- Journal List
- HHS Author Manuscripts
- PMC2530935
Five-year-old girls’ ideas about dieting are predicted by their mothers’ dieting
Abstract
Objective
To explore 5-year-old girls’ ideas, concepts, and beliefs about dieting.
Design
Girls were asked to define dieting, to describe the behaviors dieting comprised, and were queried about links between dieting, weight control, and body shape. Parents completed questionnaires addressing family health history, demographics, and issues related to food, dieting, and weight control.
Subjects/setting
Participants were 197 girls aged 5 years and their parents. All girls lived with both biological parents, and were without food allergies or chronic medical problems.
Statistical analyses performed
For 5 open-ended questions related to dieting, girls were categorized as either having or not having ideas about dieting. These ideas, concepts, and beliefs were categorized, and logistic regression examined predictors of girls’ ideas about dieting.
Results
Depending on the question, from 34% to 65% of girls aged 5 years had ideas about dieting. Compared to girls whose mothers did not diet, girls whose mothers reported current or recent dieting were more than twice as likely to have ideas about dieting, suggesting that mothers’ dieting behavior is a source of young girls’ ideas, concepts, and beliefs about dieting. Among mothers, more than 90% reported recent dieting, and most reported use of both health-promoting and health-compromising dieting behaviors.
Applications
Women should be informed that weight control attempts may influence their young daughters’ emerging ideas, concepts, and beliefs about dieting. Mothers should be encouraged to use health-promoting rather than health-compromising weight control strategies, not only for their own well being, but to reduce the likelihood that daughters will incorporate health-compromising dieting behaviors into their concepts, ideas, and beliefs about dieting.
In popular parlance, “dieting” refers to a wide range of behaviors and strategies for weight control. Dieting is a diffuse construct that encompasses behaviors and strategies that may either promote health or increase risk for health problems (1,2). Survey data have revealed that dieting for weight control is now normative among adult women and adolescent girls, and constitutes a multibillion-dollar industry in the United States today (3). Recently, there is increasing evidence that dieting is beginning during childhood, even before the onset of puberty. Dieting, weight concerns, and body dissatisfaction have all been reported in children as young as age 7 to 9 years, and these reports are more common among girls than boys, with approximately 40% of elementary school-aged girls reporting that they have tried to diet to lose weight (4–7). However, little is known about younger children’s emerging awareness of dieting and weight concerns.
The family is the primary social context for early socialization, and mothers play an important role in the development of young children’s eating behavior (8,9). Maternal influence includes directly instructing children about food and eating (10), using child-feeding practices to control what and how much children eat, and serving as models for children’s eating (10,11). Mothers play a central role in transmitting cultural values regarding weight, shape, and appearance to daughters (12), and mothers’ own dieting and eating problems are linked to their daughters’ dieting and eating problems (13). Hill and colleagues (14) also found that mothers’ dietary restraint was highly correlated with their 10-year old daughters’ dietary restraint. As early as the preschool period, mothers’ own dieting and eating behaviors may begin to influence children’s eating and overweight, and these effects may be direct, or mediated via their effects on mothers’ child-feeding practices (9,15,16). Given that dieting for weight control is pervasive among women, their young children have many opportunities to observe and learn from maternal weight-loss attempts, which may include both health-promoting and health-compromising weight loss strategies.
Story et al (2) recently reported that for adolescents, both health-promoting and health-compromising behaviors were included in the set of behaviors dieting comprised, and that adolescents who engaged in more health-compromising dieting strategies had less adequate nutrient intakes. To determine whether or not increased reports of dieting by children may represent a health risk for children, information on the behaviors that children believe constitute dieting is needed. Current findings indicate that when older children say they are dieting, they are referring to both health-promoting and health-compromising behaviors. For example, among those aged 9 to 11 years, the most frequently reported dieting behaviors reported were avoiding fattening foods and skipping meals (5,17). Although the former strategy is consistent with current dietary guidelines and should promote good health, chronic meal skipping may be problematic for children, and could place them at risk for problems of energy balance and eating disorders. This study investigates very young girls’ emerging ideas, concepts, and beliefs about dieting at a point in development before the actual initiation of dieting attempts, and examines predictors and sources of influence of girls’ ideas about dieting. In particular, the possibility that exposure to maternal dieting may facilitate their young daughters’ developing awareness and ideas about dieting is explored.
METHODS
Subjects
Participants were 197 girls aged 5 years (mean=5.36±0.30 years, range=4.69 to 6.44) and their parents. Girls lived with both biological parents, and did not have severe food allergies or chronic medical problems affecting food intake. Families were recruited for participation in a study of girls’ health and growth during middle childhood, and were paid $100. The mean age of mothers and fathers was 35.36±4.77 years (range 24.10 to 46.59) and 37.42±5.39 years (range 26.13 to 66.37), respectively. Sixty-three percent of mothers and nearly all fathers (97%) were currently employed, reporting an average of 20 and 45 hours of work per week, respectively. Twenty-nine percent of reported family incomes were below $35,000, 35% were between $35,000 and $50,000, and 36% were above $50,000. Parents were well educated, with 33% of mothers and 37% of fathers reporting high school diplomas, 38% of mothers and 48% of fathers reported an associate’s, technical, or bachelor’s degree, and 15% of mothers and 20% of fathers reported postgraduate degrees. This project was approved for human subjects by The Pennsylvania State University Office of Regulatory Compliance.
Child Measures
Dieting ideas questionnaire
This instrument assessed 5- to 9-year-old girls’ knowledge and ideas related to weight control, links between eating and weight status, and dieting. It included 5 open-ended interview questions about dieting, administered to girls in the following order: What can make people thin?, What can make people weigh too much?, What is a diet?, Why do people diet?, and What do people do when they are on a diet? This order of presentation was used to minimize the effects of the questions about dieting on girls’ responses to first 2 questions, which addressed the links between weight, body shape, and eating.
Dieting messages and behavior questionnaire
This measure provides information regarding sources of influence on girls’ ideas about dieting, as well as their use of weight-control behaviors. A list of possible influences (eg, mom, dad, media) was given to the girl with each question.
Weight concerns scale
This measure included questions that assessed worry about weight and body shape, importance of weight, perceived fatness, and current dieting (18). For young adolescents, Killen, et al (18) reported high reliability (r=0.71) across a 7-month interval, and validity based on relationships with the revised restraint scale and the drive for thinness and body dissatisfaction subscales of the Eating Disorder Inventory (EDI). The weight concerns scale had not been previously used with children as young as age 5 years, and because several of the questions were not age appropriate for this sample, 3 questions out of the original 6 questions were used: “Do you worry about your weight (how fat or skinny you are) and body shape?”, “How afraid are you of getting fat?”, and “Do you think a lot about being fat or skinny?” Three-point scales were used for these 3 questions. Questions on current dieting status were eliminated because there was little variability in our sample, with only 13 of 197 girls indicating that they were currently dieting. A question on the relative importance of weight compared to other things in life was also eliminated because many of the 5-year-old girls had trouble understanding this question. The final Cronbach’s alpha for the 3-item scale was an acceptable 0.65.
Anthropometrics
Weight and height were measured in kilograms and centimeters, respectively, by a trained research assistant. Following the procedures of Lohman, et al (19), weight was measured to the nearest 0.1 kg using an electronic scale, and height was measured to the nearest 0.5 cm using a stadiometer. The mean of 3 measurements was used to calculate body mass index (BMI) (weight in kg/height in m2). In addition, daughters’ weight and height measurements were converted to weight-for-height z scores using National Center for Health Statistics (20) data (Epi Info, Version 6.04, 1997, Epi Nut Module, World Health Organization, Geneva, Switzerland).
Parent Measures
Personal health history
Questions addressed parent’s personal health history, including exercise, alcohol intake, smoking, family history of overweight and underweight, cardiovascular disease risk, weight, and dieting status.
Dieting behaviors
This included a series of questions regarding current dieting status, frequency and success of past dieting attempts, weight loss, and weight gain. Mothers also completed the Weight Control Behavior Scale, which used a checklist of weight-loss behaviors to examine specific behavioral strategies used for weight control (1). In the Weight Control Behavior Scale French and colleagues (1) factor analyzed the data to derive 2 factors, grouping 11 weight-loss behaviors as healthful (eg, eating more fruits and vegetables, reducing fat intake), and 8 as unhealthful (eg, purging, smoking, fasting) and this categorization was used in our study.
Weight concerns scale
This measure included 5 questions that, using 5-point Likert scale, assess fear of weight gain, worry about weight and body shape, importance of weight, diet history, and perceived fatness (18). The sixth item assessed current dieting status. Items were composited to create a total weight concerns score. Principal components analysis was performed to obtain a standardized composite score for the complete weight concerns scale. Chronbach’s alpha was 0.82.
Eating inventory
The Eating Inventory, developed by Stunkard and Messick (21), measures dietary restraint, dietary disinhibition, and susceptibility to hunger. The dietary restraint subscale measures the cognitive control of food intake and assesses the restriction of certain foods, energy intake, or macronutrients. The disinhibition subscale measures overeating and the extent to which eating is triggered by emotional, social, and environmental cues. Only restraint and disinhibition were used in our study; Chronbach’s alpha for dietary restraint and disinhibition was 0.85 and 0.80, respectively.
Anthropometrics
Parents’ weight was measured to the nearest 0.1 kg using an electronic scale. Height was measured to the nearest 0.5 cm using a stadiometer. The mean of 3 measurements was used to calculate BMI.
Procedures
Each girl participated in 2 half-day interview sessions, about 1 week apart. Girls arrived at the laboratory at 8 AM, after an overnight fast. Four to 8 girls participated during each day. Breakfast, morning snack, and lunch were provided. An interviewer was assigned to be each girl’s “camp buddy” for both days’ sessions. Frequent breaks and play activities were included; the camp day consisted of four 30-minute questionnaire blocks, separated by free play periods. Parents completed questionnaires during an evening session in the laboratory.
The order of presentation of the questionnaires was designed so that the Dieting Ideas Questionnaire was presented to the girls as the first item on the interview, before any other discussions of dieting and health. Answers to Dieting Ideas Questionnaire questions were recorded verbatim. Interviewers made sure that girls were aware that it was okay to give an “I don’t know” response to any of the questions. The Dieting Messages and Behavior Questionnaire, which assessed sources of information about dieting, was presented on the second camp day, to reduce the influence from other interview questions.
Statistical Analysis
Content analysis and coding for Dieting Ideas Questionnaire
For each of the dieting questions, girls were placed into 1 of 2 mutually exclusive groups, based on their response. Girls who were able to articulate ideas were placed in the “ideas” group, and those who responded “I don’t know,” or who gave no response or a nonsensical response, were placed in the “no ideas” group. To be included in the ideas group, girls had to give responses that were related to conventional ideas about dieting, weight concern, energy balance, eating behaviors, and exercise. We did not require that their responses be “correct” based on current nutrition information, only that the response was plausible.
Inter-rater reliability was assessed by having 2 raters independently code girls’ responses into the ideas and no ideas categories, and for all of the girls’ responses to the 5 questions, there was only 1 disagreement between the raters, making the percent agreement greater than 99%. This dichotomous categorization of girls into the ideas and no ideas groups was the outcome measure predicted in the logistic regression analyses.
Further coding was done for girls’ responses falling into the ideas group. For each question, a set of mutually exclusive and exhaustive categories was developed, and by merging conceptually similar categories, the number of response categories for each question was then reduced. Once categories had been established, inter-rater reliability of the coding system was assessed again by having 2 raters independently code the girls’ responses. Across the 5 questions, inter-rater reliability ranged from 89% to 93%.
All data were analyzed using SAS (version 6.12, 1996, SAS, Cary, NC). Given the dichotomous outcome variables, logistic regression was used to predict whether or not 5-year-old girls had ideas about dieting. Analyses were conducted separately for each question from the Dieting Ideas Questionnaire.
RESULTS
Weight status, Dieting, and Weight Concern among Family Members
In this sample, 69% of mothers and 42% of fathers reported at least 1 weight-loss attempt during the past year; 24% of mothers and 8% of fathers reported that they were “currently dieting” at the time of the data collection. In response to the Weight Loss Behavior Scale, 93% of mothers reported using at least 1 weight-control strategy; 31% of women reported using only healthful strategies, whereas 62% reported using both healthful and unhealthful strategies. Mothers’ mean restraint score was 8.0, and fathers’ was 5.6; restraint scores ranged from 0 to 21, with higher scores indicating higher levels of restraint. Mothers’ mean disinhibition score was 6.9, and fathers’ was 4.8; disinhibition scores ranged from 0 to 16, with higher scores indicating greater disinhibition. Mothers’ and fathers’ mean BMI scores revealed that the sample was slightly overweight on average, with mean BMI of 26 and 28, respectively. Fifty-eight percent of mothers and 83% of fathers were overweight, as indicated by a BMI greater than 25. A history of overweight in the family was reported by 37% of the sample. Daughters had a mean weight for height percentile of 63.6, with 25% of the girls categorized as overweight, based on percentile values exceeding the 85th percentile.
Daughters’ Responses to the Dieting Ideas Questionnaire
As shown in Tables 1 through through5,5, depending on the question asked, between one-third and about two-thirds of the 5-year-old girls in our sample were able to articulate ideas about dieting and related constructs. As shown in Table 1, most girls’ responses to “What is a diet” were framed in terms of specific behaviors, although a few girls did speak in more abstract terms in defining a diet; for example, “…something that makes people thin…” Responses to “What do people do when they are on a diet?” appear in Table 2, and these were similar to those elicited by “What is a diet?”. The dietary modifications and restrictions that children mentioned included both health-promoting and health-compromising behaviors.
Table 1
Girls’ responses to, “What is a diet?” (N=197)
| Response category | %a | Sample response(s) |
|---|---|---|
| Don’t know | 55.3 | “I don’t know.” |
| Modified eating behaviors | 20.8 | |
| Diet beverage | 6.6 | Drinking a diet soda; drinking Diet Cokeb |
| Liquid diet | 5.1 | A chocolate milkshake; You have to drink this milkshake |
| Eating well or healthfully | 4.1 | Eating healthful foods; Something that people have to eat. Eat good food |
| Eat specific foods | 2.5 | Something you eat like tomatoes; Just eat vegetarian foods |
| Eat what told | 2.5 | It’s something when the doctor tells you what to eat; When you eat things people want you to eat |
| Restrictive eating behaviors | 17.1 | |
| Not eating | 8.1 | Not eat anymore; On a diet, you can’t eat |
| Restrict kilocalories | 2.5 | People eat less stuff; When you don’t eat too much food |
| Restrict sugar | 2.5 | When you can’t eat any sugar; Can’t eat sweets |
| Decrease junk food | 2.0 | Can’t eat junk food; No more chips |
| Restrict fat | 2.0 | Eating less fat; When somebody eats no fat |
| Lose weight or get skinny | 7.6 | Something that makes people thin; Want to lose some weight |
| Exercise | 1.5 | Getting some exercise; Exercising |
Table 2
Girls’ responses to, “What do people do when they are on a diet?” (N=197)
| Response category | %a | Sample response(s) |
|---|---|---|
| Don’t know | 45.7 | I don’t know |
| Modified eating behaviors | 19.3 | |
| Liquid diet | 5.1 | They drink something; Drink(diet) milkshakes |
| Eating well or healthfully | 4.6 | Eat more good things; Eat healthfully and good foods |
| Increase fruit and vegetable intake | 3.6 | Eat fruit; Eat foods like tomatoes, celery, cauliflower, and broccoli |
| Diet beverage | 3.0 | Drink diet coke; Drink diet sodas |
| Eat specific foods | 3.0 | Eat certain food; Eat foods in special packages |
| Restrictive eating behaviors | 18.7 | |
| Not eating | 8.6 | They don’t eat anything; Cook for their kids but they don’t eat the food they make |
| Restrict kilocalories | 4.6 | They don’t eat that much; They stop eating. Eat a little bit |
| Restrict fat | 3.0 | They eat less fat; No fat |
| Restrict sweets | 2.5 | Not too much chocolate; They don’t eat, especially sweets |
| Exercise | 7.1 | Runs on a treadmill; They get some exercise |
| Other behavior change | 5.5 | |
| Avoidance | 2.5 | They don’t get too close to the refrigerator; They run away from the diet |
| Other | 10.5 | Can’t eat any more snacks; Smoke instead |
Table 5
Girls’ responses to, “What can make people weigh too much?” (N=197)
| Response category | %a | Sample response(s) |
|---|---|---|
| Don’t know | 35.0 | I don’t know |
| Modified eating behaviors | 53.9 | |
| Eating, eating too much | 35.5 | They eat too much; They eat lots of food |
| Too many sweets | 5.6 | If they eat a lot of sweets; Too much candy |
| Junk food | 4.6 | Eating lots of junk food |
| Eating fat | 4.1 | Eating “fatness” stuff; Lot of fattening foods |
| Unhealthful or wrong foods | 4.1 | To eat “not healthy” foods that weigh too much |
| Weight-related behaviors | 11.7 | |
| Getting or being fat | 5.1 | Getting fat; Being too fat |
| Growing too much | 6.6 | They are growing too fast; Because they grow too much |
| Exercise | 4.0 | If you exercise; Because they exercise; Not exercising |
| Other | 1.0 | Having babies |
Responses to “What can make people thin?” and “Why do people diet?” appear in Tables 3 and and4,4, and these questions were designed to elicit information about girls’ understanding of links between dieting and attaining a thin(ner) body shape. “What can make people weigh too much?” elicited beliefs, concepts, and ideas from the largest percentage of girls (65%). Table 5 shows the majority of the girls indicated that eating too much could make people weigh too much, and suggests that many of these girls believed that being overweight was a result of eating too much, or eating the wrong foods. Only a few girls mentioned lack of exercise as a reason why people weigh too much.
Table 3
Girls’ responses to, “What can make people thin?” (N=197)
| Response category | %a | Sample response(s) |
|---|---|---|
| Don’t know | 65.0 | I don’t know |
| Modified eating behaviors | 12.1 | |
| Eating diet food | 3.0 | Eating special foods |
| Eating well or healthfully | 3.6 | Eating good foods (carrots, broccoli); Eating healthful stuff |
| Eating low-fat or nonfat | 2.5 | Eating not-fattening stuff; Eating low-fat foods |
| Eat specific foods | 3.0 | Skim milk; Spaghetti; Apples |
| Restrictive eating behaviors | 14.2 | |
| Not eating (enough) | 12.7 | Don’t eat any food; Not eating a lot |
| Eat bad-tasting food | 1.5 | Eat yucky foods |
| Exercise | 6.1 | Exercise; Walking |
| Losing weight | 3.6 | Losing weight; They get bone skinny |
| Other | 3.0 | When you don’t eat too much candy; Being tall |
Table 4
Girls’ responses to, “Why do people diet?” (N=197)
| Response category | % | Sample response(s) |
|---|---|---|
| Don’t know | 65.5 | I don’t know |
| Weight-related reasons | 15.7 | |
| Want to lose weight or get skinny | 10.2 | Because they want to lose weight; Because they want to get thin |
| Feel too fat or weigh too much | 3.0 | Because they don’t want to weigh too much; Because they are too fat |
| Don’t want to get fat | 2.5 | Because they don’t want to get a lot of fat; So they don’t get fat |
| Modified eating behaviors | 11.6 | |
| Eat too much | 4.1 | Because she ate too much before; This will help her lose weight |
| Healthful/good for them | 3.5 | Because it’s good for them; Because they want to stay healthy; So they’re healthy |
| Have no food | 1.5 | Because they don’t have any food; They’re starving |
| Like diet beverages | 2.5 | They do diets because they like Diet Pepsi; b Because they’re thirsty |
| Other | 8.0 | Because they don’t like anything; Because they like going on some dates; Because they want to; Because they have to |
Predicting Girls’ Ideas about Dieting
Logistic regression analysis was used to predict whether or not girls had ideas about dieting. Predictors included girls’ and parents’ weight status, weight concerns, parents’ recent dieting, dieting history, dietary restraint and disinhibition, family history of overweight, and mothers’ weight-control behaviors, as well as parental education, work status, and income. As a proxy for media influences we used hours of television watched as a predictor of girls’ ideas about dieting. The dichotomous outcome variable predicted was whether or not the question elicited ideas from the girls.
The results of the logistic regression analyses predicting girls’ ideas about dieting are presented in Table 6, which contains both odds ratios and 95% confidence intervals. For the first 4 questions, the only significant predictors of girls’ ideas were indicators of mothers’ recent diet history, including current dieting status and frequency of weight-loss attempts during the past year. Compared to girls whose mothers did not report recent dieting, girls whose mothers reported recent dieting were more than twice as likely to have ideas, concepts, and beliefs about dieting. Girls with a family history of over- weight were more than twice as likely to have ideas about diets than girls without a family history of overweight. Girls’ weight concern did not predict girls’ ideas about dieting, but was significantly related to mothers’ weight concern, r=0.16, P<.05.
Table 6
Summary of logistic regression results predicting daughters’ ideas about dietinga (N=197)
| Predictor | What is a diet?
| What do people do when they are on a diet?
| What can make people thin?
| Why do people diet?
| ||||
|---|---|---|---|---|---|---|---|---|
| Odds ratio | 95% CI | Odds ratio | 95% CI | Odds ratio | 95% CI | Odds ratio | 95% CI | |
| Mother currently dieting | 2.2* | 1.7–2.7 | 2.7** | 2.3–3.1 | 2.6** | 2.0–3.2 | ||
| Mother’s number of diets in the past year | 2.1* | 1.5–2.6 | ||||||
| Family history of obesity | 2.3** | 1.7–3.0 | ||||||
DISCUSSION
Although a significant proportion of 5-year-old girls in our sample were still naive about dieting and weight loss, a substantial proportion of them were able to express a wide range of ideas, concepts, and beliefs about dieting, and they articulated a range of weight-loss behaviors. Many of the girls also gave responses that revealed their understanding of the links between dieting and attaining a thin body shape. The girls’ answers in response to the question, “What do people do when they are on a diet?” were similar to the dieting behaviors reported by older children (5,17,22), and to the variety of weight-control strategies reported by adolescents and adult women (1,2,23–25). Girls’ responses included modifications of eating behaviors (drink skim milk; eat more fruits and vegetables), and restrictive eating practices (eat less “junk” food; don’t eat). Using the distinctions recently introduced by Story and colleagues (2) and by French et al (1), the behaviors mentioned by 5-year-old girls could also be categorized as health-promoting, “eat more fruits and vegetables,” or as health-compromising, “don’t eat.”
Among all the family factors we investigated, only a family history of overweight and mothers’ current or recent dieting predicted daughters’ emerging ideas, concepts, and beliefs about dieting. None of the information on fathers’ dieting history or weight status predicted girls’ ideas about dieting. Compared to girls in families where mothers did not report current or recent dieting, girls whose mothers reported current or recent dieting were more than twice as likely to articulate ideas, concepts, and beliefs about dieting. These findings indicate that in addition to potential effects on maternal health and weight status, maternal dieting also influences their young daughters’ emerging ideas about dieting.
Ideas about dieting are necessary but not sufficient for dieting to occur, but whether girls who have ideas about dieting at an early age are more likely to begin dieting early remains to be seen. These findings do reveal that the transmission of information about dieting from mothers to daughters begins early, before girls actually begin to adopt dieting behaviors. This intergenerational transmission of dieting practices and attitudes may be 1 factor contributing to familial patterns of adiposity, problems of energy balance, chronic dieting, and disordered eating (26). Transmission from mothers to daughters may occur via a variety of possible processes, including explicit instruction and observational learning. There is evidence that mothers of adolescent daughters may actually teach their daughters about how to diet (13), but explicit teaching seems unlikely among mothers of 5-year-old girls. It is also possible that 5-year-old girls’ direct observations of their mothers’ dieting behaviors may form the basis for girls’ emerging ideas, concepts, and beliefs about dieting. This possibility is suggested by 1 girl’s response to, “What do people do when they are on a diet?”, when she said, [Moms] cook for their kids but they don’t eat the food they make.”
Daughters of dieting mothers may also be exposed to maternal weight concern and body dissatisfaction, which covary with dieting and may influence daughters’ weight concern. Among adolescents, weight concern is a powerful predictor of dieting (18), and among the mothers in our sample, weight concern was the strongest predictor of dieting, especially of the use of unhealthful weight-control strategies (25). For these 5-year-old girls, neither maternal weight concern nor daughters’ weight concern predicted girls’ ideas about dieting, although mothers’ weight concern predicted maternal dieting, r=0.44, P<.001. Mothers’ weight concern was a significant predictor of daughters’ weight concern, suggesting the possibility that the transmission of weight concerns from mothers to daughters may play a critical role in determining daughters’ risk for the onset of dieting. However, at age 5 years, daughters’ weight concerns did not predict either their ideas about dieting or their reports of dieting, although many of the 5-year-old girls in the sample were already aware that dieting was associated with thinness. Among older girls and adolescents, dieting behavior is related to a desire to be thin (27–29), and to higher levels of weight concern. This desire for thinness is already prevalent in 7- to 9-year-old girls (30,31). As girls move into middle childhood, increased weight concerns and desire for thinness, in combination with their ideas that dieting is the means to a thin body shape, may trigger the initiation of dieting attempts.
Peer and media influences are probably implicated in the emergence of dieting and eating problems, but we obtained only very limited information on these factors because our focus was on family factors. Hours of TV watched did not predict girls’ ideas about dieting, and we did not obtain detailed information about peer influences on dieting in part because girls were interviewed before their entry into kindergarten, a time at which peer influences are still minimal relative to later childhood and adolescence. Additional research is needed to determine whether increased peer and media influence may fuel dieting ideas, or provide the impetus for increased weight concern and the initiation of dieting, and whether mothers’ own dieting and weight concerns continue to play a central role in the development of girls’ eating behavior.
By middle childhood, reports of dieting are common among school-aged girls (5), with about 30% of third graders, and 60% of sixth graders reporting that they have dieted (32). In another study, 22% of first-grade children reported to have dieted (33). Among the 5-year-old girls in our sample, only 13 (7%) reported that they had already dieted. Particularly with young children, such reports need to be interpreted with caution, because of children’s bias to give positive responses to questions (see Huon, Godden, and Brown [35] for a discussion of this issue). For the girls in this sample who reported dieting at age 5 years, responses to the follow-up probe questions revealed little evidence to support their reported use of weight-control practices. The lack of evidence for dieting among these 5-year olds, in combination with other evidence on the prevalence of dieting among older children, suggests that the onset of dieting behavior has not occurred at age 5 years, and may not typically occur until girls reach middle childhood.
The National Institutes of Health Consensus Statement (35) warns that weight-loss attempts may have an adverse effect on the physical and psychological health of young women. Our findings suggest that maternal dieting also has an effect on their young daughters’ developing ideas about dieting. Although the 5-year-old girls in our sample were apparently still too young to actually begin to engage in dieting behaviors, many of them had already acquired a wide range of ideas, concepts, and beliefs about dieting, and that their dieting mothers were a primary source of information about dieting. A substantial proportion of these very young girls could already articulate that dieting involved restrictive or modified eating behaviors, which were intended to produce weight loss. The relationship between mothers’ and daughters’ weight concerns suggests that in addition to acquiring ideas and information about dieting from their mothers, the motivation to diet may also be transmitted from mother to daughter, and that this transmission begins when girls are very young. Many dieting mothers may unintentionally serve as models and sources of information on dieting for their very young daughters, who first learn about dieting by observing their mothers’ use of weight-control practices. Five-year-old daughters of dieting mothers are more likely to have ideas about dieting at an early age, but it remains to be seen whether mothers’ dieting also places their daughters at risk for the early onset of dieting behaviors and heightened weight concern.
APPLICATIONS
Prevalence data reveal that a substantial proportion of women of child bearing age are dieting. In our sample, 58% of mothers had BMIs greater than 25, an indication of overweight, and for those women, weight loss may be warranted. Diet counseling directed at mothers should introduce the idea that maternal dieting may influence their daughters’ emerging ideas and beliefs about dieting, and this argues strongly for the use of positive, health-promoting weight-loss strategies, such as increasing fruit and vegetable consumption and increasing physical activity, which may have positive effects on both mothers’ and daughters’ health and weight status. Evidence that daughters are beginning to learn about dieting at a very early age provides additional impetus for discouraging maternal use of health-compromising weight-loss strategies. Practitioners could use this information to motivate mothers to eschew their use of health-compromising weight-loss strategies in favor of more healthful approaches to weight control. This approach could have the effect of decreasing the likelihood that health-compromising weight-loss strategies would become an integral part of young girls’ developing ideas and beliefs about dieting and weight control.
Acknowledgments
This study was supported by National Institutes of Health grant No. RO1 HD32973.
Facebook
Twitter
Google+