• eating attitudes;
  • dietary calcium;
  • bone density;
  • puberty;
  • prospective;
  • girls


  1. Top of page
  2. Abstract
  7. Acknowledgements

This 2-year prospective study examined associations among bone mineral acquisition and physical, maturational, and lifestyle variables during the pubertal transition in healthy girls. Forty-five girls, initially 10.5 ± 0.6 years, participated. Body composition and bone mineral content (BMC) at the spine and total body (TB) were assessed at baseline and annually thereafter using dual-energy X-ray absorptiometry (DXA). Nutrient intakes were assessed using 3-day diet records and a calcium food frequency questionnaire (FFQ), physical activity by questionnaire, sexual maturation using Tanner's stages of breast and pubic hair maturation, growth by height and weight, and eating attitudes using the children's Eating Attitudes Test (Children's EAT). Mean children's EAT subscale scores (dieting, oral control [OC], and bulimia) were stable over time. Median split of OC subscale scores was used to form high and low OC groups. Groups had similar body composition, dietary intake, activity, and Tanner stage at baseline and 2 years. Using height, weight, and Tanner breast stage as covariates, girls with low OC scores had greater TB BMC at baseline (1452 ± 221 g vs. 1387 ± 197 g; p = 0.030) and 2 years (2003 ± 323 g vs. 1909 ± 299 g; p = 0.049) and greater lumbar spine (LS) BMC at 2 years (45.2 ± 8.8 g vs. 41.2 ± 9.6 g; p = 0.042). In multiple regression analysis, OC score predicted baseline, 2 years, and 2-year change in TB and spinal BMC, contributing 0.9-7.6% to explained variance. Calcium intake predicted baseline, 2 years, and 2-year change in TB BMC, explaining 1.6-5.3% of variance. We conclude that both OC and habitual calcium intake may influence bone mineral acquisition.


  1. Top of page
  2. Abstract
  7. Acknowledgements

IN RECENT years, considerable research attention has been directed toward developing a greater understanding of variables that influence bone mineral accretion.(1) Maximizing peak bone mass reduces the impact of bone loss that inevitably occurs with aging.(2–4) It is recognized that genetic variables predominate in terms of contributing to variability in peak bone mass.(5, 6) However, lifestyle variables, including nutrition and physical activity, also have an influence and are important because they can be modified.(1, 7)

Within the nutrition area, considerable work has been done in children and adolescents in an effort to identify intakes of nutrients, particularly calcium, to support optimal bone mineral accretion.(8) However, the effect that nutrition has on bone may extend beyond the intakes of nutrients and energy and into the realm of cognitions. Attitudes about food and eating can be assessed reliably using psychometric instruments such as the Eating Attitudes Test (EAT),(9) the Eating Disorders Inventory (EDI),(10) and the Three-Factor Eating Questionnaire (TFEQ).(11) Now, it is recognized that eating “attitudes” may have the potential to influence bone metabolism through hormonal mechanisms,(12–21) independently of the nutrient content of foods consumed.

The possibility that variability in eating attitudes within the normal range could affect bone accretion is based on a number of diverse findings in adult women. First, we have found that high scores on the restraint subscale of the TFEQ are associated with increased 24-h cortisol excretion in young women(12) and that restraint score was an independent (negative) predictor of their total body (TB) bone mineral density (BMD) and bone mineral content (BMC).(13) Others have found increased cortisol excretion in the normal range is associated with lower BMD(14, 15) and in older adults, it is associated with increased fracture rate.(16) Second, we and others have found that women with higher scores on the TFEQ restraint subscale have a higher prevalence of subclinical ovulatory and menstrual cycle disturbances.(17–20) Ovulatory disturbances in turn have been related to increased 1-year spinal trabecular bone loss.(21) In these studies, the elevated eating attitude scores were not in the clinical range, the ovulatory disturbances were subclinical, and the women's relative weights were normal.(17–20) Although a luteal phase-related mechanism would not be operative in prepubertal children, it could come into play during and after the pubertal transition. Finally, women athletes with stress fractures had higher scores on the EAT than athletes without stress fractures and with similar relative weights and training volumes.(22) However, once again, the elevated scores were within the normal range. Thus, variability in eating attitudes in normal weight women may have the potential to affect bone mineral status.

Whether eating attitudes could be associated with bone mineral accretion during growth has not been studied, although considerable work has been done to assess children's eating attitudes.(23–29) These studies were conducted using a children's version of the EAT (children's EAT),(23) and observations that girls in particular are becoming preoccupied with weight at very young ages have raised concerns.(24, 28, 29) To date, however, little is known about changes in eating attitudes during the pubertal transition or whether variability in eating attitudes within the normal range has the potential to affect bone accretion.

This study was conducted to monitor prospectively bone mineral accretion during the pubertal transition in girls and its association with growth, maturation, dietary calcium intake, and eating attitudes. We monitored eating attitudes using the children's EAT to determine whether eating attitudes change over time. We also assessed whether eating attitudes were associated with bone mineral status at baseline or with change in BMC over 2 years.


  1. Top of page
  2. Abstract
  7. Acknowledgements


Healthy premenarcheal white girls were recruited using community advertisements for a prospective study of bone mineral accrual. Girls were excluded from participation if they smoked, had chronic use of medications, or had any chronic health conditions that may influence bone metabolism. The study protocol was approved by the University's Clinical Screening Committee for Research and Other Studies Involving Human Subjects. A parent or legal guardian provided signed informed consent and all girls provided written assent.

A total of 51 girls ranging in age from 9 to 12 years (mean, 10.5 ± 0.7 years) participated in baseline measurements. Six girls did not return for follow-up measures because they moved from the area (n = 2) or were too busy to come in for testing (n = 4). Data are presented for the 45 girls followed for 2 years.


Anthropometric measures and maturity assessments were made twice yearly. Height (without shoes) was measured to the nearest 0.1 cm using a stadiometer. Weight was measured to the nearest 0.1 kg using a medical beam balance scale. Duplicate measurements of height and weight were made and if measurements differed by more than 0.4 cm or 0.2 kg, a third measurement was made. The average of two or median of three measures are reported. Height and weight were used to calculate body mass index (BMI; kg/m2). A woman reproductive endocrinologist assessed breast and pubic hair maturation using the stages described by Tanner.(30) If discrepancies existed between breast and pubic hair stage, breast stage was used.(31) TB and lumbar spine (LS) BMC was measured annually on a Lunar DPX dual-energy X-ray absorptiometer (DXA) (software version 3.1; Lunar, Inc., Madison, WI, USA). Body composition (% fat and % bone mineral-free lean tissue) also was derived from DXA measurements. The in vivo coefficient of variation (CV) on 16 premenopausal women was 1.3% for the LS BMD on this machine(32) and TB CV was 0.7%.(33)


Calcium intake was estimated twice yearly from a food-frequency questionnaire (FFQ), which has been validated in this age group.(34) The questionnaire was administered orally, with portion sizes and frequency of intake clarified by one of the researchers. Participants also were requested to complete 3-day food records, (including 2 weekdays and 1 weekend day) twice each year for a total of up to five 3-day records per subject. Girls were instructed individually on how to complete the records, which were analyzed for energy, macronutrients, and calcium using Food Processor II (ESHA Research, Salem, OR, USA). Physical activity was assessed twice yearly using a modification of Slemenda's questionnaire.(35) The questionnaire, also administered orally, asked children the number of times per week and minutes per time they participated in physical education, walked or cycled to school, and spent time doing physical activity outside of school (including walking, running, swimming, cycling, basketball, volleyball, etc.). Estimated hours of physical activity per week are reported.

Eating attitudes

Participants' perceptions of their eating behavior were assessed annually using the children's version(23) of the 26-item EAT.(9) The children's EAT, like the original EAT, has subscales for dieting (13 items), bulimia (6 items), and oral control (OC; 7 items). Items in the dieting subscale reflect shape preoccupations and an avoidance of fattening foods; those in the bulimia subscale reflect the tendency to overeat and/or purge; those in the OC subscale reflect self-control of food intake and social pressures to eat.(9) Items in the children's EAT have six possible response options (ranging from “never” to “always”) and subscale scores were derived in two ways: (1) using the conventional scoring system, in which responses of “never,” “rarely,” and “sometimes” are scored as 0, and “often,” “very often,” and “always” are scored as 1-3, respectively(9); and (2) using a continuous scoring system of 1-6 for “never,” “rarely,” “sometimes,” “often,” “very often,” and “always.” The continuous scoring system was used to provide the potential to detect subtle changes in eating attitudes, should they occur; for example, using the conventional scoring system, a change over time from “never” to “sometimes” would not be detected. The continuous scoring system was used in correlation analysis and regression models. In addition, the conventional scoring system was used to facilitate comparisons to the literature.

Statistical analyses

The data were analyzed using programs available in the Statistical Package for the Social Sciences (version 8.0, 1997; SPSS, Inc., Chicago, IL, USA). After examination of the data for normality, Pearson correlation analysis was conducted to examine bivariate associations between the variables monitored and whole body and LS BMC at baseline, BMC at 2 years, and 2-year change in BMC. Specifically, associations with eating attitudes were assessed using the total children's EAT score and each of the subscale scores. This analysis revealed significant associations between BMC and the OC subscale score. These relationships were examined further by comparing physical characteristics, lifestyle variables, and bone data between girls whose average OC scores (i.e., mean score throughout the study) were above (n = 23) or below (n = 22) the mean average score. These groups are referred to as having high (score, 17.0-25.0) or low (score, 10.3-16.7) OC scores, respectively. Analysis of covariance also was used when examining differences in BMC between groups with high and low OC scores. For absolute BMC at baseline and 2 years, height, weight, and Tanner breast stage were entered as covariates. When change in BMC was the dependent variable, covariates were baseline BMC, change in height, change in weight, and Tanner breast stage at 2 years.

Regression models were used to assess the effects of established and theoretical predictors of BMC, including height, weight, physical activity, energy intake, calcium intake by FFQ, Tanner breast stage, and OC score. FFQ calcium intake was used instead of calcium intake by 3-day record because the number of 3-day records completed by participants varied. For baseline and 2-year BMC, height, weight, and Tanner breast stage at baseline or 2 years were entered as a block in the first step of all models, and the remaining variables were selected using a stepwise procedure. Models predicting 2-year changes in BMC, baseline BMC, change in height, change in weight, and Tanner breast stage at 2 years were entered as a block in the first step. Because OC, energy intake, calcium intake, and physical activity did not change significantly over time, average values for these variables were used in all analyses.

The level of significance was set at p < 0.05 for group comparisons and p < 0.10 for regression models, because the latter were considered exploratory. All comparisons were two-tailed.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Baseline characteristics for the 45 girls are presented in Table 1. With the exception of the bulimia subscale, children's EAT total and subscale scores were significantly related at baseline and at 2 years (Table 2), whether calculated using the conventional scoring system(9) or the continuous system.

Table Table 1.. Characteristicsa of Peripubertal Girls (n = 45) at Study Enrollment
Thumbnail image of
Table Table 2.. Average Total and Subscale Scores on the Children's EAT Calculated Using Conventionala and Continuousb Scoring and Correlations of Scores at Baseline and 2 Years
Thumbnail image of

Physical and lifestyle characteristics of OC subgroups

Physical characteristics at baseline, at 2 years, and the change over 2 years are compared between groups of girls with above- and below-average OC scores (Table 3). The two groups were similar in age, height, weight, BMI, percent body fat, and percent lean mass at baseline and at 2 years. The 2-year changes in these variables also were similar between groups. At baseline, the percentage of girls in each group at each Tanner breast stage was similar, with a majority of the girls at Tanner breast stage 1 or 2 (n = 38) and the remaining 7 girls at stage 3. At 2 years, only 4 girls remained at Tanner breast stages 1 or 2, and of the remainder, 12 were stage 3, 20 were stage 4, and 9 were stage 5. The distribution between OC groups was again similar.

Table Table 3.. Physical and Lifestyle Characteristics of Peripubertal Girls with Low and High Scores for OCa at Baseline, 2 Years, and Change over 2 Years (Mean ± SD)
Thumbnail image of

Average hours of physical activity were similar between the low and high OC groups (9.8 ± 4.1 h/wk vs. 9.8 ± 2.6 h/wk; p = 0.954) and were correlated at baseline and 2 years (r = 0.408; p = 0.005). Calcium intakes, whether assessed by FFQ or 3-day diet records, also were similar between groups and over time. Average calcium intakes by FFQ were 856 ± 391 mg/day for the low OC group versus 807 ± 316 mg/day for the high OC group (p = 0.645). By 3-day records, corresponding values were 979 ± 366 mg/day versus 974 ± 263 mg/day (p = 0.954). Average intakes as assessed using the two methods were correlated (r = 0.765; p < 0.001). Finally, reported energy intakes of the two groups did not differ, and averaged 1893 ± 311 kcal/day and 1808 ± 328 kcal/day (p = 0.377) for the low and high OC groups, respectively.


When height, weight, and Tanner breast stage were included as covariates, girls with low scores for OC had greater TB BMC at baseline (1452 ± 221 g vs. 1387 ± 197 g; p = 0.030) and 2 years (2003 ± 323 g vs. 1909 ± 299 g; p = 0.049). In addition, with the same covariates, girls with low OC had greater LS BMC at 2 years (45.2 ± 8.8 g vs. 41.2 ± 9.6 g; p = 0.042). However, LS BMC at baseline was not significantly different (29.2 ± 5.3 g vs. 27.0 ± 5.1 g; p = 0.093).

In addition to having lower baseline TB BMC, girls with high OC gained 1-2% less TB and LS BMC than girls with low OC over 2 years. However, when baseline BMC, change in height, change in weight, and Tanner breast stage were included as covariates the differences in BMC gain were not statistically significant (p > 0.05).

Regression models

Regression models for BMC at baseline, at year 2, and for 2-year change are presented in Table 4. OC was a significant predictor of LS and TB BMC at baseline and 2 years and of 2-year change in LS and TB BMC. In every case, after body size (height and weight), maturation (Tanner breast stage), and, as appropriate, baseline BMC were entered into the model, OC explained an additional 0.9-7.6% of the variance. Average calcium intake was also a significant predictor of TB BMC at baseline and 2 years and of the 2-year change TB BMC change. Calcium intake contributed an additional 1.6-5.3% to explained variance.

Table Table 4.. Stepwise Regression Models for TB and LS BMC at Baseline, 2 Years, and 2-Year Change in BMCa
Thumbnail image of


  1. Top of page
  2. Abstract
  7. Acknowledgements

Our results established the expected associations among growth, maturation, and BMC in peripubertal children. As has been shown previously,(3, 36) physical parameters, growth, and maturation were quantitatively the most important predictors of initial BMC and of its change over time. However, our study differed from previous work in our observation that eating attitudes, specifically scores on the OC subscale of the children's EAT, were associated negatively with baseline bone status and 2-year changes in bone. To our knowledge, this is the first study to detect relationships between BMC and eating attitudes in normal, healthy, peripubertal girls. We also found that habitual calcium intake was associated independently with change in TB BMC, and to our knowledge this not been shown previously in a prospective observational study.

Previous work in adolescent and adult women has established that clinical eating disorders, most notably anorexia nervosa, are associated with bone loss and/or failure to gain bone.(37–39) The mechanism of the bone loss is multifactorial, but at a minimum includes the effects of weight loss, inadequate nutrient intakes, and reduced or virtually absent exposure to reproductive hormones. Psychosocial stress also may contribute because hypercortisolism has been established in eating disorder patients and can have an impact on bone.(40, 41)

However, our peripubertal participants did not have clinical eating disorders. Their total scores on the children's EAT, which we used to assess eating attitudes in this study, were comparable with or even slightly lower than those reported for healthy girls of a similar age in other studies.(24–29) Scores above 20 on the conventionally scored (0-3) EAT are consistent with eating attitudes seen in patients with eating disorders(9); however, none of the girls in this study had scores above 16. It is worth noting that children's EAT scores did not increase over time in this sample, despite substantial changes in body dimensions over the 2 years of the study. However, it is possible that changes in eating attitudes may not occur coincidentally with changes in body size, and scores of these girls could increase at a later age.(25, 26) Nevertheless, the stability of the scores over the pubertal transition suggests that eating attitudes may be a stable personality trait and established relatively early in life. In this regard, mothers' beliefs and behaviors may be important: 5-year-old girls' beliefs about the relationship of food intake to body weight recently were shown to be associated with their mothers' weight control attempts.(42) At present, it is not known whether eating attitudes can be modified, and whether such modification would prevent or reverse the apparent adverse associations with bone. Nevertheless, at a minimum, efforts to develop positive healthy eating attitudes in young girls appear warranted.

We did not have data on psychosocial stress or cortisol secretion in our participants and thus cannot determine whether differences in these domains may have contributed to the observed differences in bone. This merits further research, because we have found high levels of cognitive dietary restraint in college-aged women to be associated with increased 24-h urinary cortisol excretion.(12) Over the long-term, increased cortisol excretions or levels even within the normal range may predispose to reduced BMD and fragility fractures.(14–16) It should be emphasized that cognitive dietary restraint, assessed in the college women, is not synonymous with OC that was assessed in the present study. However, both are dimensions of attitudes about eating and reflect an individual's perceptions about eating behavior. At present, a children's version of the TFEQ, from which the cognitive dietary restraint subscale is derived, is not available.

Because our finding of an association between eating attitudes and bone in peripubertal girls is novel, the possibility that the observation may be confounded by differences in other variables, such as reproductive hormones, anthropometric variables, or nutritional variables, needs to be examined carefully. Differences in reproductive hormone status do not appear to explain our results. We did not have access to blood hormone levels; however, no differences were observed in Tanner breast stage between girls with high and low OC scores at baseline, at 2 years, or in the change in breast stage over time. Although we cannot discount the possibility that subtle differences in reproductive hormones may have existed between groups, they were certainly not of the magnitude associated with clinical eating disorders. Moreover, OC scores remained associated with BMC even after controlling for Tanner breast stage in the analysis.

Similarly, it appears that anthropometric differences do not confound the observation. There were no differences in body weight between groups at baseline, at 2 years, or in the 2-year change in weight. Body composition (i.e., lean mass vs. fat mass) also was similar between groups, both at baseline and over time. Girls with low OC scores were nonsignificantly larger than girls with high scores, but percentage body composition (lean vs. fat) and BMI were virtually identical between groups. Finally, OC remained a significant predictor in multiple regression analysis that controlled for body size and growth.

Energy intakes and reported physical activity levels, which could affect energy availability for growth, also appeared similar between girls with high and low OC scores, as were macronutrient and calcium intakes. Thus, differences in growth, maturational and nutritional variables, with the potential to influence bone were not observed between OC groups, and the associations between OC and bone remained significant even when these variables were controlled. Accordingly, it is unlikely that they confounded the observed association between bone and OC scores. Although OC is not a trait that can be randomized, it should be emphasized that the observational nature of the study precludes making causal inferences. Other psychosocial variables also may relate to changes in bone and, indeed, may be stronger predictors than OC scores.

Another finding of this study was that calcium intake was associated with BMC at baseline and its change over time. We believe that this is an important observation and supports recommendations for generous calcium intake during growth.(8) The available data on the effects of calcium intake on bone mineral accretion are unclear; although randomized trials have demonstrated consistently that increased calcium intake during growth leads to greater BMC gains during the period of increased intake, in most cases the increases appear transitory and diminish after supplementation ends.(43) This has been interpreted as meaning either that the additional calcium was of no benefit or that the higher intake needs to be maintained for benefits to persist. Similarly, previous prospective studies have not found habitual calcium intake to be an independent predictor of bone mineral accrual during growth.(44–46)

The reason for our ability to detect associations between calcium intake and BMC change and the absence of such observations in other studies cannot be established, but the method we used to estimate calcium intake might play a role. Most other prospective studies used repeated 24-h recalls or food records to quantitate calcium intake, and in most cases, the number of records obtained was not adequate to estimate individual intakes with accuracy.(47, 48) Our primary analysis for calcium was based on calcium intake as assessed using an FFQ that reflected calcium intake over the previous month. This questionnaire was administered five times during the 2-year study; accordingly, it is possible that we were able to estimate usual calcium intake with greater precision than has occurred in many studies. Frequent administration of the questionnaire also may have been important, because nutrient intakes of adolescents do not appear to track over time.(49) A second possibility is that the habitual calcium intake of our participants, whether assessed by FFQ or 3-day record, was below the putative threshold at which additional calcium is no longer beneficial in this age group.(50, 51) Habitual calcium intakes were higher in other studies(44, 45) than in our study.

In conclusion, the results of this study indicate that cognitive factors, in this case OC scores on the children's EAT, may be associated negatively with bone accretion. Further research is necessary to determine the mechanism of this phenomenon, as well as what can be done to prevent or reverse the apparently harmful effect on bone. The data also provide confirmatory evidence of the importance of habitual calcium intake in contributing to bone mineral accretion during growth.


  1. Top of page
  2. Abstract
  7. Acknowledgements

We sincerely acknowledge the interest, enthusiasm, and cooperation of the girls who took part in this study and the family members who supported their participation. Sincere thanks are expressed to Cathy Langdon and Maya Spaeth for assistance with the analysis of dietary intake data. We also thank Dr. Judith Hall for her assistance during the initiation of the study. This study was supported by grants from the British Columbia Health Care Research Foundation and the British Columbia Children's Hospital Foundation.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  • 1
    Ralston SH 1997 What determines peak bone mass and bone loss? Bailliere's Clin Rheumatol 11:479494.
  • 2
    Bailey DA, Faulkner RA, McKay HA 1996 Growth, physical activity, and bone mineral acquisition. In: HolloszyJO (ed.). Exercise and Sport Sciences Reviews. Williams & Wilkins, Baltimore, MD, USA, pp. 233266.
  • 3
    Bailey DA 1997 The Saskatchewan Pediatric Bone Mineral Accrual Study: Bone mineral acquisition during the growing years. Int J Sports Med 18:S191S194.
  • 4
    Matkovic V, Kostial K, Simonovic I, Buzina R, Brodarec A, Nordin C 1979 Bone status and fracture rates in two regions of Yugoslavia. Am J Clin Nutr 32:540549.
  • 5
    Gueguen R, Jouanny P, Guillemin F, Kuntz C, Pourel J, Siest G 1995 Segregation analysis and variance components analysis of BMD in healthy families. J Bone Miner Res 10:20172022.
  • 6
    Krall EA, Dawson-Hughes B 1993 Heritable and lifestyle determinants of BMD. J Bone Miner Res 8:19.
  • 7
    Barr SI, McKay HA 1998 Nutrition, exercise and bone status in youth. Int J Sport Nutr 8:124142.
  • 8
    Standing Committee on the Scientific Evaluation of Dietary Reference Intakes, Food and Nutrition Board, Institute of Medicine 1997 Calcium. In: Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride. National Academy Press, Washington, DC, USA, pp. 71145.
  • 9
    Garner DM, Olmsted MP, Bohr Y, Garfinkel PE 1982 The Eating Attitudes Test: Psychometric features and clinical correlates. Psychol Med 12:871878.
  • 10
    Garner DM, Olmsted MP, Polivy J 1983 Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia. Int J Eat Disord 2:1534.
  • 11
    Stunkard AJ, Messick S 1985 The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. J Psychosom Res 29:7183.
  • 12
    McLean JA, Barr SI, Prior JC 2001 Cognitive dietary restraint is associated with higher urinary cortisol excretion in healthy premenopausal women. Am J Clin Nutr 73:712.
  • 13
    McLean JA, Barr SI, Prior JC 2001 Dietary restraint, exercise and bone density in young women: Are they related? Med Sci Sports Exerc (in press.)
  • 14
    Torlontano M, Chiodini I, Pileri M, Guglielmi G, Cammisa M, Modoni S, Carnevale V, Trischitta V, Scillitani A 1999 Altered bone mass and turnover in female patients with adrenal incidentaloma: The effect of subclinical hypercortisolism. J Clin Endocrinol Metab 84:23822385.
  • 15
    Dennison E, Hindmarsh P, Fall C, Kellingray S, Barker D, Phillips D, Cooper C 1999 Profiles of endogenous circulating cortisol and BMD in healthy elderly men. J Clin Endocrinol Metab 84:30583063.
  • 16
    Greendale GA, Unger JB, Rose JW, Seeman TE 1999 The relationship between cortisol excretion and fractures in healthy older people: Results from the MacArthur Studies—Mac. J Am Geriatr Soc 47:799803.
  • 17
    Schweiger U, Tuschl RJ, Platte P, Broocks A, Laessle RG, Pirke KM 1992 Everyday eating behavior and menstrual function in young women. Fertil Steril 57:771775.
  • 18
    Barr SI, Janelle K, Prior JC 1994 Vegetarian versus nonvegetarian diets, dietary restraint, and subclinical ovulatory disturbances: Prospective six-month study. Am J Clin Nutr 60:887894.
  • 19
    Barr SI, Prior JC, Vigna YM 1994 Restrained eating and ovulatory disturbances: Possible implications for bone health. Am J Clin Nutr 59:9297.
  • 20
    Lebenstedt M, Platte P, Pirke KM 1999 Reduced resting metabolic rate in athletes with menstrual disturbances. Med Sci Sports Exerc 31:12501256.
  • 21
    Prior JC, Vigna YM, Schechter MT, Burgess AE 1990 Spinal bone loss and ovulatory disturbances. N Engl J Med 323:12211227.
  • 22
    Bennell KL, Malcolm SA, Thomas SA, Ebeling PR, McCrory PR, Wark JD, Brukner PD 1995 Risk factors for stress fractures in female track-and-field athletes: A retrospective analysis. Clin J Sport Med 5:229235.
  • 23
    Maloney MJ, McGuire J, Daniels SR 1988 Reliability testing of a children's version of the Eating Attitudes Test. J Am Acad Clin Adolesc Psychiatr 25:541543.
  • 24
    Maloney MJ, McGuire J, Daniels SR, Specker B 1989 Dieting behavior and eating attitudes in children. Pediatrics 84:482489.
  • 25
    Sasson A, Lewin C, Roth D 1995 Dieting behavior and eating attitudes in Israeli children. Int J Eat Disord 17:6772.
  • 26
    De Castro JM, Goldstein SJ 1995 Eating attitudes and behaviors of pre- and postpubertal females: Clues to the etiology of eating disorders. Physiol Behav 58:1523.
  • 27
    Edlund B, Hallqvist G, Sjoden P-O 1994 Attitudes to food, eating and dieting behavior in 11- and 14-year-old Swedish children. Acta Paediatr 83:572577.
  • 28
    Markovic J, Votava-Raic A, Nikolic S 1998 Study of eating attitudes and body image perception in the preadolescent age. Coll Antropol 1:221232.
  • 29
    Rolland K, Farnill D, Griffiths RA 1997 Body figure perceptions and eating attitudes among Australian school children aged 8 to 12 years. Int J Eat Disord 21:273278.
  • 30
    Tanner, JM 1955 Growth at Adolescence, Oxford, Blackwell Scientific Publishing, Oxford, UK.
  • 31
    Warren MP 1983 Effects of undernutrition on reproductive function in the human. Endocr Rev 4:363377.
  • 32
    Prior JC, Vigna YM, Barr SI, Rexworthy C, Lentle BC 1994 Cyclic medroxyprogesterone treatment increases bone density: A controlled trial in active women with menstrual cycle disturbances. Am J Med 96:521530.
  • 33
    Prior JC, Vigna YM, Wark JD, Eyre DR, Lentle BC, Li DK, Ebeling PR, Atley L 1997 Premenopausal ovariectomy-related bone loss: A randomized, double-blind one year trial of conjugated estrogen or medroxyprogesterone acetate. J Bone Miner Res 12:18511863.
  • 34
    Barr SI 1994 Associations of social and demographic variables with calcium intakes of high school students. J Am Diet Assoc 94:260266, 269.
  • 35
    Slemenda CW, Miller JZ, Hui SL, Reister TK, Johnston CC 1991 Role of physical activity in the development of skeletal mass in children. J Bone Miner Res 6:12271233.
  • 36
    Lloyd T, Chinchilli VM, Eggli DF, Rollings N, Kulin HE 1998 Body composition development of adolescent white females. The Penn State Young Women's Health Study. Arch Pediatr Adolesc Med 152:9981002.
  • 37
    Grinspoon S, Miller K, Coyle C, Krempin J, Armstrong C, Pitts S, Herzog D, Klibanski A 1999 Severity of osteopenia in estrogen-deficient women with anorexia nervosa and hypothalamic amenorrhea. J Clin Endocrinol Metab 84:20492055.
  • 38
    Ward A, Brown N, Treasure J 1997 Persistent osteopenia after recovery from anorexia nervosa. Int J Eat Disord 22:7175.
  • 39
    Powers PS 1999 Osteoporosis and eating disorders. J Pediatr Adol Gynecol 12:5157.
  • 40
    Vierhapper H, Kiss A, Nowotny P, Wiesnagrotzki S, Monder C, Waldhausl W 1990 Metabolism of cortisol in anorexia nervosa. Acta Endocrinol 122:753758.
  • 41
    Newman MM, Halmi KA 1989 Relationship of bone density to estradiol and cortisol in anorexia nervosa and bulimia. Psychiatry Res 29:105112.
  • 42
    Abramovitz BA, Birch LL 2000 Five-year-old girls' ideas about dieting are predicted by their mothers' dieting. J Am Diet Assoc 100:11571163.
  • 43
    Barr SI, McKay HA 1998 Nutrition, exercise and bone status in youth. Int J Sport Nutr 8:124142.
  • 44
    Welten DC, Kemper HCG, Post GB, Van Mechelen W, Twisk J, Lips P, Teule GJ 1994 Weight-bearing activity during youth is a more important factor for peak bone mass than calcium intake. J Bone Miner Res 9:10891096.
  • 45
    Bailey DA, McKay HA, Mirwald RL, Crocker PRE, Faulkner RA 1999 A six-year longitudinal study of the relationship of physical activity to bone mineral accrual in growing children: The University of Saskatchewan Bone Mineral Accrual Study. J Bone Miner Res 14:16721679.
  • 46
    Lloyd T, Chinchilli VM, Johnson-Rollings N, Kieselhorst K, Eggli DF, Marcus R 2000 Adult female hip bone density reflects teenage sports-exercise patterns but not teenage calcium intake. Pediatrics 106:4044.
  • 47
    Miller J, Kimes T, Hui S, Andon MB, Johnston CC 1991 Nutrient intake variability in a pediatric population: Implications for study design. J Nutr 121:265274.
  • 48
    Basiotis PP, Welsh SO, Cronin FJ, Kelsay JL, Mertz W 1987 Number of days of food intake records required to estimate individual and group nutrient intakes with defined confidence. J Nutr 117:16381641.
  • 49
    Cusatis DC, Chinchilli VM, Johnson-Rollings N, Kieselhorst K, Stallings VA, Lloyd T 2000 Longitudinal nutrient intake patterns of U.S. adolescent women: The Penn State Young Women's Health Study. J Adolesc Health 26:194204.
  • 50
    Jackman LA, Millane SS, Martin BR, Wood OB, McCabe GP, Peacock M, Weaver CM 1997 Calcium retention in relation to calcium intake and postmenarchal age in adolescent females. Am J Clin Nutr 66:327333.
  • 51
    Matkovic V, Heaney RP 1992 Calcium balance during growth: Evidence for threshold behavior. Am J Clin Nutr 55:992996.