- Email: [email protected]
Self-Weighing and Weight Control Behaviors Among Adolescents with a History of Overweight
Journal of Adolescent Health 44 (2009) 424–430
Original article
Self-Weighing and Weight Control Behaviors Among Adolescents with a History of Overweight Mary E. Alm, Ph.D.a, Dianne Neumark-Sztainer, Ph.D., M.P.H., R.D.a,b, Mary Story, Ph.D., R.D.a,b, and Kerri N. Boutelle, Ph.D.a,b,c,d,* b
a Department of Pediatrics and Adolescent Health, University of Minnesota, Minneapolis, Minnesota Department of Epidemiology and Community Health, University of Minnesota, Minneapolis, Minnesota c Department of Pediatrics and Psychiatry, University of California–San Diego, LaJolla, California d Rady Children’s Hospital, San Diego, California Manuscript received April 11, 2008; manuscript accepted August 27, 2008
Abstract
Purpose: To assess the relationships between self-weighing frequency, weight control behaviors, and weight status among male and female adolescents who have a history of being overweight. Methods: This study compared weight control behaviors between two groups of adolescents with a history of being overweight (body mass index [BMI] >85th percentile): those who weighed themselves weekly or more (frequent self-weighers) and those who weighed themselves monthly or less (infrequent self-weighers). Participants completed a survey on weight control behaviors, dietary intake, physical activity, and sedentary activity. Height and weight were also measured. Logistic regression analyses were used for categorical outcomes and linear regressions for continuous outcomes. Results: Of the 130 adolescents, 43% were frequent weighers and 57% were infrequent weighers. In comparison to infrequent self-weighers, frequent self-weighers were more likely to report using behavior change strategies, following a structured diet, and engaging in healthy weight control behaviors, especially decreasing caloric intake, high fat food intake, and ‘‘junk food’’ intake. Also, more frequent self-weighers reported engaging in more strenuous physical activity and spending less time playing videogames than infrequent self-weighers. Although not significant, a trend resulted indicating lower average BMI percentile among frequent self-weighers. No significant differences were found between the two groups in unhealthy weight control behaviors. Conclusions: These results suggest that adolescents with a history of overweight who self-weigh at least weekly are more likely to report using healthy weight control behaviors than adolescents who self-weigh monthly or less frequently. Self-monitoring of weight may be a useful component of a comprehensive weight management plan for some overweight adolescents. Ó 2009 Society for Adolescent Medicine. All rights reserved.
Keywords:
Weight monitoring; Self-monitoring; Pediatrics; Weight management
According to national data, approximately 16% percent of adolescents have a body mass index (BMI) at or above the 95th percentile and 34% of adolescents have a BMI at or above the 85th percentile [1]. Given the high prevalence of overweight and obesity among adolescents and the high risk of obesity continuing from adolescence to adulthood, finding effective weight control strategies for youth is impor*Address correspondence to: Kerri N. Boutelle, Ph.D., University of California-San Diego, 9500 Gilman Drive MC 0985, La Jolla, CA 92093. E-mail address: [email protected]
tant [2]. Early intervention may lessen the grave health and economic consequences associated with adult obesity [3,4]. Research suggests that adults who are successful in weight management self-monitor their food intake, physical activity, and weight [5–7]. Furthermore, studies with adults suggest that regular weight monitoring may be an important component of effective weight loss and long-term weight control [7–9]. For example, one study found that a majority of adults (75%) who were successful at long-term weight loss maintenance reported self-weighing at least weekly [7]. Results of another investigation that evaluated two longitudinal datasets
1054-139X/09/$ – see front matter Ó 2009 Society for Adolescent Medicine. All rights reserved. doi:10.1016/j.jadohealth.2008.08.016
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430
of adults who were enrolled in either weight loss or weight gain prevention trials found that daily self-weighing was associated with greater weight loss and weight loss maintenance [8]. This finding was consistent with a study that found daily self-weighing was associated with improved weight loss maintenance among adults participating in a self-regulation program delivered either over the Internet or in person [9]. Few studies have explored self-weighing and weight control among adolescents. The effects of self-weighing may differ for adolescents, as being heavier than one’s peers or societal standards can negatively impact body image among teenagers [10]. There is concern that repeated weight monitoring may lead to weight preoccupation and unhealthy weight control practices (e.g., fasting, skipping meals, selfvomiting, and laxative use). Furthermore, considering the high prevalence of body dissatisfaction and weight preoccupation among adolescents, especially females [10], an intensified focus on weight may increase the risk for harmful weight control behaviors and disordered eating. Findings from the few studies investigating self-weighing and weight control among adolescents have been mixed. The benefits of self-weighing were noted by one longitudinal study that found daily self-weighing by overweight children ages 6 to 12 participating in a weight loss program were significantly associated with larger decreases in overweight status 10 years posttreatment [11]. This finding was consistent with a study of female undergraduate college students that found daily self-weighing was effective in preventing weight gain [12]. However, these studies did not evaluate the relationship of self-weighing and unhealthy weight control behaviors. Findings from a 5-year longitudinal study assessing self-weighing and weight control practices in a general population of adolescent females indicated that frequent self-weighing was predictive of unhealthy weight control behaviors and was not predictive of better weight management [13]. Regular weight monitoring may increase an adolescent’s awareness of weight fluctuations or gradual weight gain, thus allowing the adolescent to adjust diet and physical activity accordingly and engage in self-regulation [14]. Therefore, frequent self-weighing may be beneficial to long-term weight control. On the other hand, frequent weight monitoring may have unintended consequences and cause negative affect and body dissatisfaction, which may contribute to the development of disordered eating, especially among adolescent females. Although recommending regular self-monitoring of weight may be of concern for a general population of adolescent girls, little is known about the effects of self-weighing on weight control practices among overweight adolescents. Results from investigating these effects would indicate whether self-weighing should be a recommended weight control tool for overweight adolescents, as it is for overweight adults. Therefore, the purpose of this study was to further the knowledge base about the relationship between self-weighing and weight control behaviors among adolescents who have a history of being overweight. The present study aims
425
to examine the relationship between frequent self-weighing and weight control behaviors among adolescents with a history of overweight and to address the research question: Is the frequency of self-weighing related to weight status and weight control behaviors in adolescents with a history of being overweight? Methods Study design Data for this investigation was drawn from the Successful Adolescent Weight Losers (SAL) study, a cross-sectional study of overweight adolescents comparing those who reported weight loss and those who do not. The institutional review board of the University of Minnesota approved the study protocol, and informed consents were obtained from all participants. Dietary intake, physical activity levels, and weight control behaviors were assessed through questionnaires, and height and weight were measured in a clinic setting. Although parent data was also collected for the SAL study, only items from the adolescent’s data was used for the current investigation. Participants One hundred thirty adolescents were recruited through a variety of techniques, including newspaper advertisements, flyers, mailings, and physician referrals. Participation eligibility included being 12 to 20 years old at the time of the assessment and meeting the criteria for overweight (BMI > 85th percentile) within the past 2 years. Exclusion from participating included severe psychiatric disorders and medical conditions with potential to interfere with results, such as depression, eating disorders, diabetes, and endocrine disorders. Of the entire sample, almost half (n ¼ 62) of the adolescents reported weight loss of at least 10 pounds in the past 2 years with no subsequent weight regain for 3 months, and 68 adolescents reported no weight loss (Table 1). The sample included 84 females and 46 males ages 12 to 20, and the mean age of participants at the time of the assessment was 15.2 years (SD ¼ 2.14). Self-reported ethnic identification indicated that over half of the participants were Caucasian (59.2%); 13.1% were African American, 6.9% were Native American, 2.3% were Asian, 0.8% were Hispanic, and 14.6% were ethnically mixed. Measures A questionnaire was administered to participating children at the time of the assessment that addressed the following measures. Self-weighing. Self-weighing frequency was assessed by the following question: ‘‘How often do you weigh yourself?’’ Participants selected one of five responses, which included never, about once a year or less, every few months, every month, every week, every day, or more than once a day.
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430
426
Table 1 Demographics of entire sample and by reported self-weighing frequency Total (N ¼ 130) Gender % Females Mean age Ethnicity % Caucasian % African American % Hispanic % Asian/Pacific Islander % Native American % Other % Mixed Weight loss group status % Weight loss % No weight loss Mean weight (lbs.) Mean BMI percentile
Infrequent weighing (N ¼ 74)
Frequent Weighing (N ¼ 56)
64.6 15.2
53.6 15.2
36.4 15.3
59.2 13.1 0.8 2.3 6.9 3.1 14.6
55.4 17.6 0.0 0.0 5.4 4.1 17.6
64.3 7.1 1.8 5.4 8.9 1.8 10.7
47.7 52.3 184.5 91.3
37.8 62.2 187.6 93.4
60.7 39.3 180.3 88.6
calories were summed from all reported consumption of daily food intake. Other dietary variables included percent calories from fat, percent calories from protein, percent calories from carbohydrate, grams of fiber, and grams of total sugar.
p .30a .68b .31a
.01a
.34b .08b
BMI ¼ body mas index. a Chi-square. b t-test.
Weight control behaviors. The assessment of weight control behaviors (WCB) was determined through the question, ‘‘During the past year, did you do any of the following things in order to lose weight or keep from gaining weight?’’ A 1-year time increment was used to obtain recent weight control behaviors while allowing for changes in weight control strategies. Responses to 32 individual WCB were dichotomous (yes/no). Four categories were determined through factor analysis: Healthy WCB, Unhealthy WCB, Behavior Change Strategies, and Other Dietary Changes. Healthy WCB included eating fewer calories, increasing exercise, increasing fruit and vegetable intake, cutting out in between meal snacking, eating less high fat food, eating less junk food, watching less television, drinking less soda and sugared beverages, drinking more water, walking more and climbing stairs, spending less time on the computer and playing videogames, and doing different kinds of exercise (Cronbach a ¼ .81). Unhealthy WCB included fasting, skipping meals, smoking cigarettes, using laxatives, using diuretics, using diet pills, and self-vomiting (Cronbach a ¼ .70). Behavior Change Strategies included using attending a weight loss group, writing down what was eaten, working with a professional for weight management, limiting daily caloric intake, counting calories or fat, and following a structured diet (Cronbach a ¼ .61). Other Dietary Changes included eating less meat, eating less carbohydrates, using liquid diet supplements, eating more protein, following the Atkins diet, and following the South Beach diet (Cronbach a ¼ .65). Dietary intake. Dietary intake was assessed using the YouthAdolescent Food Frequency Questionnaire (YAQ). The YAQ has an average Pearson correlation (r ¼ .54) that is similar for adult food frequency questionnaires [15]. Total
Physical activity. Physical activity level was assessed by the Godin Leisure-Time Exercise Questionnaire [16], a measure of leisure time physical activity that has acceptable test–retest reliability (r ¼ .64) [17]. Participants self-reported their frequencies of strenuous, moderate, and mild exercise for more than 15-minute bouts during a typical week. The weekly frequencies of strenuous, moderate, and mild exercise multiplied by nine, five, and three metabolic equivalents, respectively, and summed to calculate the Godin Leisure Time Score. Sedentary activity. Items assessing the hours per week of television viewing and computer usage were developed and validated for the Project EAT study [18]. Test–retest correlations were calculated for weekday television viewing (r ¼ .80), weekend television viewing (r ¼ .69), weekday computer usage (r ¼ .66), and weekend computer usage (r ¼ .71). Self-reported responses for each sedentary activity ranged from 0 hours to more than 5 hours per week. The investigators developed the items pertaining to the hours of video game playing during the weekday and weekend using the same question and response formats as those from Project EAT. For each sedentary activity, weekday and weekend hours were combined to yield hours-per-week measures of television viewing, computer usage, and video game playing. The weekly frequencies of television viewing, computer usage, and video game playing were summed to yield total weekly screen time. Weight and BMI. Trained staff members measured height and body weight using a stadiometer and calibrated scale, respectively, at the time of the assessment. Adolescents selfreported weight and height at their highest weight within the past 2 years and the amount of weight loss. Parents confirmed these self-reported measurements of weight, height, and weight loss. BMI percentiles for corresponding ages of participants at their highest weight before the assessment and at the time of the assessment were calculated from height and weight measurements using the Centers for Disease Control and Prevention (CDC) growth curves [19,20]. Demographic characteristics. Gender, age, and race/ethnicity were self-reported. Data analysis All statistical analyses were conducted using SPSS version 11.0 (SPSS, Inc., Chicago, IL). For all analyses of outcome measures, self-weighing frequency was dichotomized into ‘‘frequent self-weighing’’ (self-weighing every week, every day, or more than once a day) and ‘‘infrequent selfweighing’’ (self-weighing never, about once a year or less, every few months, or every month). To approximate normal
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430
distributions for analysis, the cutoff points for healthy weight control behaviors were determined by distribution medians. Healthy weight control behaviors were dichotomized into 6 or more versus 5 or less. Unhealthy weight control behaviors, behavior change strategies, and other dietary changes were dichotomized into any versus none. Demographic characteristics were compared by selfweighing frequency using t-test and chi-square analyses. Significant differences resulted between self-weighing groups in weight loss status (weight loss vs. no weight loss). To assess the necessity of controlling for these variables, the associations between weight loss status and the outcome measures were evaluated. Given that weight loss status was also significantly related with the outcome measures, weight loss status was controlled in logistic and linear regression analyses. Logistic regression analysis assessed differences in weight control behaviors by self-weighing frequency. Odds ratios (OR) and 95% confidence intervals (CI) that are adjusted for weight loss status are provided, with infrequent self-weighers (IFW) as the reference group. The prevalence of weight control behaviors for frequent and in IFW is provided to aid in interpretation of the odds ratios. Differences in dietary and physical activity variables were evaluated by linear regression. Results
427
IFW. Results are presented for scaled scores and individual weight control behaviors. Healthy weight control behaviors Higher percentages of FWs reported engaging in six or more healthy weight control behaviors during the past year than IFWs (FW ¼ 77.1%; IFW ¼ 26.2%; p ¼ .001). In comparison to IFW, FW were nearly four times more likely to report engaging in six or more of the healthy weight control behaviors during the past year (OR ¼ 3.6, CI ¼ 1.6–7.9). Compared to IFW, FW were more likely to decrease daily caloric intake (FW ¼ 60.6%; IFW ¼ 33.3%; p < .001; OR ¼ 4.0, CI ¼ 1.9–8.6), eat less high fat foods (FW ¼ 57.1%; IFW ¼ 48.4%; p ¼ .02; OR ¼ 2.4, CI ¼ 1.1–5.0), and eat less junk food (FW ¼ 72.8%; IFW ¼ 63.9%; p ¼ .04; OR ¼ 2.2, CI ¼ 1.0–4.9) in order to control their weight. Behavior change strategies In comparison to IFW, more FW used at least one behavior change strategy within the past year (FW ¼ 78.6%; IFW ¼ 38.2%; p ¼ .02). FW had odds of using any behavior strategy that are 2.5 times the odds of IFW (CI ¼ 1.2–5.5). In evaluating specific behaviors in this scale, more FW reported following a structured diet (FW ¼ 12.1%; IFW ¼ 2.3%; p ¼ .03; OR ¼ 4.8, CI ¼ 1.2–19.8).
Frequency of self-weighing Forty two percent (n ¼ 74) of adolescents were categorized as frequent self-weighers (FW) (30.8% weekly, 10.0% daily, and 2.3% more than once a day) and 56.9% (n ¼ 56) were infrequent weighers (11.5% never selfweighed, 11.5% once a year, 18.5% every few months, 15.4 % monthly).
Unhealthy weight control behaviors and other dietary changes No significant relationships were found between unhealthy weight control behaviors and self-weighing frequency or other dietary changes. Dietary intake
Weight status Overall, the FW did not significantly differ from the IFW in terms of gender, mean age, ethnicity, weight, and mean BMI percentile (Table 1). Although not statistically significant, a trend was found between self-weighing groups for BMI percentile, suggesting that frequent weighers had a lower average BMI percentile (FW ¼ 88.6th percentile; IFW ¼ 93.4th percentile; p ¼ .08). BMI percentile ranged from 18.6 to 99.9 for FW and 29.3 to 99.8 for IFW. The mean weight of FW was 180.3 pounds (range of 95.5 to 285.0 pounds), whereas the mean weight of IFS was 187.6 pounds (range of 107.0 to 293.5 pounds). FW were also more likely to have lost 10 pounds in the past 2 years with no subsequent weight regain (FW ¼ 60.7%; IFW ¼ 37.8%; p ¼ .01). Weight control behaviors Table 2 describes the results from the regression analysis, and odds ratios among weight control behaviors for FW and
As shown in Table 3, the mean percentage of total calories from fat was significantly less for FW than IFW (p ¼ .04). A statistical trend was found as FW consumed a greater percentage of total calories from carbohydrates (p ¼ .07). No significance resulted between groups in total caloric intake and percent of total calories from protein. Physical activity In comparison to IFW, FW reported a significantly greater amount of strenuous physical activity as shown in Table 3 (FW ¼ 3.5 15-minute bouts/week; IFW ¼ 2.4 15-minute bouts/week; p ¼ .03). Although not significant, a statistical trend was found in FW reporting more overall time engaged in physical activity than IFW, as measured by the composite Godin Leisure Time score (p ¼ .09). Regarding sedentary activity, the number of hours playing video games was significantly less for FW than IFW (p ¼ .02). Although no significant differences resulted in time spent watching television and using a computer, a trend arose as FW reported
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430
428
Table 2 Prevalence, odds ratios, and 95% confidence intervals for weight control behaviors among frequent self-weighers compared to infrequent self-weighers Prevalence
At least six healthy WCB Decrease calories Eat less high fat foods Eat less junk food/sweets Drink less soda/sugared beverages No between meal snacks Eat more fruits/vegetables Drink more water Increase exercise Do different kinds of exercise Walk more/climb stairs Watch less TV Less computer/video games Any unhealthy WCB Laxatives use Diuretics use Vomiting Skip meals Fasting Diet pills Smoke cigarettes Any behavior change strategy Write what was eaten Count calories or fat Eat a certain amount of calories Structured diet Attend weight loss group Work with professional Any other dietary change Eat more protein Eat less meat Eat less carbohydrates/bread Liquid diet supplements Atkins diet South Beach diet
Odds ratios
Infrequent weighing % (N ¼ 74)
Frequent weighing % (N ¼ 56)
Infrequent weighing OR (N ¼ 74)
Frequent weighing OR (N ¼ 56)
95% CI
26.2
77.1*
1.0
3.6
1.6–7.9
33.3 48.4
60.6* 57.1*
1.0 1.0
4.0 2.4
1.9–8.6 1.1–5.0
63.9
72.8*
1.0
2.2
1.0–4.9
52.2
52.8
1.0
2.0
1.0–4.2
38.9
54.1
1.0
1.8
0.9–3.7
40.2 72.7 75.9 44.9
58.4 75.0 78.4 46.0
1.0 1.0 1.0 1.0
1.9 1.3 2.2 1.6
0.9–4.0 0.6–2.9 0.9–5.4 0.8–3.4
54.4 21.4 15.3 57.9 2.9 3.6 2.8 30.7 12.2 2.0 4.9 38.2
59.2 31.8 15.6 58.3 0.0 0.0 2.2 34.5 15.0 7.1 8.1 78.6*
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
1.6 1.5 1.2 1.0 0.4 0.0 0.5 1.0 1.4 2.2 1.4 2.5
0.8–3.3 0.7–3.5 0.4–3.2 0.5–2.1 0.0–4.6 0.0–7.7 0.0–5.1 0.5–2.4 0.5–4.1 0.3–14.2 0.4–4.3 1.2–5.5
10.1
25.6
1.0
2.0
0.8–4.9
10.2
11.6
1.0
1.6
0.5–5.1
8.4
10.7
1.0
1.8
0.5–6.1
2.3 3.7
12.1* 4.8
1.0 1.0
4.8 2.3
1.2–19.8 0.4–13.8
9.4 53.4 19.7 11.7 15.2 4.3 1.9 4.4
21.0 63.4 20.5 9.8 27.0 1.0 2.5 4.5
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
0.4 1.8 1.4 0.7 1.4 0.5 2.3 1.8
0.3–3.0 0.9–3.7 0.6–3.5 0.2–2.3 0.6–3.1 0.0–6.2 0.2–27.4 0.3–11.8
OR ¼ odds ratio; CI ¼ conficence interval; WCB ¼ weight control behavior. Odds ratios and 95% confidence intervals adjusted for weight loss group status. * p < .05 compared to the infrequent self-weighing group.
fewer overall screen time hours per week (television, computer, and videogames) than IFW (p ¼ .06). Discussion This study examined the relationship between self-weighing frequency and weight control behaviors among adolescents who have a history of being overweight. The primary
finding was that adolescents with a history of overweight who self-weigh at least weekly were four times more likely to report engaging in healthy weight control behaviors and over twice as likely to use behavior change strategies within the past year than adolescents self-weighing monthly or less frequently. Specifically, more frequent self-weighers reported less daily calorie intake, less high-fat food intake, less junk food consumption, and greater use of a structured diet. In
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430 Table 3 Means and standard deviations of dietary intake, physical activity level, and sedentary activity by self-weighing frequency, adjusting for weight loss group status Infrequent weighing Frequent weighing (N ¼ 74) (N ¼ 56) Means (SD) Means (SD) F Dietary intake Calories 1907.8 (1078.6) 1993.0 (933.6) % Calories 31.0 (5.1) 29.2 (5.0) from fat % Calories 16.5 (3.4) 16.3 (2.8) from protein % Calories 53.8 (6.9) 55.9 (5.8) from carbohydrate Physical activity level (15-minute bouts/week) Strenuous 2.4 (2.1) 3.5 (2.5) Moderate 3.8 (2.1) 4.0 (2.5) Mild 4.1 (2.7) 4.8 (4.2) Godin Leisure 53.4 (28.7) 64.6 (33.8) Time Score (Strenuous 3 9) þ (Moderate 3 5) þ (Mild 3 3) Sedentary activity (hours/week) Television 5.5 (3.0) 4.5 (2.4) Computer 3.4 (3.2) 3.4 (3.0) Video games 1.9 (3.1) 0.8 (1.5) Screen time 10.8 (7.0) 8.7 (4.4) (TV þ computer þ video games)
p
1.1 .89 2.1 .04 0.1 .87 1.7 .07
4.2 1.1 1.2 2.2
.03 .41 .45 .09
3.2 0.8 3.1 1.9
.13 .78 .02 .06
addition, FW reported engaging in more vigorous physical activity and spending less time playing videogames than IFW. Interestingly, no significant differences arose between selfweighing groups in unhealthy weight control behaviors. Of note, the majority of adolescents (30.8%) reported weighing themselves on a weekly basis. Few (12.3%) reported weighing themselves daily or more than once a day. These findings support research indicating positive associations between self-weighing frequency and healthy weight control behaviors among overweight adults [8]. Similar to the results among adults, our study measured selfweighing along a behavior construct of frequency, ‘‘How often do you weigh yourself.’’ Conceivably, frequent selfweighing may have been used by participants as a form of self-monitoring for weight control. Self-monitoring has been cited as one of the key components of behavioral weight management [5,7,9]. Perhaps, individuals who regularly monitor their weight are also more likely to actively monitor other healthy weight-related behaviors, such as dietary intake and physical activity. Adolescents reporting self-monitoring of weight at least weekly had lower intakes of dietary fat, more strenuous activity, and showed a trend of a lower BMI compared to IFW. Associations between selfweighing and these positive behaviors may explain why some studies have found that frequent self-weighing predicts sustained weight loss among adults [8] and children [11]. The findings from this study differ from a longitudinal analysis [13] that found frequent self-weighing predictive of unhealthy weight control behaviors among adolescents,
429
including overweight females, and found that self-weighing was not predictive of weight change. Different study designs, sampling, outcome variables, and measures of self-weighing frequency may explain this discrepancy. The previous analysis was longitudinal, allowing for the examination of temporality. The sample for the longitudinal study was populationbased, whereas self-selected adolescents with a history of overweight comprised the sample of our cross-sectional study. Also, self-weighing was assessed by a behavioral measure in our study (‘‘How many times do you weigh yourself’’), whereas the longitudinal study assessed agreement (strongly agree, agree, disagree, strongly disagree) with a cognitive perception variable (‘‘I weigh myself often’’). In addition, the longitudinal study analyzed results by gender, but this study did not as a result of sample size limitations. More research is clearly needed to resolve this discrepancy. It may very well be appropriate for overweight adolescents to monitor their weight on a weekly basis to avoid excessive weight gain. Strengths of this study include a focus on an overweight population of adolescents, including those who had lost weight, and the assessment of a comprehensive list of weight control behaviors and self-weighing. This sample of adolescents who have a history of overweight allowed for a unique analysis of self-weighing frequency and weight control behaviors that would not be possible from a population-based sample. Another strength of this study was the ability to analyze associations between self-weighing frequency and a range of dietary, physical activity, and behavioral variables. As in all studies, limitations need to be recognized. This study is limited by its use of self-report assessments, categorical evaluation of self-weighing, and cross-sectional study design. One discrepancy resulting from the use of self-report measures was found in FW reporting less caloric intake for weight control, while consuming greater daily calories than IFW as assessed by the food frequency questionnaire. Considering more FW reported self-monitoring their food intake, IFW may have tended to underreport their daily dietary intake. Measures that include observed behaviors (e.g., covertly watching adolescents eat) would not be subject to self-report bias. However, measuring observed behaviors would be impractical, costly, and time consuming. Also, the self-weighing frequency assessment was limited to five categories, thus forcing participants to select a response that may not represent actual self-weighing frequency. Of most importance, the current study was cross-sectional. Thus, we cannot state whether self-weighing preceded the use of specific weight control behaviors and weight loss or rather those who had lost weight in the past and/or were currently involved in a weight management plan were weighing themselves more frequently. Despite these limitations, the findings from this study provide important information about self-weighing and weight control behaviors among adolescents who have a history of being overweight. Results indicate that frequent self-weighing was associated with more positive behaviors among the
430
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430
adolescents in our sample. Therefore, for some overweight adolescents, self-weighing may be a useful weight control tool. However, further studies are needed to explore the relationships between self-weighing, weight control behaviors, and psychological well-being among overweight adolescents. Future studies can explore differences between overweight adolescents who report self-weighing weekly, daily, and more than once a day. This analysis could not be conducted in the current study given the few numbers of adolescents reporting self-weighing daily or more than daily. Overall, clinical intervention trials comparing interventions that incorporate selfweighing and weight control behaviors are needed to determine whether self-weighing is helpful or harmful for overweight adolescents, weight management, behavior changes, and overall sense of self and psychological health. Considering the current study was a cross-sectional analysis, longitudinal studies using behavior measures would better predict the long-term impact of frequent self-weighing on weight control behaviors and longterm weight management.
References [1] Ogden CL, Carroll MD, Curtin LR, et al. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA 2006;295:1549–55. [2] Serdula MK, Ivery D, Coates RJ, et al. Do obese children become obese adults? A review of the literature. Prev Med 1993;22(2):167–77. [3] Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity related health risk factors, 2001. JAMA 2003; 289(1):76–9. [4] Finkelstein EA, Fiebelkorn IC, Wang G. National medical spending attributable to overweight and obesity: How much, and who’s paying? Health Affairs 2003;W3:219–26. [5] Boutelle KN, Kirschenbaum DS. Further support for consistent selfmonitoring as a vital component of successful weight control. Obesity Res 1998;6(3):219–24. [6] Weber J, Wertheim EH. Relationships of self-monitoring, special attention, body fat percent and self-motivation to attendance at a community gymnasium. J Sport Exerc Psychol 1989;11:105–14.
[7] Klem ML, Wing RR, McGuire MT, et al. A descriptive study of individuals successful at long-term maintenance of substantial weight loss. Am J Clin Nutr 1998;62:336–45. [8] Linde JA, Jeffery RW, French SA, et al. Self-weighing in weight gain prevention and weight loss trials. Ann Behav Med 2005;30(3):210–6. [9] Wing RR, Tate DF, Gorin AA, et al. A self-regulation program for maintenance of weight loss. N Engl J Med 2006;355:1563–71. [10] Jones DC, Vigfusdottir TH, Lee Y. Body image and the appearance culture among adolescent girls and boys: an examination of friend conversations, peer criticism, appearance magazines, and the internalization of appearance ideals. J Adolesc Res 2004;19:323–39. [11] Epstein LH, Valoski A, Wing RR, et al. Ten-year outcomes of behavioral family-based treatment for childhood obesity. Health Psychol 1994;13:373–83. [12] Levitsky DA, Garay J, Nausbaum M, et al. Monitoring weight daily blocks the freshman weight gain: a model for combating the epidemic of obesity. Int J Obes 2006;30:1003–10. [13] Neumark-Sztainer D, van den Berg P, Hannan PJ, et al. Self-weighing in adolescents; helpful or harmful? Longitudinal associations weigh body weight changes and disordered eating. J Adolesc Health 2006; 39:811–8. [14] Kanfer FH, Karoly P. Self-control: a behavioristic excursion into the lion’s den. Behav Ther 1972;3:398–416. [15] Rockett HRH, Breitenbach M, Frazier AL, et al. Validation of a youth/adolescent food frequency questionnaire. Prev Med 1997; 26(6):808–16. [16] Godin G, Shephard RJ. A simple method to assess exercise behavior in the community. Can J Appl Sport Sci 1985;10(3):141–6. [17] Godin G, Jobin J, Bouillon J. Assessment of leisure time exercise behavior by self-report: a concurrent validity study. Can J Public Health 1986;77(5):359–62. [18] Utter J, Neumark-Sztainer D, Jeffery R, et al. Couch potatoes or French fries: are sedentary behaviors associated with body mass index, physical activity, and dietary behaviors among adolescents? J Am Diet Assoc 2003;103:1298–305. [19] Himes JH, Dietz WH, for the Expert Committee on Clinical Guidelines for Overweight in Adolescent Prevention Services. Guidelines for overweight in adolescent preventive services. Am J Clin Nutr 1994; 59:307–16. [20] Kuczmarski RJ, Ogden Grummer-Strawn LM, et al. CDC Growth Charts: United States. Advance Data from Vital and Health Statistics, No. 314. Hyattsville, MD: National Center for Health Statistics, 2000. 314.
Original article
Self-Weighing and Weight Control Behaviors Among Adolescents with a History of Overweight Mary E. Alm, Ph.D.a, Dianne Neumark-Sztainer, Ph.D., M.P.H., R.D.a,b, Mary Story, Ph.D., R.D.a,b, and Kerri N. Boutelle, Ph.D.a,b,c,d,* b
a Department of Pediatrics and Adolescent Health, University of Minnesota, Minneapolis, Minnesota Department of Epidemiology and Community Health, University of Minnesota, Minneapolis, Minnesota c Department of Pediatrics and Psychiatry, University of California–San Diego, LaJolla, California d Rady Children’s Hospital, San Diego, California Manuscript received April 11, 2008; manuscript accepted August 27, 2008
Abstract
Purpose: To assess the relationships between self-weighing frequency, weight control behaviors, and weight status among male and female adolescents who have a history of being overweight. Methods: This study compared weight control behaviors between two groups of adolescents with a history of being overweight (body mass index [BMI] >85th percentile): those who weighed themselves weekly or more (frequent self-weighers) and those who weighed themselves monthly or less (infrequent self-weighers). Participants completed a survey on weight control behaviors, dietary intake, physical activity, and sedentary activity. Height and weight were also measured. Logistic regression analyses were used for categorical outcomes and linear regressions for continuous outcomes. Results: Of the 130 adolescents, 43% were frequent weighers and 57% were infrequent weighers. In comparison to infrequent self-weighers, frequent self-weighers were more likely to report using behavior change strategies, following a structured diet, and engaging in healthy weight control behaviors, especially decreasing caloric intake, high fat food intake, and ‘‘junk food’’ intake. Also, more frequent self-weighers reported engaging in more strenuous physical activity and spending less time playing videogames than infrequent self-weighers. Although not significant, a trend resulted indicating lower average BMI percentile among frequent self-weighers. No significant differences were found between the two groups in unhealthy weight control behaviors. Conclusions: These results suggest that adolescents with a history of overweight who self-weigh at least weekly are more likely to report using healthy weight control behaviors than adolescents who self-weigh monthly or less frequently. Self-monitoring of weight may be a useful component of a comprehensive weight management plan for some overweight adolescents. Ó 2009 Society for Adolescent Medicine. All rights reserved.
Keywords:
Weight monitoring; Self-monitoring; Pediatrics; Weight management
According to national data, approximately 16% percent of adolescents have a body mass index (BMI) at or above the 95th percentile and 34% of adolescents have a BMI at or above the 85th percentile [1]. Given the high prevalence of overweight and obesity among adolescents and the high risk of obesity continuing from adolescence to adulthood, finding effective weight control strategies for youth is impor*Address correspondence to: Kerri N. Boutelle, Ph.D., University of California-San Diego, 9500 Gilman Drive MC 0985, La Jolla, CA 92093. E-mail address: [email protected]
tant [2]. Early intervention may lessen the grave health and economic consequences associated with adult obesity [3,4]. Research suggests that adults who are successful in weight management self-monitor their food intake, physical activity, and weight [5–7]. Furthermore, studies with adults suggest that regular weight monitoring may be an important component of effective weight loss and long-term weight control [7–9]. For example, one study found that a majority of adults (75%) who were successful at long-term weight loss maintenance reported self-weighing at least weekly [7]. Results of another investigation that evaluated two longitudinal datasets
1054-139X/09/$ – see front matter Ó 2009 Society for Adolescent Medicine. All rights reserved. doi:10.1016/j.jadohealth.2008.08.016
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430
of adults who were enrolled in either weight loss or weight gain prevention trials found that daily self-weighing was associated with greater weight loss and weight loss maintenance [8]. This finding was consistent with a study that found daily self-weighing was associated with improved weight loss maintenance among adults participating in a self-regulation program delivered either over the Internet or in person [9]. Few studies have explored self-weighing and weight control among adolescents. The effects of self-weighing may differ for adolescents, as being heavier than one’s peers or societal standards can negatively impact body image among teenagers [10]. There is concern that repeated weight monitoring may lead to weight preoccupation and unhealthy weight control practices (e.g., fasting, skipping meals, selfvomiting, and laxative use). Furthermore, considering the high prevalence of body dissatisfaction and weight preoccupation among adolescents, especially females [10], an intensified focus on weight may increase the risk for harmful weight control behaviors and disordered eating. Findings from the few studies investigating self-weighing and weight control among adolescents have been mixed. The benefits of self-weighing were noted by one longitudinal study that found daily self-weighing by overweight children ages 6 to 12 participating in a weight loss program were significantly associated with larger decreases in overweight status 10 years posttreatment [11]. This finding was consistent with a study of female undergraduate college students that found daily self-weighing was effective in preventing weight gain [12]. However, these studies did not evaluate the relationship of self-weighing and unhealthy weight control behaviors. Findings from a 5-year longitudinal study assessing self-weighing and weight control practices in a general population of adolescent females indicated that frequent self-weighing was predictive of unhealthy weight control behaviors and was not predictive of better weight management [13]. Regular weight monitoring may increase an adolescent’s awareness of weight fluctuations or gradual weight gain, thus allowing the adolescent to adjust diet and physical activity accordingly and engage in self-regulation [14]. Therefore, frequent self-weighing may be beneficial to long-term weight control. On the other hand, frequent weight monitoring may have unintended consequences and cause negative affect and body dissatisfaction, which may contribute to the development of disordered eating, especially among adolescent females. Although recommending regular self-monitoring of weight may be of concern for a general population of adolescent girls, little is known about the effects of self-weighing on weight control practices among overweight adolescents. Results from investigating these effects would indicate whether self-weighing should be a recommended weight control tool for overweight adolescents, as it is for overweight adults. Therefore, the purpose of this study was to further the knowledge base about the relationship between self-weighing and weight control behaviors among adolescents who have a history of being overweight. The present study aims
425
to examine the relationship between frequent self-weighing and weight control behaviors among adolescents with a history of overweight and to address the research question: Is the frequency of self-weighing related to weight status and weight control behaviors in adolescents with a history of being overweight? Methods Study design Data for this investigation was drawn from the Successful Adolescent Weight Losers (SAL) study, a cross-sectional study of overweight adolescents comparing those who reported weight loss and those who do not. The institutional review board of the University of Minnesota approved the study protocol, and informed consents were obtained from all participants. Dietary intake, physical activity levels, and weight control behaviors were assessed through questionnaires, and height and weight were measured in a clinic setting. Although parent data was also collected for the SAL study, only items from the adolescent’s data was used for the current investigation. Participants One hundred thirty adolescents were recruited through a variety of techniques, including newspaper advertisements, flyers, mailings, and physician referrals. Participation eligibility included being 12 to 20 years old at the time of the assessment and meeting the criteria for overweight (BMI > 85th percentile) within the past 2 years. Exclusion from participating included severe psychiatric disorders and medical conditions with potential to interfere with results, such as depression, eating disorders, diabetes, and endocrine disorders. Of the entire sample, almost half (n ¼ 62) of the adolescents reported weight loss of at least 10 pounds in the past 2 years with no subsequent weight regain for 3 months, and 68 adolescents reported no weight loss (Table 1). The sample included 84 females and 46 males ages 12 to 20, and the mean age of participants at the time of the assessment was 15.2 years (SD ¼ 2.14). Self-reported ethnic identification indicated that over half of the participants were Caucasian (59.2%); 13.1% were African American, 6.9% were Native American, 2.3% were Asian, 0.8% were Hispanic, and 14.6% were ethnically mixed. Measures A questionnaire was administered to participating children at the time of the assessment that addressed the following measures. Self-weighing. Self-weighing frequency was assessed by the following question: ‘‘How often do you weigh yourself?’’ Participants selected one of five responses, which included never, about once a year or less, every few months, every month, every week, every day, or more than once a day.
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430
426
Table 1 Demographics of entire sample and by reported self-weighing frequency Total (N ¼ 130) Gender % Females Mean age Ethnicity % Caucasian % African American % Hispanic % Asian/Pacific Islander % Native American % Other % Mixed Weight loss group status % Weight loss % No weight loss Mean weight (lbs.) Mean BMI percentile
Infrequent weighing (N ¼ 74)
Frequent Weighing (N ¼ 56)
64.6 15.2
53.6 15.2
36.4 15.3
59.2 13.1 0.8 2.3 6.9 3.1 14.6
55.4 17.6 0.0 0.0 5.4 4.1 17.6
64.3 7.1 1.8 5.4 8.9 1.8 10.7
47.7 52.3 184.5 91.3
37.8 62.2 187.6 93.4
60.7 39.3 180.3 88.6
calories were summed from all reported consumption of daily food intake. Other dietary variables included percent calories from fat, percent calories from protein, percent calories from carbohydrate, grams of fiber, and grams of total sugar.
p .30a .68b .31a
.01a
.34b .08b
BMI ¼ body mas index. a Chi-square. b t-test.
Weight control behaviors. The assessment of weight control behaviors (WCB) was determined through the question, ‘‘During the past year, did you do any of the following things in order to lose weight or keep from gaining weight?’’ A 1-year time increment was used to obtain recent weight control behaviors while allowing for changes in weight control strategies. Responses to 32 individual WCB were dichotomous (yes/no). Four categories were determined through factor analysis: Healthy WCB, Unhealthy WCB, Behavior Change Strategies, and Other Dietary Changes. Healthy WCB included eating fewer calories, increasing exercise, increasing fruit and vegetable intake, cutting out in between meal snacking, eating less high fat food, eating less junk food, watching less television, drinking less soda and sugared beverages, drinking more water, walking more and climbing stairs, spending less time on the computer and playing videogames, and doing different kinds of exercise (Cronbach a ¼ .81). Unhealthy WCB included fasting, skipping meals, smoking cigarettes, using laxatives, using diuretics, using diet pills, and self-vomiting (Cronbach a ¼ .70). Behavior Change Strategies included using attending a weight loss group, writing down what was eaten, working with a professional for weight management, limiting daily caloric intake, counting calories or fat, and following a structured diet (Cronbach a ¼ .61). Other Dietary Changes included eating less meat, eating less carbohydrates, using liquid diet supplements, eating more protein, following the Atkins diet, and following the South Beach diet (Cronbach a ¼ .65). Dietary intake. Dietary intake was assessed using the YouthAdolescent Food Frequency Questionnaire (YAQ). The YAQ has an average Pearson correlation (r ¼ .54) that is similar for adult food frequency questionnaires [15]. Total
Physical activity. Physical activity level was assessed by the Godin Leisure-Time Exercise Questionnaire [16], a measure of leisure time physical activity that has acceptable test–retest reliability (r ¼ .64) [17]. Participants self-reported their frequencies of strenuous, moderate, and mild exercise for more than 15-minute bouts during a typical week. The weekly frequencies of strenuous, moderate, and mild exercise multiplied by nine, five, and three metabolic equivalents, respectively, and summed to calculate the Godin Leisure Time Score. Sedentary activity. Items assessing the hours per week of television viewing and computer usage were developed and validated for the Project EAT study [18]. Test–retest correlations were calculated for weekday television viewing (r ¼ .80), weekend television viewing (r ¼ .69), weekday computer usage (r ¼ .66), and weekend computer usage (r ¼ .71). Self-reported responses for each sedentary activity ranged from 0 hours to more than 5 hours per week. The investigators developed the items pertaining to the hours of video game playing during the weekday and weekend using the same question and response formats as those from Project EAT. For each sedentary activity, weekday and weekend hours were combined to yield hours-per-week measures of television viewing, computer usage, and video game playing. The weekly frequencies of television viewing, computer usage, and video game playing were summed to yield total weekly screen time. Weight and BMI. Trained staff members measured height and body weight using a stadiometer and calibrated scale, respectively, at the time of the assessment. Adolescents selfreported weight and height at their highest weight within the past 2 years and the amount of weight loss. Parents confirmed these self-reported measurements of weight, height, and weight loss. BMI percentiles for corresponding ages of participants at their highest weight before the assessment and at the time of the assessment were calculated from height and weight measurements using the Centers for Disease Control and Prevention (CDC) growth curves [19,20]. Demographic characteristics. Gender, age, and race/ethnicity were self-reported. Data analysis All statistical analyses were conducted using SPSS version 11.0 (SPSS, Inc., Chicago, IL). For all analyses of outcome measures, self-weighing frequency was dichotomized into ‘‘frequent self-weighing’’ (self-weighing every week, every day, or more than once a day) and ‘‘infrequent selfweighing’’ (self-weighing never, about once a year or less, every few months, or every month). To approximate normal
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430
distributions for analysis, the cutoff points for healthy weight control behaviors were determined by distribution medians. Healthy weight control behaviors were dichotomized into 6 or more versus 5 or less. Unhealthy weight control behaviors, behavior change strategies, and other dietary changes were dichotomized into any versus none. Demographic characteristics were compared by selfweighing frequency using t-test and chi-square analyses. Significant differences resulted between self-weighing groups in weight loss status (weight loss vs. no weight loss). To assess the necessity of controlling for these variables, the associations between weight loss status and the outcome measures were evaluated. Given that weight loss status was also significantly related with the outcome measures, weight loss status was controlled in logistic and linear regression analyses. Logistic regression analysis assessed differences in weight control behaviors by self-weighing frequency. Odds ratios (OR) and 95% confidence intervals (CI) that are adjusted for weight loss status are provided, with infrequent self-weighers (IFW) as the reference group. The prevalence of weight control behaviors for frequent and in IFW is provided to aid in interpretation of the odds ratios. Differences in dietary and physical activity variables were evaluated by linear regression. Results
427
IFW. Results are presented for scaled scores and individual weight control behaviors. Healthy weight control behaviors Higher percentages of FWs reported engaging in six or more healthy weight control behaviors during the past year than IFWs (FW ¼ 77.1%; IFW ¼ 26.2%; p ¼ .001). In comparison to IFW, FW were nearly four times more likely to report engaging in six or more of the healthy weight control behaviors during the past year (OR ¼ 3.6, CI ¼ 1.6–7.9). Compared to IFW, FW were more likely to decrease daily caloric intake (FW ¼ 60.6%; IFW ¼ 33.3%; p < .001; OR ¼ 4.0, CI ¼ 1.9–8.6), eat less high fat foods (FW ¼ 57.1%; IFW ¼ 48.4%; p ¼ .02; OR ¼ 2.4, CI ¼ 1.1–5.0), and eat less junk food (FW ¼ 72.8%; IFW ¼ 63.9%; p ¼ .04; OR ¼ 2.2, CI ¼ 1.0–4.9) in order to control their weight. Behavior change strategies In comparison to IFW, more FW used at least one behavior change strategy within the past year (FW ¼ 78.6%; IFW ¼ 38.2%; p ¼ .02). FW had odds of using any behavior strategy that are 2.5 times the odds of IFW (CI ¼ 1.2–5.5). In evaluating specific behaviors in this scale, more FW reported following a structured diet (FW ¼ 12.1%; IFW ¼ 2.3%; p ¼ .03; OR ¼ 4.8, CI ¼ 1.2–19.8).
Frequency of self-weighing Forty two percent (n ¼ 74) of adolescents were categorized as frequent self-weighers (FW) (30.8% weekly, 10.0% daily, and 2.3% more than once a day) and 56.9% (n ¼ 56) were infrequent weighers (11.5% never selfweighed, 11.5% once a year, 18.5% every few months, 15.4 % monthly).
Unhealthy weight control behaviors and other dietary changes No significant relationships were found between unhealthy weight control behaviors and self-weighing frequency or other dietary changes. Dietary intake
Weight status Overall, the FW did not significantly differ from the IFW in terms of gender, mean age, ethnicity, weight, and mean BMI percentile (Table 1). Although not statistically significant, a trend was found between self-weighing groups for BMI percentile, suggesting that frequent weighers had a lower average BMI percentile (FW ¼ 88.6th percentile; IFW ¼ 93.4th percentile; p ¼ .08). BMI percentile ranged from 18.6 to 99.9 for FW and 29.3 to 99.8 for IFW. The mean weight of FW was 180.3 pounds (range of 95.5 to 285.0 pounds), whereas the mean weight of IFS was 187.6 pounds (range of 107.0 to 293.5 pounds). FW were also more likely to have lost 10 pounds in the past 2 years with no subsequent weight regain (FW ¼ 60.7%; IFW ¼ 37.8%; p ¼ .01). Weight control behaviors Table 2 describes the results from the regression analysis, and odds ratios among weight control behaviors for FW and
As shown in Table 3, the mean percentage of total calories from fat was significantly less for FW than IFW (p ¼ .04). A statistical trend was found as FW consumed a greater percentage of total calories from carbohydrates (p ¼ .07). No significance resulted between groups in total caloric intake and percent of total calories from protein. Physical activity In comparison to IFW, FW reported a significantly greater amount of strenuous physical activity as shown in Table 3 (FW ¼ 3.5 15-minute bouts/week; IFW ¼ 2.4 15-minute bouts/week; p ¼ .03). Although not significant, a statistical trend was found in FW reporting more overall time engaged in physical activity than IFW, as measured by the composite Godin Leisure Time score (p ¼ .09). Regarding sedentary activity, the number of hours playing video games was significantly less for FW than IFW (p ¼ .02). Although no significant differences resulted in time spent watching television and using a computer, a trend arose as FW reported
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430
428
Table 2 Prevalence, odds ratios, and 95% confidence intervals for weight control behaviors among frequent self-weighers compared to infrequent self-weighers Prevalence
At least six healthy WCB Decrease calories Eat less high fat foods Eat less junk food/sweets Drink less soda/sugared beverages No between meal snacks Eat more fruits/vegetables Drink more water Increase exercise Do different kinds of exercise Walk more/climb stairs Watch less TV Less computer/video games Any unhealthy WCB Laxatives use Diuretics use Vomiting Skip meals Fasting Diet pills Smoke cigarettes Any behavior change strategy Write what was eaten Count calories or fat Eat a certain amount of calories Structured diet Attend weight loss group Work with professional Any other dietary change Eat more protein Eat less meat Eat less carbohydrates/bread Liquid diet supplements Atkins diet South Beach diet
Odds ratios
Infrequent weighing % (N ¼ 74)
Frequent weighing % (N ¼ 56)
Infrequent weighing OR (N ¼ 74)
Frequent weighing OR (N ¼ 56)
95% CI
26.2
77.1*
1.0
3.6
1.6–7.9
33.3 48.4
60.6* 57.1*
1.0 1.0
4.0 2.4
1.9–8.6 1.1–5.0
63.9
72.8*
1.0
2.2
1.0–4.9
52.2
52.8
1.0
2.0
1.0–4.2
38.9
54.1
1.0
1.8
0.9–3.7
40.2 72.7 75.9 44.9
58.4 75.0 78.4 46.0
1.0 1.0 1.0 1.0
1.9 1.3 2.2 1.6
0.9–4.0 0.6–2.9 0.9–5.4 0.8–3.4
54.4 21.4 15.3 57.9 2.9 3.6 2.8 30.7 12.2 2.0 4.9 38.2
59.2 31.8 15.6 58.3 0.0 0.0 2.2 34.5 15.0 7.1 8.1 78.6*
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
1.6 1.5 1.2 1.0 0.4 0.0 0.5 1.0 1.4 2.2 1.4 2.5
0.8–3.3 0.7–3.5 0.4–3.2 0.5–2.1 0.0–4.6 0.0–7.7 0.0–5.1 0.5–2.4 0.5–4.1 0.3–14.2 0.4–4.3 1.2–5.5
10.1
25.6
1.0
2.0
0.8–4.9
10.2
11.6
1.0
1.6
0.5–5.1
8.4
10.7
1.0
1.8
0.5–6.1
2.3 3.7
12.1* 4.8
1.0 1.0
4.8 2.3
1.2–19.8 0.4–13.8
9.4 53.4 19.7 11.7 15.2 4.3 1.9 4.4
21.0 63.4 20.5 9.8 27.0 1.0 2.5 4.5
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
0.4 1.8 1.4 0.7 1.4 0.5 2.3 1.8
0.3–3.0 0.9–3.7 0.6–3.5 0.2–2.3 0.6–3.1 0.0–6.2 0.2–27.4 0.3–11.8
OR ¼ odds ratio; CI ¼ conficence interval; WCB ¼ weight control behavior. Odds ratios and 95% confidence intervals adjusted for weight loss group status. * p < .05 compared to the infrequent self-weighing group.
fewer overall screen time hours per week (television, computer, and videogames) than IFW (p ¼ .06). Discussion This study examined the relationship between self-weighing frequency and weight control behaviors among adolescents who have a history of being overweight. The primary
finding was that adolescents with a history of overweight who self-weigh at least weekly were four times more likely to report engaging in healthy weight control behaviors and over twice as likely to use behavior change strategies within the past year than adolescents self-weighing monthly or less frequently. Specifically, more frequent self-weighers reported less daily calorie intake, less high-fat food intake, less junk food consumption, and greater use of a structured diet. In
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430 Table 3 Means and standard deviations of dietary intake, physical activity level, and sedentary activity by self-weighing frequency, adjusting for weight loss group status Infrequent weighing Frequent weighing (N ¼ 74) (N ¼ 56) Means (SD) Means (SD) F Dietary intake Calories 1907.8 (1078.6) 1993.0 (933.6) % Calories 31.0 (5.1) 29.2 (5.0) from fat % Calories 16.5 (3.4) 16.3 (2.8) from protein % Calories 53.8 (6.9) 55.9 (5.8) from carbohydrate Physical activity level (15-minute bouts/week) Strenuous 2.4 (2.1) 3.5 (2.5) Moderate 3.8 (2.1) 4.0 (2.5) Mild 4.1 (2.7) 4.8 (4.2) Godin Leisure 53.4 (28.7) 64.6 (33.8) Time Score (Strenuous 3 9) þ (Moderate 3 5) þ (Mild 3 3) Sedentary activity (hours/week) Television 5.5 (3.0) 4.5 (2.4) Computer 3.4 (3.2) 3.4 (3.0) Video games 1.9 (3.1) 0.8 (1.5) Screen time 10.8 (7.0) 8.7 (4.4) (TV þ computer þ video games)
p
1.1 .89 2.1 .04 0.1 .87 1.7 .07
4.2 1.1 1.2 2.2
.03 .41 .45 .09
3.2 0.8 3.1 1.9
.13 .78 .02 .06
addition, FW reported engaging in more vigorous physical activity and spending less time playing videogames than IFW. Interestingly, no significant differences arose between selfweighing groups in unhealthy weight control behaviors. Of note, the majority of adolescents (30.8%) reported weighing themselves on a weekly basis. Few (12.3%) reported weighing themselves daily or more than once a day. These findings support research indicating positive associations between self-weighing frequency and healthy weight control behaviors among overweight adults [8]. Similar to the results among adults, our study measured selfweighing along a behavior construct of frequency, ‘‘How often do you weigh yourself.’’ Conceivably, frequent selfweighing may have been used by participants as a form of self-monitoring for weight control. Self-monitoring has been cited as one of the key components of behavioral weight management [5,7,9]. Perhaps, individuals who regularly monitor their weight are also more likely to actively monitor other healthy weight-related behaviors, such as dietary intake and physical activity. Adolescents reporting self-monitoring of weight at least weekly had lower intakes of dietary fat, more strenuous activity, and showed a trend of a lower BMI compared to IFW. Associations between selfweighing and these positive behaviors may explain why some studies have found that frequent self-weighing predicts sustained weight loss among adults [8] and children [11]. The findings from this study differ from a longitudinal analysis [13] that found frequent self-weighing predictive of unhealthy weight control behaviors among adolescents,
429
including overweight females, and found that self-weighing was not predictive of weight change. Different study designs, sampling, outcome variables, and measures of self-weighing frequency may explain this discrepancy. The previous analysis was longitudinal, allowing for the examination of temporality. The sample for the longitudinal study was populationbased, whereas self-selected adolescents with a history of overweight comprised the sample of our cross-sectional study. Also, self-weighing was assessed by a behavioral measure in our study (‘‘How many times do you weigh yourself’’), whereas the longitudinal study assessed agreement (strongly agree, agree, disagree, strongly disagree) with a cognitive perception variable (‘‘I weigh myself often’’). In addition, the longitudinal study analyzed results by gender, but this study did not as a result of sample size limitations. More research is clearly needed to resolve this discrepancy. It may very well be appropriate for overweight adolescents to monitor their weight on a weekly basis to avoid excessive weight gain. Strengths of this study include a focus on an overweight population of adolescents, including those who had lost weight, and the assessment of a comprehensive list of weight control behaviors and self-weighing. This sample of adolescents who have a history of overweight allowed for a unique analysis of self-weighing frequency and weight control behaviors that would not be possible from a population-based sample. Another strength of this study was the ability to analyze associations between self-weighing frequency and a range of dietary, physical activity, and behavioral variables. As in all studies, limitations need to be recognized. This study is limited by its use of self-report assessments, categorical evaluation of self-weighing, and cross-sectional study design. One discrepancy resulting from the use of self-report measures was found in FW reporting less caloric intake for weight control, while consuming greater daily calories than IFW as assessed by the food frequency questionnaire. Considering more FW reported self-monitoring their food intake, IFW may have tended to underreport their daily dietary intake. Measures that include observed behaviors (e.g., covertly watching adolescents eat) would not be subject to self-report bias. However, measuring observed behaviors would be impractical, costly, and time consuming. Also, the self-weighing frequency assessment was limited to five categories, thus forcing participants to select a response that may not represent actual self-weighing frequency. Of most importance, the current study was cross-sectional. Thus, we cannot state whether self-weighing preceded the use of specific weight control behaviors and weight loss or rather those who had lost weight in the past and/or were currently involved in a weight management plan were weighing themselves more frequently. Despite these limitations, the findings from this study provide important information about self-weighing and weight control behaviors among adolescents who have a history of being overweight. Results indicate that frequent self-weighing was associated with more positive behaviors among the
430
M.E. Alm et al. / Journal of Adolescent Health 44 (2009) 424–430
adolescents in our sample. Therefore, for some overweight adolescents, self-weighing may be a useful weight control tool. However, further studies are needed to explore the relationships between self-weighing, weight control behaviors, and psychological well-being among overweight adolescents. Future studies can explore differences between overweight adolescents who report self-weighing weekly, daily, and more than once a day. This analysis could not be conducted in the current study given the few numbers of adolescents reporting self-weighing daily or more than daily. Overall, clinical intervention trials comparing interventions that incorporate selfweighing and weight control behaviors are needed to determine whether self-weighing is helpful or harmful for overweight adolescents, weight management, behavior changes, and overall sense of self and psychological health. Considering the current study was a cross-sectional analysis, longitudinal studies using behavior measures would better predict the long-term impact of frequent self-weighing on weight control behaviors and longterm weight management.
References [1] Ogden CL, Carroll MD, Curtin LR, et al. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA 2006;295:1549–55. [2] Serdula MK, Ivery D, Coates RJ, et al. Do obese children become obese adults? A review of the literature. Prev Med 1993;22(2):167–77. [3] Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity related health risk factors, 2001. JAMA 2003; 289(1):76–9. [4] Finkelstein EA, Fiebelkorn IC, Wang G. National medical spending attributable to overweight and obesity: How much, and who’s paying? Health Affairs 2003;W3:219–26. [5] Boutelle KN, Kirschenbaum DS. Further support for consistent selfmonitoring as a vital component of successful weight control. Obesity Res 1998;6(3):219–24. [6] Weber J, Wertheim EH. Relationships of self-monitoring, special attention, body fat percent and self-motivation to attendance at a community gymnasium. J Sport Exerc Psychol 1989;11:105–14.
[7] Klem ML, Wing RR, McGuire MT, et al. A descriptive study of individuals successful at long-term maintenance of substantial weight loss. Am J Clin Nutr 1998;62:336–45. [8] Linde JA, Jeffery RW, French SA, et al. Self-weighing in weight gain prevention and weight loss trials. Ann Behav Med 2005;30(3):210–6. [9] Wing RR, Tate DF, Gorin AA, et al. A self-regulation program for maintenance of weight loss. N Engl J Med 2006;355:1563–71. [10] Jones DC, Vigfusdottir TH, Lee Y. Body image and the appearance culture among adolescent girls and boys: an examination of friend conversations, peer criticism, appearance magazines, and the internalization of appearance ideals. J Adolesc Res 2004;19:323–39. [11] Epstein LH, Valoski A, Wing RR, et al. Ten-year outcomes of behavioral family-based treatment for childhood obesity. Health Psychol 1994;13:373–83. [12] Levitsky DA, Garay J, Nausbaum M, et al. Monitoring weight daily blocks the freshman weight gain: a model for combating the epidemic of obesity. Int J Obes 2006;30:1003–10. [13] Neumark-Sztainer D, van den Berg P, Hannan PJ, et al. Self-weighing in adolescents; helpful or harmful? Longitudinal associations weigh body weight changes and disordered eating. J Adolesc Health 2006; 39:811–8. [14] Kanfer FH, Karoly P. Self-control: a behavioristic excursion into the lion’s den. Behav Ther 1972;3:398–416. [15] Rockett HRH, Breitenbach M, Frazier AL, et al. Validation of a youth/adolescent food frequency questionnaire. Prev Med 1997; 26(6):808–16. [16] Godin G, Shephard RJ. A simple method to assess exercise behavior in the community. Can J Appl Sport Sci 1985;10(3):141–6. [17] Godin G, Jobin J, Bouillon J. Assessment of leisure time exercise behavior by self-report: a concurrent validity study. Can J Public Health 1986;77(5):359–62. [18] Utter J, Neumark-Sztainer D, Jeffery R, et al. Couch potatoes or French fries: are sedentary behaviors associated with body mass index, physical activity, and dietary behaviors among adolescents? J Am Diet Assoc 2003;103:1298–305. [19] Himes JH, Dietz WH, for the Expert Committee on Clinical Guidelines for Overweight in Adolescent Prevention Services. Guidelines for overweight in adolescent preventive services. Am J Clin Nutr 1994; 59:307–16. [20] Kuczmarski RJ, Ogden Grummer-Strawn LM, et al. CDC Growth Charts: United States. Advance Data from Vital and Health Statistics, No. 314. Hyattsville, MD: National Center for Health Statistics, 2000. 314.