientific field, to the public, is planned after our final data analyses in collaboration with the Northern New Jersey Community Foundation as part of their Obesity Prevention Initiative and with the Bergen Family Center, our two community partners for this project. PARTICIPANTS: Katharine L. Loeb, PhD (PD) and Cynthia Radnitz, PhD (Co-PD) have overseen the design and implementation of the experimental paradigms, the procurement of IRB approvals and continuation paperwork, and data collection, entry and protection. Specifically, in consultation with Marlene Schwartz, PhD and Kathleen Keller, PhD, the PDs wrote the
scripts for and produced the eight videos for the priming manipulation in the experiments; designed the menus for the food experiment consistent with published nutritional guidelines and the activities for the exercise paradigm; created materials and procedures for choice designation; wrote detailed operations procedures for implementation of each experiment; and recruited participants for and conducted the experiments. Dr. Loeb is responsible for all programmatic reporting and implementation of best research practices. In consultation with Sue Marcus, PhD, she has conducted the preliminary data analyses. Dr. Loeb has presented the optimal default model and experiments at multiple universities and conferences. The PDs have also completed a manuscript on optimal defaults in the prevention of childhood obesity, which has been submitted to a peer reviewed journal. The PDs have received a
FDU Provosts Seed Grant to pilot the application optimal defaults to elementary school lunch programs as a next stage of this program of research. The PDs have also piloted a home-based implementation of the optimal default paradigm in collaboration with Dr. Keller in a design that combines the PDs work on the current study with her recent findings on food packaging manipulations to increase fruit and vegetable consumption in young children. An Integrated grant extending from the independent and combined experiments is being submitted to NIFA/USDA. Finally, the PDs have received an obesity prevention seed grant from NIFA/USDA applying optimal defaults for late adolescents in the first year college dining experience. Through our nonprofit partner organization, the Northern New Jersey Community Foundation - collaborators for this seed grant as well as for the Extension arm of a future USDA
Integrated project - we have promoted the study and the broader potential for optimal defaults in the prevention of childhood obesity. We have also established relationships with multiple collaborators and contacts through their extensive network of health care facilities, boards of health, schools, organizations, programs, and affiliates throughout NJ. The Bergen Family Center, a NJ nonprofit, community-based center, has supplied the necessary facilities to conduct the study and functioned as our performance and primary recruitment site. This seed grant has also contributed to training and professional development of PhD candidates in Clinical Psychology and undergraduate Psychology majors at FDU, who have been taught the scientific underpinnings of the study and trained in research ethics, design, and methodology in their work on the project. One graduate student is a co-author on the
aforementioned scientific paper and her dissertation will involve an optimal default experiment to increase energy expenditure in late adolescents. TARGET AUDIENCES: The ethnic identity of the participants in the food paradigm study is primarily Latino (62%); the racial identity is primarily white (40%), followed by black or African American (21%), American Indian or Alaskan Native (6%), Asian (5%), and Native Hawaiian or other Pacific Islander (1%), with 28% not reporting a racial identity. We conducted both experiments in English and Spanish. Our ultimate target audience is the general population of parents and young children, as our study is designed to inform parenting practices around food and activity presentation to reduce obesogenic factors in the home and early education environments. In particular, we are interested in targeting the segments of this subpopulation that are most
vulnerable to childhood obesity; ultimate extension and outreach efforts from this project should reflect the significant racial, ethnic, and socioeconomic disparities in obesity prevalence among US children and adolescents. The Integrated grant extending from this research to be submitted to NIFA/USDA shortly (described above), will in particular target low income families, and, for the Extension component, focus on participants from the Expanded Food and Nutrition Education Program (EFNEP). The Northern New Jersey Community Foundation (NNJCF) and the Bergen Family Center, our community partner organizations, serve low-income communities and will be an integral part of parent education efforts based on the findings of the studies and in connection with the NNJCF Obesity Prevention Initiative. PROJECT MODIFICATIONS: There were no major changes in the design or implementations of the
experiments. Recruitment was slower than anticipated, and given that target enrollment had accordingly not been met by the end of the 18 month grant period, we applied for and received a no-cost extension during which we recruited another 39 parent-child dyads. While we ultimately did not meet our original projected recruitment goals for the experiments, this no-cost extension period yielded a final sample size of 120 participant dyads with sufficient power to detect a strong main effect for default condition.
<p>PROGRESS: 2011/02/15 TO 2012/02/14<br/>OUTPUTS: This study applies behavioral economics to optimize child food and activity choices via parent-driven decisions. Specifically, this project tests an optimal defaults paradigm, representing new and innovative research aimed at co-opting the default-influenced response bias to stay with an existing, pre-selected, or easier choice (rather than seeking out an available alternative) as a critical strategy in the prevention of pediatric obesity. We are running two randomized 2x2 experiments to test the effects of (a) shifts toward optimal defaults in the food and exercise domains, and (b) parent education that teaches the concept of optimal defaults and empowers parents to facilitate healthy choices for their children. We hypothesize that making the default option more optimal (less obesogenic) will yield more frequent choice
of healthier foods and behaviors, and that parent empowerment education will potentiate these effects, in parent-child (age 3-8) dyads. Outputs of the project to date are as follows: Activities have included conducting the food experiment with 62 families and administering an assessment battery to participants, the variables from which will be tested as predictors and moderators in planned secondary data analyses; and analyzing the preliminary data from this condition for the main effect of default food condition (optimal vs. suboptimal) (see Outcomes below). We have also commenced the physical activity paradigm and run 7 parent-child dyads. Events have included presenting the theoretical model of optimal defaults as applied to the prevention of childhood obesity, and the proof-of-concept experiments incorporated in this seed grant, within invited talks at New York University's Child
Study Center, Columbia University's New York Obesity Nutrition Research Center at St. Luke's-Roosevelt Hospital, the City University of New York, and Boston University, as well as at the AFRI annual obesity project director's meeting in November 2011. We will also be presenting data from the food experiment at the 2012 Society for Nutrition Education and Behavior. Under current peer review is a symposium presentation at the Association for Behavioral and Cognitive Therapies conference. In addition, we have been invited to present our findings at the Rudd Center for Obesity and Food Policy at Yale University. Products from the project have included 8 recorded educational videos being used in the experiments (nutrition and exercise health, with and without specific messages to empower parents to implement optimal defaults on their children's behalf, in English and Spanish); and a database
of our ongoing results. Dissemination of findings beyond the scientific field, to the public, is planned at the completion of each experiment in collaboration with the Northern New Jersey Community Foundation (formerly the Englewood Area Community Foundation) as part of their Obesity Prevention Initiative and with the Bergen Family Center, our two community partners for this project. PARTICIPANTS: Katharine L. Loeb, PhD (PD) and Cynthia Radnitz, PhD (Co-PD) have overseen the design and implantation of the experimental paradigms, the procurement of Institutional Review Board approvals and continuation paperwork, and data collection, entry and protection. Specifically, in consultation with Marlene Schwartz, PhD and Kathleen Keller, PhD, the project directors wrote the scripts for and produced the eight videos being used for the priming manipulation in the experiments; designed the menus
for the food experiment in reference to published nutritional guidelines and the activities for the exercise paradigm; created materials and procedures for choice designation (i.e., for participants to remain in or opt out of the default choice - our primary dependent variable); wrote detailed operations procedures for implementation of each experiment; and recruited participants for and conducted the experiments. Dr. Loeb is also responsible for all programmatic reporting and implementation of best research practices. In consultation with Sue Marcus, PhD, she has also conducted preliminary data analyses. Dr. Loeb has presented the theoretical model of optimal defaults as applied to the prevention of childhood obesity, and the proof-of-concept experiments incorporated in this seed grant within invited talks at New York University, Columbia University, the City University of New York, and
Boston University, as well as at the AFRI annual obesity project director's meeting. Several additional talks are pending. Dr. Loeb and Dr. Radnitz have also completed a manuscript on applying optimal defaults in the prevention of childhood obesity, which will be submitted shortly to a peer reviewed journal. The project directors have also designed a set of experiments extending from the current work to apply optimal defaults in elementary school-based settings as the next stage of this program of research, and have drafted a grant for later submission to NIFA/USDA. Through our nonprofit partner organization, the Northern New Jersey Community Foundation - collaborators for this seed grant as well as for the Extension arm of a future USDA Integrated project - we have promoted the study and the broader potential for optimal defaults in the prevention of childhood obesity. We have also
recruited for the experiments, and established relationships with multiple collaborators and contacts, through their extensive network of health care facilities, boards of health, schools, organizations, programs, and affiliates throughout the state of New Jersey. The Bergen Family Center, a nonprofit, community-based family center in Englewood, NJ, has supplied the necessary facilities to conduct the study and functioned as our performance and primary recruitment site. This seed grant has also contributed to training and professional development of PhD candidates in Clinical Psychology and undergraduate Psychology majors at FDU, who have been taught the scientific underpinnings of the study and are being trained in research ethics, design, and methodology in their work on the project. One graduate student is a co-author on the aforementioned scientific paper. TARGET AUDIENCES: The
ethnic identity of the participants in the food paradigm study is primarily Latino (63%); the racial identity is primarily white (55%), followed by black or African American (21%), American Indian or Alaskan Native (8%), and Asian (3%), with 13% not reporting a racial identity. We are conducting both experiments in English and Spanish. Our ultimate target audience is the general population of parents and young children, as our study is designed to inform parenting practices around food and activity presentation to reduce obesogenic factors in the home and early education environments. In particular, we are interested in targeting the segments of this subpopulation that are most vulnerable to childhood obesity; ultimate extension and outreach efforts from this project should reflect the significant racial, ethnic, and socioeconomic disparities in obesity prevalence among US children and
adolescents. The Northern New Jersey Community Foundation (NNJCF) and the Bergen Family Center, our community partner organizations, serve low-income communities and would be an integral part of parent education efforts based on the findings of the studies and in connection with the NNJCF Obesity Prevention Initiative. PROJECT MODIFICATIONS: There have been no major changes in the design or implementations of the experiments. Recruitment has been slower than anticipated, and given that target enrollment has accordingly not been met, we may apply for a no-cost extension.