Food Practices and Preferences in Youth With Diabetes
Objectives. To determine where children with type 1 diabetes eat and with whom, how foods are prepared, their favorite foods, and food issues or conflicts regarding both portioning foods and types of food eaten and to examine the multiple factors that influence food intake in this population.
Methods. Parent-child pairs (n = 299) completed a questionnaire that focused on where and with whom children eat and how their food is prepared. χ2 tests and ANOVA were calculated to determine differences between child and parent responses.
Results. As reported by parents, children (78%) eat meals at home with the family 4–6 or more meals per week; 21% eat at home alone 4–6 or more meals per week; 27% eat at fastfood restaurants 2–3 or more meals per week, and 72% eat school lunch 2–3 or more times per week (58% almost every day). Confusion regarding portion skills was reported by 13%, 52% measure by “eye balling” only (estimating portions sizes), and 47% would like more information on portioning. Among parents, 37% reported disagreements with children over choices of food, and 28% reported disagreements over food portion sizes; 5% did not care about food portioning at all. Cooking from scratch was reported by 71% to occur ≥ 4 times per week (out of 21 meals per week), and 65% reported cooking prepackaged meals at least 2–3 times per week.
Conclusions. Suggestions for eating at fast-food restaurants and schools; food-portioning skills for home, school, restaurants, and homes of friends; cooking from scratch and from prepackaged foods; incorporating favorite foods; and reading labels on prepackaged foods should be addressed for effective educational sessions.
Medical nutrition therapy (MNT) is an important aspect of diabetes self-care management and diabetes self-care education.1 The American Diabetes Association position statement on MNT includes the following goals: 1) attaining/maintaining optimal metabolic outcomes (glucose and lipid levels); 2) prevention and treatment of chronic complications of diabetes; 3) improvement of health through healthy food choices and physical activity; and 4) taking individual nutritional needs into consideration.1 Choosing appropriate foods and food portion sizes is integral to achieving these goals, as is meeting the individual needs of clients. The Nutrition Subcommittee of the Diabetes Care Advisory Committee of Diabetes UK2 concurs with its American counterpart and adds that advice in counseling must respect an individual's wishes and willingness to change.
Many factors influence an individual's choice of foods. Food choices and managing healthy eating was studied by Falk et al.,3 who categorized seven themes for healthy eating: low-fat, natural/unprocessed foods, balance, prevention of disease, nutrient balance, disease management, and weight control. Besides health considerations, other values influencing food choices are convenience, taste, and cultural preferences.
When considering which factors influence a child's choice of foods, short-term gratification may be more important than long-term health outcomes.4 When adolescents' choices of foods were studied,5 the influences they reported were hunger, food cravings, food appeal, convenience of food, food availability, parental influence (culture or religion of the family), health benefits of food, body image, habit, cost, media, and vegetarian beliefs.
Because of the importance of nutrition for growth and long-term health promotion, nutrition programs need to be effective. Suggestions for effective education programs are to address a broad range of factors, including environmental factors such as food availability, and consideration of individual schedules, as well as acquiring knowledge and skills to appropriately choose appealing, convenient foods within homes, restaurants, and schools.4,6
Although there have been studies on nutrient consumption to help assess the epidemic of obesity in youth,7,8 research is lacking on food choices, attitudes about portioning foods, and techniques used for cooking in the childhood or adolescent population with diabetes. Cole et al.7 collected dietary recalls from 50 high school students on the Zuni Indian reservation. Meat, low-fiber grains, and sugared beverages were major contributors to adolescents' total energy intakes. Soft drinks contributed 27% of total carbohydrate intake for females and 21% for males. Williams et al.8 reported that in a group of 165 nondiabetic Pima Indians, the risk of developing diabetes was 2.5 times higher for the group consuming a western or Anglo diet instead of the traditional Indian diet (e.g., legumes, chili stew, tortillas, and squash). The influence of the food environment on preventing childhood obesity was reviewed by Haire-Joshu and Nanney,9 with the conclusion that parent modeling and availability of foods in the home, school, and at fast-food restaurants are critical to food choices for children. These studies did not examine food portioning or cooking techniques for the child and adolescent age groups.
Diabetes educators need to learn where they should focus their efforts in nutrition education. Diabetes educators have limited interaction time (especially time reimbursed by insurance) with patients to teach the numerous aspects of diabetes care. It is essential that diabetes educators focus on the important and useful aspects of nutrition that will be helpful to improving diabetes control.
The purposes of this study were to identify current eating habits of youth with type 1 diabetes by 1) determining where food is eaten, how food portioning is conducted and by whom, how often children/adolescents eat traditional versus convenience foods, favorite foods eaten, and food issues or conflicts in the home and 2) examining the multiple factors that influence food intake in this population.
Questionnaire. A questionnaire was developed to determine favorite food choices of children with type 1 diabetes, where they eat, and what type of cooking techniques are used. The open-ended questionnaire asked by frequency category (< 1 per month, 1 per month, 1 per week, 2–3 times per week, 4–6 times per week, or every day), where and with whom the children eat meals, how the food is cooked, and what measuring techniques are used. There was also an open-ended question asking for a list of favorite foods. Three test questions were included to assess knowledge of portion sizes (correct portion size to obtain a 15-g carbohydrate serving). The questionnaire for the child was worded to obtain information on their own behaviors, and the one for parents was worded to obtain information regarding their child's behavior.
Two nutritional epidemiologists critiqued the format of the questionnaire to enhance ease of use and accuracy of data collection. The questionnaire was then administered to 30 clinic patients in Utah to determine ease of use and face validity. Revisions were made to the questionnaire based on the feedback received to create the final form used for the study. It took approximately 10–15 minutes for parents to complete the questionnaire and 15–20 minutes for children to complete, unless the children needed assistance with reading. The questionnaire was administered to children and parents at the beginning of a diabetes camp session. This project was reviewed by the first author's institutional review board.
Subjects. Children with type 1 diabetes and their parents (n = 299) who attended summer diabetes camps in Tennessee, Utah, and Idaho participated in the study. Administering the questionnaire was incorporated into the camp check-in process, and no parent or child refused to complete it. A few children and their parents who arranged to come late to camp did not complete the questionnaire. Incorporating this study into the check-in process resulted in a high response rate.
The Tennessee camp was a 2-week overnight summer experience for children with diabetes between the ages of 9 and 15 years. The Utah and Idaho camps were 1-week overnight camps (four camp sessions per summer with campers separated by age group), with campers ranging in ages from 8 to 18 years. The average duration of diabetes for campers at all three locations was 5.1 ± 3.5 years, with 47% being male and 53% female.
Parents and children were asked to fill out the questionnaire in separate locations at camp check-in. Both parents and children were asked to fill out the questionnaire to compare differences in their responses. As calculated through the SMOG (simplified measure of gobbledygook) readability formula, the questionnaire was written at a 7.7-grade reading level (with long words, such as hypoglycemia, increasing the reading level).10 This knowledge prepared the researchers in advance to be ready to help younger children complete the questionnaire if required.
Counselors at the camps also filled out the questionnaire. Counselor ages ranged from 16 to 41 years, with a mean of 18.25 years and a median age of 17 years (n = 39, with 35 having diabetes). This counselor group was to serve as a control group to compare the effect of age on the response to questions. However, the sample size was small, and statistical comparisons were not possible to calculate.
Statistics. The questionnaires were matched for parent and child and were analyzed using Pearson's χ2 calculation. ANOVA was conducted between geographic locations and age groups to determine any differences. Correlations were calculated between the knowledge of portion sizes and frequency of measuring foods, cooking from scratch, cooking with prepackaged foods, and location of eating.
Results. The statistical analysis showed that there were no differences between parents' and children's responses when analyzed by geographic location (Tennessee, Utah, and Idaho) or younger versus older age groups of children. Specific results, as shown in Table 1, revealed that when parents and children were asked the frequency of where children eat meals, 78% of the parents (parents' responses are listed here, but all participant responses are listed in the tables) reported eating meals at home with the family 4–6 or more meals per week, whereas 21% eat at home alone 4–6 meals per week (43% eat alone at home 2–3 or more times per week). The frequency of eating at fast-food restaurants was 27% for 2–3 or more per week, and 6% reported eating in sit-down restaurants 2–3 or more times per week. A high percentage of meals were eaten at school, with 72% eating school lunch 2–3 or more times per week (58% almost every day), and 34% bringing a brown-bag lunch 2–3 or more times per week (only 22% almost every day).
There were statistically significant differences between child and parent responses on the questions regarding meal frequency at home and eating at a friend's home, with parents reporting a higher frequency of eating at home and children reporting a higher frequency of eating at a friend's home. There were also statistically significant differences between parent and child responses when reporting the frequency of eating at ball parks, from vending machines, or at a quick stop store. The practical importance of this is unknown, considering that parents reported eating ≥ 1 time per week at the ball park, from a vending machine, or from a quick stop store only 7, 12, and 13% of the time, respectively.
Table 2 lists the specific responses of parents to attitudes, skills, and knowledge of food portioning. Thirteen percent reported being confused about food portion skills, 52% measured by “eye balling” only (guessing portion sizes), and 47% perceived the need for more education on portioning. Regarding conflicts over food between parents and children, 37% reported fighting/disagreeing about types of food the child should eat, and 28% reported fighting/disagreeing over food portion sizes. Also, 5% of parents stated that they did not care about food portioning at all. There was a statistical difference between child and parent responses for all questions related to attitudes, skills, and knowledge about food portions.
Responses to the cooking methods question are reported in Table 3. Among parents, 71% reported cooking from scratch (questionnaire examples included mashed or baked potatoes, broiled chicken breasts, and steamed broccoli) in the home 4 or more times per week (out of 21 meals per week) compared with 46% of children, and 65% cooked prepackaged meals (examples on the questionnaire were frozen pizza, microwave dinners, frozen burritos, and canned soups) 2–3 or more times per week.
Table 3 also lists responses to questions about the use of measuring tools. The following tools were used at least 4–6 times per week: measuring cups (39%), scales (8%), diabetes exchange books (21%), and label reading (88%). There was a statistical difference between child and parent responses for frequency of cooking from scratch, cooking from prepackaged foods, and reading food labels. Carbohydrate-counting books were mislabeled as calorie-counting books on the questionnaire, which resulted in both parent and child reporting infrequent use.
When asked how often their child with diabetes has an appointment with a dietitian, parents' responses were: every 3 months (36%); every 6 months (27%); once a year (13%); one time ever (23%); and never (nutrition information given by doctor or nurse) (2%). There was no significant difference between parents and children regarding the number of times per year the child had an appointment with the dietitian.
Responses regarding where measuring skills are used are shown in Table 4. When eating at home, parents reported that the child measures foods 97% of the time; 34% when at school; 40% at restaurants; and 51% of the time at a friend's home. Table 5 shows that the mother is the person who does the cooking at home most of the time (66%).
χ2 tests between 299 matched parent-to-child questionnaires revealed that parents reported eating more meals at home with the family. In contrast, the children reported eating more frequently at a friend's home, the ball park, out of vending machines, and at quick stop stores. Parents reported reading labels more frequently, more conflicts in choosing/eating various types of foods and in choosing portion sizes for food, and cooking from scratch and cooking from prepackaged foods more frequently. Parents were also significantly more likely to request additional education on portioning skills.
Counselors (n = 39) reported more frequent use of measuring cups but less use of food labels than parents or children, as assessed from the raw data. Although the sample size for counselors was too small to include in the statistical analysis, responses are included in the tables. The counselor data are provided to show any differences (trends in raw data) that age may contribute in food-portioning habits and attitude differences between older adolescents/young adults with diabetes and younger people with diabetes.
Responding to an open-ended question, children reported their eight favorite foods in order of preference as pizza, chicken, noodles, potatoes, fruit, ice cream, spaghetti, and hamburgers. Parents listed the same favorite foods for their children in the same order, except ice cream was displaced to the end of the list.
Knowledge on portion sizes was assessed by asking for the correct portion size to provide 15 g of carbohydrate for rice, vanilla wafers, and potato chips (camp favorites). One-third–cup or 1/2-cup servings were scored as correct for rice, 5 or 6 as correct for vanilla wafers, and 1 oz or 15 potato chips as correct serving sizes for chips. Incorrect answers were 3/4 cup or 1 cup of rice, 8 or 10 vanilla wafers, and 1/2, 2, or 3 oz of potato chips. The percentage of correct answers for parents was 88.5, 80.2, and 87.2% for rice, vanilla wafers, and potato chips, respectively. The corresponding percentage of correct responses for children ages 12–18 years was 77.8, 83.4, and 77.0 %, respectively, and, for children ages 8–11 years, 61.7, 72.5, and 66.4%, respectively. χ2 analysis showed parents scored significantly higher on the portion size questions than both the older and younger children except for the older children on vanilla wafers. Adolescents scored higher, but not statistically higher, for the vanilla wafer portion size question. Older children scored significantly higher than younger children on all three food portion questions.
When asked whether the children adjust insulin for extra carbohydrate in their meal (use insulin-to-carbohydrate ratios and correction factors), 85% of the children and 90% of the parents reported that they did.
There was no correlation between the portion sizes knowledge score and frequency of measuring foods, cooking from scratch or prepackaged foods, or places of eating. There was a negative correlation between frequency of visits to the dietitian and increased perception of confusion regarding portion sizes (r = –0.134, P = 0.018).
Regularly using the skill of accurate food portioning has been shown to be cost-effective in managing diabetes.11 The community diabetes program,11 which emphasized food portioning skills, label reading, and cooking skills, showed a significant decrease in hemoglobin A1c and a significant cost savings in hospital charges according to the scale set by Gilmer et al.12
In addition to accurate portioning skills, it is also important to consider convenience and family influences on dietary, caloric, and carbohydrate intake. The results of this study were similar to one assessing the nutrient intake and food choices of adolescents when frequently eating at fast-food restaurants.13 In the current study, 25% of parents reported that the child ate at a fast-food restaurant 2–3 times per week, which is comparable to French et al.'s13 study, in which 26% of 8th grade boys and 22% of 8th grade girls reported this habit. The results of French's nutrient analysis revealed that total energy intake was 40% higher for males and 37% higher for females, and energy from fat was 9% higher for males and 13% higher in females when they frequented fast-food restaurants 2–3 times per week compared with those who seldom ate at a fast-food restaurant. In this study, 78% of parents and 60% of children reported eating at home with the family 4–6 or more times per week, which compares to Fulkerson et al.'s14 findings. She reported 75% of parents and 68% of adolescents in grades 7–12 consumed family meals 3 or more times per week.
An additional reason education on food portion skills continues to be an important factor in diabetes education and diabetes control is that the frequency of snacking in children is increasing.15 Children with diabetes will engage in social snacking with friends (with and without diabetes) more frequently than they might have 20–30 years ago. Using the Continuing Survey of Food Intake by Individuals,15,16 the frequency of snacking increased from 1.56 to 1.99 times per day, and the caloric intake increased ∼ 150 calories per day for children ages 2–18 years between 1977 and 1996. Snacks are encouraged in diabetes management and have the potential to alter diabetes and weight control. Results from this study included parents reporting that their child eats meals or snacks at the ball park at least once per week 7% of the time, from a vending machine 12% of the time, and at a quick stop store 13% of the time. Frequency and total snack calories should be considered along with the carbohydrate from full meals, blood glucose readings, exercise, and medication adjustments to achieve tight control of diabetes.
The increasing portion sizes of food offered at restaurants and in the home is another reason children with diabetes should be taught food portioning skills. The results of this survey of campers at three diabetes camps showed that 52% used visual estimation (eye balling) of portion sizes only. Stated in another format in a subsequent question, 43% of the parents reported that their child used a measuring cup once per week or less. The ability to accurately portion foods should be valued highly, considering how serving sizes delivered or offered at restaurants have increased. The American Cancer Institute commissioned a survey in 1999 (called the “Eyeball Method” Survey) to raise awareness of how large portion sizes had become.17 Although restaurant portion sizes have increased from 25% to three times what they were a decade ago, 62% of the 1,003 respondents believed restaurant food portions were the same size or smaller during the specified time frame. In addition, 80% of responders perceived that portion sizes at home were the same size or smaller than they were a decade before. Survey respondents were aware that the rate of obesity was increasing but were blind to the higher consumption of calories.
Data from this questionnaire were obtained from children with type 1 diabetes and their parents, with results in frequency of family meals and the number of times eating at fast-food restaurants being similar to other reported studies.13,14 Using this information for children with type 2 diabetes or for children with type 1 diabetes plus insulin resistance may also be helpful because it provides a better understanding of where food is eaten and with whom, the frequency of using measuring utensils, and knowledge of correct portion sizes. Certified diabetes educators should address the ease of obtaining large portion sizes, ease of convenience food, eating away from home (often at fast-food restaurants), caloric load from soft drinks,5,7–9,13–15 and importance of physical activity.
Data for this study were collected from a self-reported survey of eating behaviors. Memory of usual habits may be skewed by recall of previous teaching and knowledge or the desire to report good behavior. Camper data may have been affected by the desire to begin participation in camp activities rather than spend time filling out a questionnaire. The questionnaire was read to younger campers if there was any indication that they were having trouble answering the questions.
As children become more self-reliant, appropriate skills in diabetes management should be taught. Children are independently making food choices and food portioning decisions at school and at the homes of friends. Along with blood glucose monitoring and adjusting insulin for blood glucose levels out of target range, food portioning skills should be taught to both parents and children to aid in their decision-making process. Carbohydrate counting can only be done with accurate food portioning. Because a child or adolescent may value short-term gratification when making food choices over long-term benefits of good blood glucose control, nutrition and food education should include instruction and problem solving for eating at fast-food restaurants and schools; food portioning skills for home and at school, restaurants, and homes of friends; cooking from scratch and from prepackages; and reading labels from prepackaged food. Favorite foods and family preferences should be considered in educational sessions.
Nedra K. Christensen, PhD, RD, is a professor in the Utah State University Nutrition and Food Science Department in Salt Lake City. Elaine Boswell King, MSN, APRN, CDE, is an instructor in the practice of nursing in the Vanderbilt Diabetes Research and Training Center in Nashville, Tenn. Sherrie Hardy, RD, MS, CDE, is a clinical dietitian at Primary Children's Medical Center in Salt Lake City. Roxane Pfister, MS, is data manager of epidemiologic studies at Utah State University in Logan.
- American Diabetes Association