A randomized controlled study design was used to examine the efficacy of the ABC study's program. Children in the study completed questionnaires regarding their levels of physical activity, dietary intake, usual food choices, knowledge of nutrition and physical activity and self-efficacy related to physical activity and healthy food choices at baseline (T0) and 2 months (T1), 6 months (T2) and 8 months (T3) after the baseline assessment. The primary caregiver completed questionnaires regarding parents' demographic information and levels of acculturation. Eight to 10-year-old Chinese American children who were normal weight or overweight and their parents were eligible for enrollment if they met the following criteria: (1) The adult and child self-identify ethnicity as Chinese or of Chinese origin, and they reside in the same household. A dyad of one adult and one child was the minimum necessary for a household to participate. Two adults per child were encouraged to participate. (2) The child was able to speak and read English. (3) The child was in good health, defined as free of an acute or life-threatening disease and able to attend to activities of daily living such as going to school. Children with chronic health problems that included any dietary modifications or activity limitations (e.g. diabetes, exercise-induced asthma) were excluded. (4) Parents were able to speak English, Mandarin or Cantonese and were able to read in English or Chinese and to complete questionnaires. Overweight is defined as a child having BMI ≥85th percentile whereas normal weight is a child having a BMI between 5th–84th percentile for his/her age and gender, based on the Centers for Disease Control and Prevention (CDC) growth chart. Data were collected from September 2006 to December 2008.
Study procedure and intervention
Upon approval from the University of California, San Francisco, Committee on Human Research, 8- to 10-year-old children who self-identified as Chinese and their mothers were invited to participate in this study. Participants were recruited from Chinese language programs in the San Francisco Bay area. Research assistants described the study to potential children and gave them an introduction letter and research consent form to take home to their parents. Parents who were interested in the study signed and returned the consent form, providing their names and contact information to the research team. Children and parents were informed that they could refuse to participate or withdraw from the study at any time.
After parents gave informed consent and children provided verbal assent, baseline data were collected, and children and parents were randomly assigned to the intervention group or the waiting list control group by a computer-generated random number assignment (Fig. ). Research assistants went to the study site and administered all of the questionnaires for children to complete. Children also had their weight, height, waist and hip circumferences and blood pressure measured. Parents completed questionnaires at home and mailed them back to the research team within 2-weeks of receiving them. Children assigned to the intervention group participated in small group weekly session activities for 8 weeks, and parents in the intervention group participated in two small group workshops in 8 weeks.
Waiting-list control group children and families participated in the data collection activities at the same time as those in the intervention group. After completing the final follow-up assessment, this group received the ABC study intervention.
ABC study intervention overview
The intervention is based on the social cognitive theory developed by Bandura.18–20
The theory proposes that cognitive factors, behavioral factors (including other personal factors such as preferences and competencies), and environmental influences are interactive and integrated determinants. The learning and social influences stem both directly and vicariously from numerous resources, including parents. In this model of reciprocal causation, cognitive, behavioral and environmental factors all operate as interrelating determinants. Several key concepts such as self-efficacy, outcome expectation, skill mastery and self-regulation capabilities included in the social cognitive theory are used to explain and predict a person's behavior.18–20
This study intervention addressed these concepts by attempting to increase children's and parents' self-efficacy through setting realistic and achievable goals, providing necessary skills to achieve mastery, and improving self-regulation in maintaining healthy weight and healthy lifestyles.
Once four to six families had been randomized to the intervention group, the parents and children met separately for small-group sessions. Children participated in a 45-min session once each week for 8 weeks, and parents participated in two sessions that lasted 2 h each session during that 8 weeks. Children took part in a play-based workshop facilitated by a bicultural/bilingual research assistant. The intervention program consisted of educational play-based activities that increase children's self-efficacy and facilitate their understanding and use of critical thinking and problem-solving skills related to nutrition, physical activity and coping. The intervention for children was designed to improve their self-efficacy and self-competence via interactive activities (such as games and play), and to promote internal motivation to change health behaviors and maintain a healthy weight. For example, children learned how to select healthy meals and will be involved in role-playing and practice sessions about options and choices related to high-sugar and high-fat foods. An interactive dietary preparation software program tailored to common Chinese foods that was developed by Joslin Diabetes Center Asian American Diabetes Initiative was used for this study.21
In each session, children engaged in lessons related to nutrition, physical activity and critical thinking (See Table for overview of the intervention program components).
Intervention program for children
At the beginning of each session, children spent 15 min on physical activities. Physical activity sessions were aimed at increasing children's energy expenditure by implementing a 15-min activity session each week for 8 weeks. They engaged in different types of non-competitive activities (such as dance, brisk walking and jump rope), learned types of activities that they can do during recess and at home, and also learned alternatives to watching television. In addition, children received a pedometer, activity diary and books related to physical activity. For the remaining 30 min of each session, we focused on children's knowledge regarding nutrition and physical activity and reinforced the notion of self-efficacy regarding food choices and alternatives to high-fat and high-sugar foods and television viewing. Children received a food diary to record their food intake, books related to healthy eating and a packet of materials in both Chinese and English each week explaining the activities that highlight healthy eating and active lifestyles.
A family component (two 2-h sessions) was incorporated into this study to provide reinforcement and social support at home for the education received during the study. The parents took part in ‘Healthy Eating and Healthy Family: A Hands-on Workshop’, which was led by a bilingual/bicultural registered dietitian. The group workshops (8–10 parents) included sets of exercises to increase parents' knowledge and skills regarding healthy food preparation, discussion of issues related to dealing with children's eating habits and problems and brainstorming about specific family/children activities to improve dietary intake and physical activity. The parent intervention included a workbook, video clips and discussion of techniques. In addition, parents were encouraged to allow children to make their own choices for agreeable foods and activities. Parents were also encouraged to involve their children in grocery shopping and meal preparation.
After completion of each of the data collection activities, parents and children received a $10 gift certificate. Upon completion of the study, they received a $30 gift certificate. At the 2 month (T1), 6 month (T2) and 8 month (T3) assessments, the children completed questionnaires regarding their dietary intake, usual food choices, knowledge about nutrition and physical activity and self-efficacy regarding physical activity and food choice. The children also wore a Caltrac personal activity computer to measure their physical activity and had their weight, height, blood pressure and waist and hip circumferences measured at all the assessments. Primary caregivers (all mothers) completed questionnaires regarding their demographic information and acculturation level at baseline. Children completed their questionnaires in English, and parents completed questionnaires in either Chinese or English.
The 12-item parent questionnaire includes parent(s)' and children's ages, parents' weights and heights, parents' occupation(s), family income and parents' levels of education. The questionnaire was written at a third-grade reading level and took approximately 5 min to complete. Mothers completed this questionnaire at baseline.
Suinn-Lew Asian self-identity acculturation scale
The Suinn-Lew Asian self-identity acculturation scale (SL-ASIA) is used to examine the levels of maternal acculturation.22,23
The SL-ASIA scale is a 21-item multiple-choice questionnaire covering topics such as language (4 items), identity (4 items), friendships (4 items), behaviors (5 items), general and geographic background (3 items) and attitudes (1 item). Scores could range from a low of 1.00, indicative of low acculturation or higher Asian identity, to a high of 5.00, indicative of high acculturation or high Western identity. The scale also permits classification as ‘bicultural’, indicating that a person has adopted some Asian values, beliefs and attitudes along with some Western values, beliefs and attitudes. Validity and moderate to good reliability have been reported.22,23
The Cronbach alpha for the SL-ASIA was 0.79–0.91 for Chinese Americans.22,23
This questionnaire is available in Chinese and English and is written at a fifth-grade reading level. It takes about 10 to 15 min to complete the questionnaire.
BMI has a well-established association with stature and age among children and adolescents.24,25
BMI was calculated by dividing body mass in kilograms by height in meters squared (kg/m2
). BMI has acceptable ranges of sensitivities and specificity. Sensitivity ranged from 29 to 88%, specificity ranged from 94 to 100% and predictive value ranged from 90 to 100%, in children24,25
In this study, BMI lower than the 5th percentile was defined as underweight, between the 6th and 84th percentile was defined as normal weight and BMI above the 85th percentile was defined as overweight and 95th percentile as obese, based on the growth chart developed by CDC.
The waist-to-hip ratio was derived from the waist and hip circumferences. Waist circumference was measured midway between the lowest rib and the superior border of the iliac crest. Hip circumference was measured at the maximal protrusion of the buttocks. The circumferences were given as the mean of two measurements to the nearest 0.1 cm.
Systolic and diastolic blood pressure (DBP) were measured by using a mercury sphygmomanometer with specific cuff size appropriate for children (Baumanometer, W.A. Baum Co, Copiague, New York) to the nearest 2 mm Hg, twice in the child's right arm, with the child seated after 10 min of rest.
Caltrac personal activity computer
The Caltrac has been widely used in assessing physical activity among children and adults.26
Although there are other advanced measures of physical activity (i.e. accelerometer), we chose to use Caltrac in our study because of its relatively low cost ($70) and ease of use. The Caltrac is designed to be placed at the hip and to measure vertical acceleration. Readings from the device have been used to predict oxygen consumption and net caloric expenditure, based on the user's age, height and weight, during exercise. The Caltrac has a moderate to high validity, ranging from 0.35 to 0.97, with heart rate and observation methods.26
A high reliability of the device, ranging from 0.87 to 0.98, was also reported in children.26
In this study, children were instructed to place the Caltrac at the hip for 3 consecutive days (two weekdays and one weekend day). Children were also instructed to wear the Caltrac as soon as getting up from the bed in the morning and remove it only in shower, in water-related activities (i.e. swimming) and sleeping. Average count was used for analysis.
Three-day food diary
This quantitative self-report food diary was used to estimate dietary intake of children. Children received a 3-day food diary containing an instruction sheet, a sample completing day's food-record sheet and eight blank white dietary record forms. Each form included spaces for the child's name, day of the week, date of recording and blank lines to record food and drink grouped into the following categories: breakfast, snack, lunch, snack, dinner and snack. Children were asked to record all foods and beverages consumed for 3 consecutive days. They also recorded serving sizes. Kappa coefficients and percent of agreement for interobserver reliability ranged from 0.43 to 0.91.27,28
The diary takes approximately 10 min to complete. Parents were asked to complete the diary together with their children to increase the accuracy of dietary report.
Usual food choices
This 14-item survey was part of the Health Behavior Questionnaire developed for the Child and Adolescent Trial for Cardiovascular Health (CATCH) study. This survey asked about usual food choices (behavior) in a forced-choice format that focuses on low-fat and low-sodium foods. It measured usual food selections and what types of food a child eats most of the time. Children were given a choice between two foods and asked which one they eat more often. Sample questions are ‘Which foods do you eat most of the time: hot dog or chicken? Frozen yogurt or ice cream?’ A higher score indicated more healthy food choices. Validity was obtained by including expert review and pilot testing with a focus group. The alpha coefficient for internal consistency in the original study was 0.76.29
Physical activity knowledge
This five-item questionnaire was developed by the researcher to assess children's knowledge about physical activity. Items were adapted from recommendations from the US Department of Agriculture30
and the American Heart Association31
regarding dietary guideline, MyPyramid and children's health. Sample questions included the following: How much aerobic activity is required for a healthy heart? How many hours a day should a child watch television or play video games? The reliability coefficient for internal consistency with the sample of children in this study was 0.65. Children received one point for every question they answered correctly. A higher score indicates more accurate knowledge about physical activity needs. The total score was used for analysis.
This 14-item survey also was part of the Health Behavior Questionnaire developed for the CATCH study. It measured children's knowledge about healthy food choices. Children were asked to identify which food was ‘better for your health’. Samples of two choices included ‘whole wheat or white bread’ and ‘frozen corn or canned corn’. Content validity was examined by including expert review and focus group pilot testing from the CATCH study. This survey had a reported internal consistency ranging from 0.76 to 0.78.29
A higher score indicates more accurate dietary knowledge.
Child dietary self-efficacy
This 15-item self-report questionnaire measured children's self-confidence in their ability to choose foods low in fat and sugar.29
The questionnaire contained 15-item stems beginning with ‘How sure are you… ?’ Items were scored on a Likert scale, with options of ‘not sure’, ‘a little sure’ or ‘very sure’. Higher scores indicated higher self-efficacy. The internal consistency ranged from 0.82 to 0.87 in the third and fifth graders.29,32
The questionnaire took about five to 10 min for children to complete.
Physical activity self-efficacy
This subscale of the Health Behavior Questionnaire was used to measure the children's self-confidence in their ability to participate in various age-appropriate physical activities.29,32
The subscale included five items in which children were asked if they were ‘not sure’, ‘a little sure’ or ‘very sure’ that they could do such things as ‘keep up a steady pace without stopping for 15–20 min’. Higher scores indicated higher self-efficacy. Internal consistency ranged from 0.67 to 0.69 in the third and fifth graders.29,32
The questionnaire took about two to 5 min for children to complete.
Descriptive statistics were calculated initially for demographic characteristics and all major study variables. We used t-tests to examine any differences in variables between intervention and control groups at the baseline. We examined whether the rate of change across the four data collection time points was different for the children in the intervention group than for children in the control group by fitting linear mixed-effects models that included functions of time and group effects to the repeated child data. With a continuous outcome, the mixed-model approach to analyze longitudinal data is a method of modeling population parameters as fixed effects while simultaneously modeling individual subject parameters as random effects. The modeling of individual subjects was obtained as random deviations about the population model. Follow up t-tests were used to examine the efficacy of the intervention on three follow-up time (T0–T1, T0–T2, T0–T3). All analyses were performed in SPSS 15.0, with 0.05 set as the required level of significance.
A minimum of 30 participants per group provides power of 0.85 at an alpha of 0.05 to detect a moderate difference in BMI between the two groups between baseline and 8-month follow-up (mean BMI difference 0.8 between the groups which reflects a difference of 1 of a standard deviation [SD]).