Source: CORNELL UNIVERSITY submitted to
HEALTHY CHILDREN, HEALTHY FAMILIES: PARENTS MAKING A DIFFERENCE: A RANDOMIZED CONTROLLED TRIAL
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
TERMINATED
Funding Source
Reporting Frequency
Annual
Accession No.
1011175
Grant No.
(N/A)
Project No.
NYC-399418
Proposal No.
(N/A)
Multistate No.
(N/A)
Program Code
(N/A)
Project Start Date
Oct 4, 2016
Project End Date
Dec 30, 2019
Grant Year
(N/A)
Project Director
Dollahite, JA, S..
Recipient Organization
CORNELL UNIVERSITY
(N/A)
ITHACA,NY 14853
Performing Department
Nutritional Sciences
Non Technical Summary
This research project will rigorously test whether Healthy Children, Healthy Families: Parents Making a Difference! (HCHF) is effective in educating low-income parents about nutrition and motivating them to use parenting practices related to food and physical activity that research indicates are best aligned with positive health outcomes for children. Results from prior practice-based research suggest that this is a highly effective approach; this study will provide research-tested evidence.The research objectives are to assess 1) the impacts of HCHF on parenting practices and self-efficacy, parent diet quality and physical activity, and child diet quality and physical activity from baseline to post-education in immediate education versus delayed-education control; and 2) retention of changes in behavior for 16 weeks. If proven effective, HCHF is well suited for national dissemination in EFNEP, SNAP-Ed and other public health programs.The study is a collaboration between Cornell University's Division of Nutritional Sciences and Cornell Cooperative Extension in New York City. Approximately 300 parents and caregivers with young children 3-5 years old in Head Start and other childcare programs in New York City will be enrolled. The study will employ a randomized design in which half the groups receive HCHF education in period 1 while the other half serve as delayed-education controls receiving no education. In period 2, the delayed-education groups will receive HCHF. In period 3, no education will be offered and all groups will be followed longitudinally to assess retention of behavior changes.HCHF is an innovative curriculum for parents and caregivers that integrates nutrition, physical activity, and parenting to help children make healthy choices and prevent obesity. The curriculum reflects a socio-ecological model of the multi-dimensional ways parents and caregivers affect children's energy intake and expenditure. It includes the knowledge, skills, and attitudes parents need to influence child behaviors. HCHF fully integrates parenting and nutrition education in a dialogue-based, hands-on format and includes training to ensure effective delivery by paraprofessional educators.HCHF was developed in close collaboration with nutrition and parenting Extension educators and revised based on stakeholder input, so it is grounded in the experiences of low-income parents and honed by educators. It has been shown to be feasible in practice, and engaging and acceptable to participants and EFNEP staff.
Animal Health Component
0%
Research Effort Categories
Basic
(N/A)
Applied
100%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
70360991010100%
Goals / Objectives
Determine the impacts of HCHF on how low-income parents enrolled in EFNEP use effective parenting practices to influence children's healthy eating and active play behavior, as compared to a delayed intervention control group.Investigate parents' retention of changes in parenting practices and their influence on child diet quality and physical activity when assessed up to 16 weeks after HCHF sessions have ended.
Project Methods
Setting and participants. This study will be carried out in Head Start and childcare programs in New York City. About 300 parents/primary caregivers with children 3-5 years old will be enrolled. Cash incentives will not be provided at baseline, but at the 3 subsequent data collection points, with increasing amounts ($15, 25, 35) for each point to ensure participant retention throughout the study. These amounts are consistent with previous research that our group has conducted with this population. Half the groups will be English- and half Spanish-speaking due to different outcomes from these populations in standard EFNEP evaluation. HCHF is available in both English and Spanish. Education will be delivered by educators trained and experienced in the delivery of HCHF and fluent in the language of the participants in a given group.Design. The research will use a group-randomized design carried out over three 8 week periods (see table below). In period 1, half the groups will receive HCHF education (immediate education, IE) and half will serve as delayed-education controls receiving no education (delayed education, DE). In period 2, the treatments will cross over, so IE will receive no education, and will be followed longitudinally for periods 2 and 3 (16 weeks total) to assess retention of behavior change. In period 2, DE will receive HCHF. In period 3, neither group will receive education, and both will be followed longitudinally to assess retention of behavior change. Each DE group will be paired in time with an IE group at the same site to diminish site and seasonal effects. Data will be collected at baseline (T1) and at the end of each of the periods (T2, T3, T4).Numbers of groups and participants. Based on the design and HCHF practice-based evidence the estimated sample size is 32 groups (confidence Interval=95%, alpha=5%; power=80%). The NYC EFNEP has a participant retention rate of 70-80%, so 15 participants will be recruited for each group, with the anticipation that 10-12 will be retained, for a total of at least 300 participants.Hypotheses to be tested include the following:Objective 1. Assess the impacts of HCHF from baseline to post-education1.1 Parenting practices and self-efficacy will improve more in HCHF than in control groups1.2 Parent diet quality and physical activity will improve more in HCHF than in control groups1.3 Child diet quality and physical activity will improve more in HCHF than in control.Objective 2. Retention of changes in behavior for 16 weeks2.1 Parenting practices and self-efficacy assessed immediate post-education will be retained.2.2 Parent diet quality and physical activity assessed immediate post-education will be retained.2.3 Child diet quality and physical activity assessed immediate post-education will be retainedData Collection. Demographic and other data from validated measures will be collected from parents/caregivers at each time point. Validated measures include the HCHF Checklist (validated by our team for EFNEP) and complementary measures that assess parenting feeding practices, food behavior in parents, and food behavior in children to ensure outcomes with the brief checklist are convincing when assessed in greater depth. Parent self-efficacy around obesity prevention behaviors will also be assessed. Measures will focus on behaviors included in HCHF and designed to test the hypotheses.HCHF Checklist is a brief 16-item, validated outcome evaluation tool on child and parent behaviors addressed in HCHF (Dickin et al., 2012). The instrument is available in both English and Spanish, has low respondent burden, and assesses behavioral outcomes using the format of the Nutrition Education Evaluation Reporting System (NEERS5) federally-mandated for EFNEP [84]. Item selection was based on HCHF objectives and input from an expert panel of nutrition faculty and extension professionals. The instrument underwent cognitive testing with the target population and further revision. It is feasible to use, sensitive to pre-post change in reported behaviors, had good test-retest reliability (r=0.83), and convergent validity (P<0.001) with in-depth, validated measures of modeling (r= 0.56), adult (r=0.48) and child (r=0.52) dietary intake, and physical activity (r=0.44).Comprehensive Feeding Practices Questionnaire will be used to assess parenting practices (Musher-Eizenman & Holub, 2007). This instrument was developed to assess parenting practices of young children. Items were generated from existing literature around feeding practices, then subjected to exploratory and confirmatory factor analysis, and assessment of internal consistency of resulting scales. Of the 12 scales, we will use the six (24 items) that most closely align with HCHF: encouraging balance and variety in which parents promote well-balanced food intake, parent use of food as a reward for child's behavior, parent allowing child to control eating, parental modeling of healthy eating, pressure from the parent for child to consume more food at meals, and the availability of a healthy foods in the home environment.Parental self-efficacy for obesity prevention related behaviors will be assessed with an instrument focusing on parents' confidence they can help their children with specific behaviors when various barriers are in place, e.g., "when you have too many things to worry about" (Wright et al., 2014). An item pool was generated based on theory, interviews with parents, and content experts, then subjected to factor analysis to identify scales. Construct validity was assessed by comparing the child's behavior to parent efficacy related to that behavior. All correlations were significant in expected direction. Measures of internal consistency (α > .8) and test-retest reliability (r > .8) were good. We will use the three 4-item scales that most closely align with HCHF, including physical activity, fruit intake, vegetable intake, and sugar sweetened beverage intake.Food Behavior Checklist was developed and validated with adult SNAP-Ed audiences and will be used to assess parent food behaviors. Two scales that align with HCHF will be used to assess behavior change pre- to post-intervention (Murphy et al., 2001). A 7-item item scale assesses fruit and vegetable intake, and a 4-item diet quality scale assesses intake of sugar sweetened beverages and overall diet quality. This instrument was originally validated against serum carotenoids as a biomarker of fruit and vegetable intake with high carotenoid correlations of r=0.28-0.45 (p<.001) for fruits, vegetables, sugar sweetened beverages, and overall diet quality. Convergent validity with dietary recall data was also statistically significant.Child Food and Beverage Intake Questionnaire was developed in a low-income population for parent report of preschool children's intake (Koleilat & Whaley, 2016). This 10-item food frequency focuses on fruits, vegetables, sugar-sweetened beverages, and sweetened foods, which align with HCHF. Validity was assessed through comparison with three 24-hour recalls, with intraclass correlations ranging from .49 to .56 for the four groups.Data analysis. Chi-square analysis will be used to identify demographic differences among groups Repeated measures analyses of variance will analyze differences among data collection points within a group. Multiple regression analyses will be performed to examine the outcome results while controlling for sample characteristics of the sample, i.e. demographic characteristics, specific research site, specific educator, and cohort. Analytical methods will be employed to control for intercluster correlation and cluster randomized design, including proc mixed analysis.

Progress 10/04/16 to 12/30/19

Outputs
Target Audience:The target audience was low-income parents of 3-5 year old children in Head Start in New York City. Daycares serving this population were recruited, and then staff in these daycares assisted project staff in recruiting parents to participate in the study. Half of the daycares and subsequent groups of parents spoke Spanish and half spoke English as a primary language. Participating parents with multiple children were asked to identify one child between the ages of 3 and 5 years as a focus for data collection. Changes/Problems:Numbers of participants recruited and retained was low in two sites during year 1. In one of these sites, the intervention was not able to be completed in the immediate intervention group. While 7 participants were recruited into this group, classes had to be canceled 3 times for lack of attendance. For the 3 classes held, only 2 or 3 parents attended, and they were different individuals each time. Participants must attend at least 6 of the 8 classes in the HCHF series to be considered to have completed the series. Therefore, the research team made the decision to discontinue the intervention for this site. The delayed intervention group proceeded as planned, with 6 of the 8 recruited participants successfully completing the series. Both of these sites were discontinued and replaced in year 2 , and an extra site added in order to reach the estimated sample size. What opportunities for training and professional development has the project provided?This project provided the opportunity for Cooperative Extension staff to be involved in a carefully designed randomized control trial. Formal training was held annually that included both research and extension staff. This integrated approach allowed Extension staff, in particular, to learn about research methodology, both formally and informally. The undergraduate and graduate students learned about Cooperative Extension/EFNEP programming, with the undergraduate learning about and practicing qualitative data collection and analysis, while the graduate student learned about quantitative data collection, triangulation, and analysis. How have the results been disseminated to communities of interest?A report designed for the public was developed for the staff at the Head Start sites involved in the project. This was provided to the Cooperative Extension educators working with these sites. A training was held for the educators in which the project results were reported to them, along with this report. They have subsequently passed the reports on to the Head Start staff with an explanation and time for questions to be answered. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Childhood obesity is a recognized public health threat in the United States. Prevalence has been rising since the 1980s, increases as children age, and is greater among low-income children. Recent evidence from the Centers for Disease Control and Prevention indicates that 13.9% of 2-5 year olds are obese, putting these children at risk for chronic health conditions. Poor dietary habits and lack of physical activity are the proximal causes of obesity. These behaviors stem from habits formed early, and once formed are challenging to change. Parents are the child's primary role models and teachers. Along with other caregivers, parents need to demonstrate healthy eating and physical activity behaviors, and to provide the guidance and support children need to integrate these behaviors into their lives in order to grow into healthy adults who are not overweight or obese. However, parents often lack the necessary knowledge, skills, and support to accomplish these goals. Healthy Children, Healthy Families: Parents Making a Difference! (HCHF) was designed to address these gaps. Researchers at Cornell University worked with Cooperative Extension nutrition educators across New York State to develop and test HCHF, a research-based workshop series for low-income parents of young children. Drawing on relevant research and grounded in theory, HCHF teaches parents and caregivers (hereafter referred to as parents) to model healthy behaviors, guide their children in making healthy choices, and provide supportive environments for their children that include healthy food and activity choices. HCHF focuses on promoting behaviors identified as most relevant to preventing unhealthy weight gain and diet-related chronic disease, such as limiting sugar-sweetened beverages and other high fat/high sugar foods; increasing consumption of fruits, vegetables and whole grains; and increasing physical activity. This carefully designed study was conducted with 391 diverse English- and Spanish-speaking parents of 3-5 year old Head Start students to test the effectiveness of HCHF in helping parents create and sustain healthy habits for themselves and their families. Parents who participated reported positive changes in food and activity habits for themselves and their children. Results indicate that HCHF can effectively help parents make changes for themselves and their families in order to prevent unhealthy weight gain and subsequent chronic disease. To test the effectiveness of HCHF in changing behaviors of parents and children, a rigorous research design was employed, with low-income parents (n=391; 222 Spanish-speaking, 169 English-speaking) from 11 Head Start sites enrolled. Objectives were to (1) determine the impacts of HCHF on participants' use of effective parenting practices to influence children's healthy eating and active play behavior and (2) investigate retention of these changes up to18 weeks after HCHF sessions ended. In each site, half the participants were randomized to immediate education (IE, n=195) and half to delayed education (DE, n=196). The DE group served as a control for the IE group. There were no demographic differences (gender, race/ethnicity, education, age, family size, working outside the home) between the IE and DE groups. Data were collected 4 times, approximately 9 weeks apart: for IE this occurred at baseline (pre-), post-education, 9 and 18 weeks follow up; for DE this occurred at baseline, immediately pre-, post-education, and 9 weeks follow up. Longitudinal data allowed for an assessment of maintenance of behavior change up to 18 weeks. At each data collection point, parents completed a written survey in which they reported their own and their child's behaviors. The HCHF Checklist, the primary outcome assessment, was designed to accompany the HCHF intervention and has previously been validated against other instruments. Five other instruments, validated in studies with similar populations, were used to triangulate results. The HCHF Checklist asked about frequency of intake of fruits, vegetables, low-fat dairy, and soda for parents and their children; of participation in physical activity for parents and children and in screen time for children; of the child eating take-out or fast food; and of availability of fruit and of sweets in the home. Parenting practices were assessed via frequency of parents eating with children, and of children being allowed to decide how much to eat. Results indicated that participants reported healthier practices post-education as compared to pre-education in both the IE and DE groups (p<0.05), as well as post-education (IE) compared to the 9-week control period in DE (p<0.01). Improved healthy practices were maintained 9 and 18 weeks post-education. The Parental Self-efficacy for Obesity Prevention instrument provided data on parents' confidence that they can support their children in making healthy food (fruit, vegetable, and sugar-sweetened beverage) and physical activity choices in the context of situational barriers. Results indicated that participants in both IE and DE were more confident in supporting their children in making healthy choices after attending HCHF (p<0.001). Behavioral changes were maintained at 9 and 18 weeks post-education. The Comprehensive Feeding Practices Questionnaire assessed parental practices to encourage balance and variety and promote well-balanced food intake among children, parent's use of food as a reward for child's behavior, parent allowing child to control eating, parental modeling of healthy eating, pressure from the parent for child to consume more food at meals, and availability of healthy foods in the home environment. The IE group reported a greater positive change after completing HCHF (p<0.0001) as compared to the DE group with no intervention (p<0.01). The DE group reported a further change post-intervention, as compared to the end of the control period (p<0.001). Behavioral changes were maintained at 9 and 18 weeks post-education. Scales from the Food Behavior Checklist were used to assess parent's fruit and vegetable intake, sugar-sweetened beverage intake, and overall diet quality. The summative score indicated the IE group reported a greater positive change after completing HCHF (p<0.0001), as compared to the DE group with no intervention (p<0.05). The DE group reported a further change post-intervention, as compared to the end of the control period (p<0.0001). Behavioral changes were maintained at 9 and 18 weeks post-education. Child Food and Beverage Intake Questionnaire was used for parents to report their child's frequency of intake of vegetables, sugar-sweetened beverages, and other sweetened foods. The IE group reported significantly healthier scores post- as compared to pre-education (p<.01), with changes maintained at each subsequent time point. DE scores did not show significant changes at any time point. In summary, the HCHF intervention was successful in changing behaviors of parents and children according to the HCHF Checklist, the main outcome measure. Parents reported that these behavior changes were maintained for 18 weeks post-intervention. Baseline values were not significantly different between the IE and DE groups confirming adequate randomization. Using the delayed intervention model, each group was its own control (i.e. baseline data with no intervention). In addition, a real-time control was provided by the DE group during the first period when they received no intervention while the IE group was receiving HCHF. This confirms that changes in reported behavior were due to the intervention itself rather than secular changes or social desirability of participant responses. Other instruments successfully supported these results, with the possible exception of the Child Food and Beverage Questionnaire, although even here results in the IE group were consistent with the HCHF Checklist.

Publications

  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Hill T, Dickin K, Scott-Pierce M, Shapiro P, Parker C, Dollahite J. Healthy Children, Healthy Families: Parents Making a Difference: A Randomized Controlled Trial. J Nutr Ed Behav 2019;51:S110-111.


Progress 10/01/17 to 09/30/18

Outputs
Target Audience:The target audience is low-income parents with children 3-5 years old living in New York City. Six of the 8 Head Start sites from Year 1, and 3 additional daycare sites were recruited, with a total of 192 additional parent participants enrolled in the study. Parents in 5 of the sites were English speaking and in 4 of the sites were Spanish speaking. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project has provided the opportunity for Cooperative Extension staff to be involved in a carefully designed randomized control trial. At the beginning of Year 2, an additional training (beyond that provided at project initiation) was provided to staff to review and reinforce the research methodology integral to this trial. The undergraduate students who conducted and coded the participant interviews were able to learn about Cooperative Extension/EFNEP programming, as well as to learn about and practice qualitative data collection and analysis. The graduate student is learning about Cooperative Extension/EFNEP, randomized control trials, and data analyses. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?The graduate student will be writing a methods paper. No additional data will be collected. The analysis plan for quantitative data is under review. The graduate student will conduct analysis on the demographic characteristics of the group - immediate intervention and delayed intervention, within and between trial years, and across data collection periods. The remaining analyses that will provide outcomes of the HCHF intervention will be completed. Analysis of the qualitative data collected in year 1 will be completed. Ultimately those data will provide insights into the success of the intervention and will be used in the outcomes paper.

Impacts
What was accomplished under these goals? Six of the 8 Head Start sites from Year 1, and 3 additional sites were recruited, with a total of 192 additional parent participants enrolled in the study. In each site, half of the participants were randomized to the immediate intervention group (n=96) and half to the delayed intervention group (n=96). Participants attending at least 6 of the 8 intervention educational sessions, equivalent to graduation in EFNEP, included 44 in the immediate intervention group and 38 in the delayed intervention group. Data were collected 4 times, approximately 9 weeks apart, with the final data collection providing information about the retention of behavior change in the medium term. A similar number of participants provided data at all 4 data collection points (n=61 in immediate and n=58 in delayed intervention groups, respectively). An undergraduate student worked with another student and two faculty members to code the transcripts of participant interviews both to triangulate the quantitative data and to identify emergent themes related to participant experience of HCHF and the response of participants and their families to HCHF-related changes. Data analyses were begun.

Publications


    Progress 10/04/16 to 09/30/17

    Outputs
    Target Audience:The target audience is low-income parents of 3-5 year old children in daycare (primarily Head Start) in New York City. Daycares serving this population were recruited, and then staff in these daycares assisted project staff in recruiting parents to participate in the study. Half of the daycares and subsequent groups of parents spoke Spanish and half spoke English as a primary language. Participating parents with multiple children were asked to identify one child between the ages of 3 and 5 as a focus for data collection. Changes/Problems:Numbers of participants recruited and retained was low in two sites. In one of these sites, the intervention was not able to be completed in the immediate intervention group. While 7 participants were recruited into this group, classes had to be canceled 3 times for lack of attendance. For the 3 classes held, only 2 or 3 parents attended, and they were different individuals each time. Participants must attend at least 6 of the 8 classes in the HCHF series to be considered to have completed the series. Therefore, the research team made the decision to disband the intervention for this group. The delayed intervention group proceeded as planned, with 6 of the 8 recruited participants successfully completing the series. Both of these sites are being replaced in year 2, and an extra site added in hopes of reaching our estimated group sample size of 16 total groups (8 immediate intervention and 8 delayed intervention). No other changes are anticipated. What opportunities for training and professional development has the project provided?This project has provided the opportunity for extension staff to be involved in a carefully designed randomized control trial. A formal training was held at the beginning of the project that included both research and extension staff. This integrated approach has allowed extension staff, in particular, to learn about research methodology, both formally and informally as the trial has progressed. The undergraduate student was able to learn about Cooperative Extension/EFNEP programming, as well as to learn about and practice qualitative data collection and analysis. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?A second and final wave of daycares and participants will be recruited for the intervention during year 2. Six of the 8 daycares from year 1 will participate again and 3 new daycares will be added. The goal is to recruit at least 24 participants per site, with half randomized into immediate intervention and half into delayed intervention. Data collection will be conducted as it was done in year 1. A graduate student will work on analysis of the survey data for her master's thesis. She will be involved in writing a methods paper, as well as the results paper for the study. The undergraduate student who collected qualitative data in year 1 will complete the analysis of these data in year 2. Ultimately those data will provide insights into the success of the intervention and will be used in the outcomes paper.

    Impacts
    What was accomplished under these goals? The design of this randomized control trial includes two groups, one that receives the intervention immediately following baseline data collection, and the other that serves as a control between baseline and data collection 2. Eight daycare sites were recruited, with a total of 199 parent participants recruited into the study. In each site, half of the participants were randomized to the immediate intervention group (n=99) and half to the delayed intervention group (n=100). Participants attending at least 6 of the 8 intervention educational sessions, equivalent to graduation in EFNEP, included 58 in the immediate intervention group and 54 in the delayed intervention group. Data were collected 4 times, approximately 9 weeks apart, with the final data collection providing information about the retention of behavior change in the medium term. A similar number of participants provided data at all 4 data collection points (n=54 in immediate and n=56 in delayed intervention groups, respectively). Data analysis will begin in year 2. An undergraduate student received summer internship funding to conduct in-person, qualitative interviews with participants who had completed the intervention. Her study was designed to understand, in-depth, if and how participants were incorporating new practices into their parenting to influence their child's eating and active play behaviors. She was able to conduct in-depth interviews with 20 participants over the summer and is now working with faculty to analyze the results.

    Publications