Weight Concerns and Weight Control Behaviors of Adolescents and Their Mothers

Objective  To assess the association between weight concerns and weight control practices of adolescents and their mothers.

Design and Participants  Cross-sectional study of 5331 adolescent girls and 3881 adolescent boys (age range, 11.8-18.4 years) in an ongoing cohort study and their mothers. Participants were included in the analysis if both the adolescent and his or her mother returned a questionnaire mailed in 1999 and provided information on weight, height, and weight concerns.

Results  More adolescent girls (33.0%) than boys (8.1%) thought frequently about wanting to be thinner. Compared with adolescent girls who accurately perceived that their thinness was not important to their mother, girls who misperceived (odds ratio [OR] = 1.9; 95% confidence interval [CI], 1.3-2.8) or accurately perceived (OR = 2.85; 95% CI, 1.0-8.4) that it was important to their mother that they be thin were significantly more likely to think frequently about wanting to be thinner. Among the adolescent boys, only those who accurately perceived that it was important to their mother that they not be fat were more likely than their peers to think frequently about wanting to be thinner (OR = 3.8; 95% CI, 2.3-6.2). Adolescents who accurately perceived that it was important to their mother to be thin or not fat were significantly more likely to be frequent dieters than their peers who accurately perceived that their weight was not important to their mother.

Conclusion  Among adolescents, an accurate perception that weight status is important to their mother is associated with thinking frequently about wanting to be thinner and about frequent dieting.

The association between body weight of children and their parents is known to be affected by genetic and cultural factors. However, less is known about the relationship between the weight concerns and weight control practices of parents and their children. Mothers are thought to play a role in the transmission of cultural values about body weight and shape.^1-4 Studies^1,2 show that girls whose mothers diet and are concerned with their weight and shape are more likely than their peers to develop unhealthy weight control practices. Moreover, Levine et al⁴ suggest that weight control behaviors among young girls are modeled partially on their mother’s behavior.

Children not only model their behavior on the behavior of their parents but also may adopt behaviors in response to the values that they perceive are important to their parents. Weight-related comments by parents have been found to have an effect on their children’s concerns with weight and weight control behaviors. Among 2379 girls who were 9 to 10 years of age, Schreiber et al⁵ reported that girls who had a high body mass index (BMI) or were told by their mother that they were too fat were more likely to be constant dieters. Moreover, Smolak et al³ observed that comments by mothers had a larger effect than those by fathers and that daughters were more affected than sons by the comments. In contrast, in the Growing Up Today Study (GUTS), it was previously observed that preadolescent girls and boys who reported that it was important to their father that they be thin (for the girls) or not fat (for the boys) were significantly more likely than their peers to start dieting, whereas the importance of a child’s body shape to his or her mother was not predictive of dieting among either sex.⁶ However, children who perceived that their weight or shape was important to at least 1 of their parents were at an increased risk for becoming concerned with weight. Because the information was based on the child’s report, it was unclear whether the observed relationships were due to a distorted perception on his or her part. To assess the relationship between adolescents’ weight concerns and the real and perceived weight concerns of their mothers, as well as between adolescents’ dieting behavior and the real and perceived weight concerns and dieting practices of their mothers, we analyzed data from adolescents in GUTS and their mothers. Specifically, we assessed whether the misperceived importance and the true importance of the adolescent’s weight status to the mother were related to the adolescent’s weight concerns.

The GUTS was established in 1996 by recruiting children of women participating in the Nurses’ Health Study II (NHS II). The NHS II was established in 1989 and consists of 116 671 female nurses aged 25 to 43 years at entry. Follow-up questionnaires have been sent to participants biennially since 1989. Additional details about the cohort have been reported previously.⁷ Using the NHS II data, we identified mothers who had children aged 9 to 14 years. We wrote a detailed letter to the mothers, explaining the purpose of GUTS, and sought parental consent to enroll their children. Adolescents whose mothers gave us consent to invite them to participate were mailed an invitation letter and a questionnaire. Other details of the study have been previously published.⁶ Approximately 68% of the adolescent girls (n = 9039) and 58% of the adolescent boys (n = 7843) returned completed questionnaires, thereby assenting to participate in the cohort. The participants have been sent questionnaires annually. In addition, a questionnaire was sent to their mothers in 1999. The study was approved by the Human Subjects Committee at Brigham and Women’s Hospital, Boston, Mass.

Participants included 5331 female and 3881 male GUTS participants (age range, 11.8-18.4 years) who returned the 1999 questionnaire (and whose mothers returned the 1999 mothers’ questionnaire), were aged 9 to 14 years at baseline, and completed questions on wanting to be thinner, dieting, and their perception of the importance of their physique to their mother. In addition, 7591 mothers of these GUTS participants (1624 mothers have >1 child in the study) who completed a brief questionnaire in 1999 as part of the GUTS (as well as the 1999 NHS II questionnaire) were included in the analysis.

In the fall of 1996 through 1999, GUTS participants received a questionnaire assessing different factors. Among many female subjects in the United States, thinness and low body fat are associated with one another, and both are considered to be positive attributes. Among male subjects, however, the 2 are not necessarily related to one another, nor are they both considered positive attributes. As with female subjects, low body fat is desirable, but thinness may be associated with weakness or femininity and is not considered to be a desirable characteristic by young men. Male subjects are more likely to be focused on wanting to increase muscle tone rather than being thin.^8-10 To render the McKnight Risk Factor Survey¹¹ questions appropriate for adolescent boys, the questions on thinness were replaced with questions inquiring about the importance of not being fat or desiring not to be fat. Two questions assessing the importance to each of the parents that the adolescent be thin (or not fat for adolescent boys) were adapted from the McKnight Risk Factor Survey. For the analysis, the McKnight Risk Factor Survey questions were dichotomized. Thinness or lack of fatness was considered important (to a parent or peer) if the adolescent indicated the degree of importance as “a lot” or “always.” Teenagers were considered to be making a lot of effort to look like women or men in the media if they responded “a lot” or “always” to the following question: “In the past year, how much effort have you made to look like the girls or women [boys or men] you see on television, in movies, or in magazines?”

Dieting to control weight was assessed with the following question adapted from the Youth Risk Behavior Surveillance¹² questionnaire: “During the past year, how often did you diet to lose weight or to keep from gaining weight?” To be consistent with previous analyses on the relationship of dieting to weight gain,¹³ adolescents who reported that they had dieted to lose or maintain weight “2 to 6 times per week” or “every day” were considered frequent dieters. Height and weight were self-reported annually. We calculated BMI as weight in kilograms divided by the square of height in meters, using self-reported weight and height information and calculated percentiles and z scores based on the Centers for Disease Control and Prevention growth charts,¹⁴ which are age- and sex-specific. Adolescents who had a BMI in the 85th percentile or higher for their age and sex were classified as at risk of overweight or as overweight. The validity of self-reported weight and height among preadolescents and adolescents has been investigated by several groups of researchers, and the results show that young people provide valid information on weight and height.^15-17

In August 1999, the mothers were sent a brief questionnaire that was personalized with the adolescent’s first name so that mothers with more than 1 child in the cohort would not be confused when completing the survey. The mothers were asked the following 2 questions that were included on the GUTS survey: “In the past year, how often have you thought about wanting to be thinner?” and “In the past year, how often have you tried to lose weight?” The mothers were asked how important it was to her that her son or daughter be thin and how important it was to her that her son or daughter not be fat. In addition, information on weight in 1999 and adult height (reported in 1989) were collected as part of the ongoing NHS II. Body mass index was calculated from self-reported weight and height.

There were 2 main outcome measures among the adolescents. The first was thinking frequently about wanting to be thinner, and the second was frequent dieting to lose or maintain weight.

Analyses were stratified by sex and conducted using SAS software (version 8.2; SAS Institute Inc, Cary, NC). To account for the correlation between siblings, generalized estimating equations, using the logit link function, were used for all multivariate analyses (SAS PROC GENMOD). An adolescent’s age and age- and sex-specific z scores of BMI were controlled for in all generalized estimating equation analyses that assessed the correlates of thinking frequently about wanting to be thinner and the correlates of frequent dieting. In the statistical models, the BMIs of the mothers in 1999 were classified as underweight or low healthy weight (BMI, 16 to <20), healthy weight (BMI, 20 to <25), or overweight or obese (BMI, ≥25). Women who were of healthy weight were the referent group. In one set of statistical models (model 1 in Table 1), to evaluate the predictiveness of the perceived importance to a mother that her daughter be thin or that her son not be fat, as well as the true importance as reported by the mother, we entered separate covariates to represent the information provided by the adolescent and the mother. In other models (model 2 in Table 1), we used the information from the adolescent and the mother to ascertain whether the adolescent’s perception of the importance of his or her weight to his or her mother was accurate. Using the information from the mother and the adolescent, we created the following 4 new variables: (1) neither the mother nor her adolescent reported that the adolescent’s weight was important to the mother, (2) the adolescent misperceived that his or her weight was important to the mother, (3) the adolescent misperceived that his or her weight was not important to the mother, and (4) the mother and the adolescent reported that the adolescent’s weight was important to the mother. In those analyses that assessed the associations with distorted and accurate perceptions about the importance of the adolescent’s weight status to the mother, adolescents who accurately perceived that their weight status was not important to their mother were the referent group, and the 3 new variables that cross-classified the mother and adolescent information were used instead of the separate questions answered by the mother and the adolescent. In addition, in secondary analyses, the models were stratified by the adolescent’s weight status (whether the adolescent was overweight or at risk of overweight vs underweight or healthy weight). All P values are 2-sided, with P<.05 considered statistically significant.

In 1999, the mean age of the participants was 14.9 years, and 14.6% of the adolescent girls and 22.4% of the adolescent boys were at risk of overweight or were overweight. Despite the higher prevalence of high weight status among the boys, weight concerns were more common among the girls, with 33.0% of girls and 8.1% of boys thinking frequently about wanting to be thinner. Frequent weight concerns were also common among the mothers. Approximately 54% of the mothers reported that they thought about wanting to be thinner a lot or always, although only 22% reported attempting frequently to lose weight during the past year. The teenagers underestimated the importance to their mother that they be thin (girls) or not be fat (boys). Only 0.4% of the adolescent girls and 3.7% of the adolescent boys accurately perceived that their weight was important to their mother (Table 1).

About 33% of adolescent girls thought frequently about wanting to be thinner. Independent of the girl’s age, BMI, mother’s weight status, and peer and media influences, the importance to the mother that her daughter be thin was less strongly associated with her daughter’s frequency of thinking about being thinner (odds ratio [OR] = 0.91; 95% confidence interval [CI], 0.61-1.35) than the perception by the daughter (regardless of accuracy) that her weight was important to her mother (OR = 1.84; 95% CI, 1.30-2.69). However, compared with girls who accurately perceived that their thinness was not important to their mothers, girls who misperceived (OR = 1.90) or accurately perceived (OR = 2.85) that their thinness was important to their mother were more likely to report thinking frequently about wanting to be thinner (Table 1).

In models stratified by whether the adolescent had been overweight, we did not observe strong evidence of effect modification by a girl’s weight status. Among the adolescent girls who had been overweight in 1997, 1998, or 1999, the perceived importance to a mother that her daughter be thin (OR = 2.79; 95% CI, 1.44-5.39) had a stronger association with thinking frequently about wanting to be thinner than the true importance to the mother (OR = 1.89; 95% CI, 0.79-4.50). The same was true for adolescent girls who had not been overweight in 1997, 1998, or 1999 and their mothers (OR = 1.79; 95% CI, 1.14-2.79; and OR = 0.76; 95% CI, 0.47-1.23; respectively).

Approximately 8% of the adolescent boys reported thinking frequently about wanting to be thinner. A boy’s z score of BMI was the factor most highly associated with wanting to be thinner. The perception that it was important to his mother that he not be fat (OR = 2.04) was a stronger correlate of thinking frequently about wanting to be thinner than the true importance to the mother (OR = 1.25) (Table 1). However, as with the adolescent girls, those who accurately perceived that their weight status was important to their mother were more likely to think frequently about wanting to be thinner (OR = 3.80) than those adolescents who knew that not being fat was not important to their mother. In addition, there was no effect modification by the adolescent boy’s weight status in terms of the associations with the perceived and true importance to the mother that the boy not be fat (data not shown).

Eight percent of the adolescent girls (n = 423) had frequently dieted during the past year. Although the prevalence of having a mother who reported that it was important to her that her daughter be thin was similar among the girls who had been overweight in 1997, 1998, or 1999 (3.0%) and the girls who had not been overweight (3.5%), there was a large difference in the perception that it was important to their mother that they be thin (2.9% and 7.2%, respectively). Overall, there was a suggestion that adolescent girls who misperceived that being thin was important to their mothers (OR = 1.72) or accurately perceived that it was important to be thin (OR = 2.10) were more likely to be frequent dieters than their peers who accurately perceived that their thinness was not important to their mother (Table 2). However, a girl’s weight status modified those associations. In the model stratified by the adolescent girl’s weight history, among the girls who had been overweight in 1997, 1998, or 1999, the true importance to the mother that the girl be thin (OR = 2.47; 95% CI, 1.06-5.78), but not the perception of the importance by the daughter (OR = 0.81; 95% CI, 0.32-2.06), was associated with frequent dieting. In contrast, among the adolescent girls who had not been overweight, the perception of the importance by the daughter (OR = 2.60; 95% CI, 1.27-5.32), but not the true importance to the mother that the girl be thin (OR = 1.12; 95% CI, 0.49-2.59), was associated with frequent dieting. Independent of the mother’s and daughter’s BMI, frequent weight loss attempts by the mother were associated with frequent dieting by the daughter (OR = 1.43; 95% CI, 1.06-1.94); however, it appeared that the association was exclusively among the adolescent girls who had not been overweight in 1997, 1998, or 1999 (OR = 1.45 vs 0.94).

Approximately 3% of the adolescent boys frequently dieted in the past year. Unlike the adolescent girls, adolescent boys who had a mother who frequently dieted (OR = 1.22; 95% CI, 0.82-1.81) were not more likely than their peers to have frequently dieted during the past year. However, there was a suggestion that adolescent boys who accurately perceived that it was important not to be fat (OR = 2.13) were more likely to be frequent dieters (Table 2). There were too few cases to assess whether the associations varied by the adolescent boy’s weight status (ie, ever vs never overweight in the past 3 years).

The present study is the largest investigation to date on the relationship between an adolescent’s weight concerns and dieting and the importance of the adolescent’s weight to his or her mother and her dieting practices. Moreover, unlike the previous studies^1,3 with parent and child reports, we collected similar information from adolescents and from their mothers and combined the information from both sources so that accurate and distorted perceptions could be separated from one another. Although few adolescents accurately perceived that their weight was important to their mother, adolescent boys and adolescent girls who accurately perceived that their weight was important to their mother were more likely to think frequently about wanting to be thinner and to frequently diet than their peers who accurately perceived that their weight was not important to their mother.

Parents are believed to play a role in the transmission of cultural values about desirable body weight and shape.^1-4,18,19 In a cross-sectional study, Smolak et al³ studied 131 mothers and 89 fathers of girls and boys in fourth and fifth grades. They observed that a mother’s comments about weight were significantly associated with her child’s concerns about gaining weight and with the frequency of trying to lose weight. Unfortunately, the sample was too small to conduct meaningful analyses within each sex. Although neither Smolak et al³ nor Byely et al²⁰ observed an association between a mother’s concerns with her weight or her dieting and her child’s weight loss attempts or concerns about gaining weight, it is possible that the samples were too small to detect a modest association. In a study of 381 adolescent boys and 429 adolescent girls, Fulkerson et al²¹ observed that girls whose mothers dieted were more like to be concerned with weight and to try to control their weight, but the association was confounded by BMI. In contrast, adolescent boys who were encouraged to diet by their mothers were more likely than their peers to be trying to control their weight, as well as to binge eat. In the GUTS cohort, it was previously observed that preadolescents and adolescents who perceived that their weight was important to their mother or father were more likely than their peers to become concerned with their weight during a 1-year period.⁶ However, it was unclear whether it was a distorted or an accurate perception that was the risk factor for becoming concerned with weight. Findings from the present study suggest that the accuracy of the perception is more important for adolescent boys than for girls, but accurate perceptions for both sexes were more strongly associated than distorted perceptions with a desire to be thinner and with frequent dieting. However, among adolescent girls, a distorted perception that weight status is important to the mother was associated with thinking frequently about wanting to be thinner and with frequent dieting to lose weight.

Adolescents not only model their behavior on the behavior of their parents and peers but also may adopt values and behaviors in response to the values that they perceive are important to their parents and peers. Abramovitz and Birch¹⁸ observed that girls whose mothers were dieters were more than twice as likely as girls whose mothers were nondieters to have formed concepts and beliefs about dieting. The influence of peers should not be underestimated. Among the adolescent girls in our study, the perceived importance of thinness to their peers was a stronger predictor than the perceived importance to their mother for the outcomes of thinking frequently about wanting to be thinner and of frequent dieting. Among the adolescent boys, the perceived importance to peers was related to thinking frequently about wanting to be thinner but was not related to frequent dieting. However, in both sexes, wanting to look like same-sex figures in the media had a stronger association with both outcomes than the perceived importance of weight to their peers or their mother.

It was a surprise that the preadolescent and adolescent participants, particularly boys, underestimated the importance of their weight to their mother. Only 4% to 5% of the participants thought that their weight status was important to their mother, whereas almost half of the women reported that it was important to them that their sons not be fat. A similar percentage of mothers reported that it was important that their daughters not be fat, but the adolescent girls were not asked about their perception of the importance to their parents that they not be fat; therefore, no comparisons could be made. The high percentage of mothers who believed strongly that they did not want their adolescent to be fat may reflect the fact that they are nurses and may be well aware of the adverse health consequences of obesity.²² It is unclear whether the discrepancy between the reports of the mothers and their sons reflects that the mothers focused on promoting a lifestyle that would have a positive effect on weight rather that discussing weight per se. More research is needed that assesses communication regarding weight, nutrition, and physical activity, as well as efforts to promote behavior change among mothers and their adolescents to better understand the most effective means for promoting a healthy lifestyle among young people.

There are several limitations of our study. The participants are children of nurses, and the sample is more than 90% white; therefore, we are unlikely to have adolescents of low socioeconomic status in the sample population, so the results of the sample may not be readily generalizable to economically disadvantaged populations or to adolescents of color. In addition, because of the cross-sectional design, we cannot make inferences about the temporal order of the associations. It is possible that adolescents who think frequently about their weight start to believe that being thin or not fat is important to their mother, and it is also possible that having a perception (regardless of accuracy) that weight is important to their mother may lead an adolescent to think frequently about wanting to be thinner. Prospective analyses will help to better understand the direction of the association. Because the cohort continues to be followed up, eventually we will be able to assess the longitudinal associations between maternal and peer influences on weight concerns and dieting in young adults.

The present study makes an important contribution to our understanding of the association of accurate and inaccurate perceptions of the importance of weight to parents, as well as the weight concerns and the dieting frequency of preadolescents and adolescents. Future research is needed to advance our understanding of how misperceptions on the importance of weight arise. High levels of concern with weight are associated with the initiation of using unhealthy weight control behaviors.^23,24 Therefore, the recommendations from the present study are that parents should be advised that their concerns with weight may promote unhealthy weight concerns in their adolescents. Being overweight is associated with many adverse health consequences, so parents are justified in not wanting their adolescents to be overweight. However, it is essential to strike a balance between promoting a healthy weight and not placing too much emphasis on the importance of weight. Parents should be encouraged to be role models in incorporating physical activity and healthy nutrition into everyday lifestyle patterns, rather than implementing them as weight control strategies. In addition, it would be advisable for clinicians who treat overweight and weight-concerned adolescents and young adults to promote physical activity for benefits other than weight control, such as helping to improve self-esteem.²⁵

Correspondence: Alison E. Field, ScD, Division of Adolescent/Young Adult Medicine, Department of Medicine, Children’s Hospital and Harvard Medical School, 300 Longwood Ave, Boston, MA 02115 (Alison.Field@childrens.harvard.edu).

Accepted for Publication: June 13, 2005.

Author Contributions: Ms Louie had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Funding/Support: This study was supported by grant DK46200 from the Boston Obesity Nutrition Research Center, Boston; by grants DK46834 and DK59570 from the National Institutes of Health, Bethesda, Md; by the Kellogg Company, Battle Creek, Mich; and by grant 6T71 MC00009-11-01 from the Maternal and Child Health Bureau, Rockville, Md.

Acknowledgment: We thank Marie E. Louie, PhD, for verifying the accuracy of the statistical analysis and results.