- Email: [email protected]
Adolescents’ Low-Carbohydrate-Density Diets Are Related to Poorer Dietary Intakes
RESEARCH Research and Professional Briefs
Adolescents’ Low-Carbohydrate-Density Diets Are Related to Poorer Dietary Intakes LINDA S. GREENE-FINESTONE, PhD, RD; M. KAREN CAMPBELL, PhD; SUSAN E. EVERS, PhD, RD; IRIS A. GUTMANIS, PhD
Web site exclusive! Editor’s note: Table 4 that accompanies this article is available online at www.adajournal.org. ABSTRACT This study was undertaken to assess how low-carbohydrate-density diets below the acceptable macronutrient distribution range relate to food and micronutrient intake and sociodemographic and health-related characteristics. The multistage stratified cluster design in the 1990 Ontario Health Survey was used. There were 5,194 subjects, 12 to 18 years of age, in sampled households. Dietary data were collected via a food frequency questionnaire. Lowcarbohydrate-density diets were consumed by 27.6% of males and 24.1% of females. Low-carbohydrate-density diets were related (P⬍.05) to reduced sufficiency of vegetables and fruit and higher consumption of meat and alternatives and added fats. The low-carbohydrate-density diet resulted in intakes lower in vitamin C and fiber and higher in cholesterol and total fat. The low-carbohydrate-density diet was directly associated with being Canadian-born (odds ratio [OR]⫽1.78, 95% confidence interval [CI]⫽1.27 to 2.50), overweight status (OR⫽1.27, 95%
L. S. Greene-Finestone is nutrition advisor in the Office of Nutrition Policy and Promotion, Health Canada, Ottawa, Ontario, Canada, and was previously enrolled in a doctoral program in the Department of Epidemiology and Biostatistics, University of Western Ontario, London, Ontario, Canada. M. K. Campbell is chair of the Department of Epidemiology and Biostatistics, University of Western Ontario, London, Ontario, Canada. S. E. Evers is a professor in the Department of Family Relations and Applied Nutrition, University of Guelph, MacDonald Institute, Guelph, Ontario, Canada. I. A. Gutmanis is director, Evaluation and Research, Specialized Geriatric Services, St Joseph’s Health Care, Parkwood Hospital, London, Ontario, Canada; at the time of the study, she was director of the Southwest Region Health Information Partnership, London, Ontario, Canada. Address correspondence to: Linda S. Greene-Finestone, PhD, RD, Office of Nutrition Policy and Promotion, Health Canada, Qualicum Tower A, 2936 Baseline Rd, 3rd Floor, Ottawa, Ontario, Canada, K1A 0K9 A.L. 3303D. E-mail: [email protected] Copyright © 2005 by the American Dietetic Association. 0002-8223/05/10511-0008$30.00/0 doi: 10.1016/j.jada.2005.08.014
© 2005 by the American Dietetic Association
CI⫽1.02 to 1.57), smoking (OR⫽1.53, 95% CI⫽1.23 to 1.90), alcohol use (OR⫽1.46, 95% CI⫽1.21 to 1.75), and poorer self-rated health (OR⫽1.47, 95% CI⫽1.01 to 2.14). Use of the acceptable macronutrient distribution range identified adolescents with low-carbohydrate-density diets whose food choices and nutrient intake may impact negatively on short- and long-term health. J Am Diet Assoc. 2005;105:1783.e1-1783.e6.
T
he eating behaviors of adolescents affect their current health and may impact on adult-onset chronic disease (1,2). Fruit, vegetables, and grains are particularly implicated with decreased risk of diseases such as cancer, cardiovascular disease, and stroke (3-7). Teenagers’ poor consumption of fruit and vegetables (8-10) is of concern because eating behaviors tend to consolidate by adolescence and follow a similar pattern into young adulthood (8,11). The most recent nutrition recommendations of the United States and Canada are the Dietary Reference Intakes (12). Of the Dietary Reference Intakes, the acceptable macronutrient distribution range recommendations can be a general diet evaluation guide (13). The acceptable macronutrient distribution ranges were estimated to account for the roles of macronutrients in chronic disease development and in achievement of micronutrient sufficiency. Given the long-term consequences of eating patterns formed during adolescence and the roles of acceptable macronutrient distribution ranges in promoting health, this study sought to assess how carbohydrate intakes below the acceptable macronutrient distribution ranges relate to (a) food and micronutrient intake and (b) sociodemographic and health-related factors. METHODS The 1990 Ontario Health Survey (14) was population based, targeting subjects in private dwellings using a two-stage stratified cluster design to sample households. Detailed descriptions of the sampling methodology have been published elsewhere (14). The response rate for the first stage, during which information was gathered from one knowledgeable household member, was 87.5%. The response rates for the second stage were 75% and 68% for 12- to 15- and 16- to 24-year-old males, respectively, and 80% and 76% for 12- to 15- and 16- to 24-year-old females, respectively. Among adolescent respondents, 83% of males and 88% of females completed the food frequency questionnaire (FFQ). Respondents with incomplete FFQs were eliminated and were more likely (P⬍.05) to be male,
Journal of the AMERICAN DIETETIC ASSOCIATION
1783.e1
Table 1. Characteristics of the 1990 Ontario Health Survey population Characteristic Sociodemographic and personal Sex Male Female Age group 12–15 16–18 Household income statusb Low income ⬍$50,000 But not low income $50,000 (above average) Family type Two-parent household One-parent household Other Birthplace Canada Other Adiposityc At risk of underweight or underweight Normal weight At risk of overweight or overweight Health behavior Smoking “at the present time”d No Yes Alcohol use (monthly over the past year)e No Yes Physical Activity Indexf Active Moderate Inactive Health Good to excellent Fair to poor
Unweighteda (n)
Weighted (N)
% of population
2,459 2,735
411,800 411,800
50.0 50.0
3,008 2,186
436,900 386,700
53.1 47.0
559 1,614 2,331
77,700 226,200 369,600
11.1 32.3 56.6
4,380 609 205
685,300 97,700 40,600
83.2 11.9 4.9
4,861 312
729,400 90,900
88.9 11.1
276 3,496 870
45,800 557,500 138,600
6.2 75.1 18.7
4,322 775
684,600 121,700
84.9 15.1
3,669 1,287
582,700 200,900
74.4 25.6
2,322 1,052 1,495
357,500 170,100 245,400
46.3 22.0 31.8
4,902 254
784,100 34,700
95.8 4.2
a The frequency of missing data (either not applicable or not stated) was as follows: 0% for age, sex; 0% to ⬍2% for birthplace, weight change, smoking, self-rated health, family type; 5% to ⬍10% for alcohol use, Physical Activity Index, adiposity; 15% for household income status. b Household income status is a three-level variable derived by the Ontario Health Survey. The level “low income” is based on household income, family size, and urban/rural area of residence. c Based on body mass index and compared with age- and sex-specific norms (15). d Smoking refers to smoking “at the present time.” “No”⫽not at all, “Yes”⫽daily or occasional smoking. e Alcohol use refers to monthly drinking in the past year. “No”⫽none to less than once a month, “Yes”⫽once or twice a month or more. f Physical Activity Index is a three-level variable derived by the Ontario Health Survey. Average daily energy expenditure (kcal//kg/day): 3 kcal/kg/day⫽active, 1.5–2.9⫽moderate, and ⬍1.5⫽inactive.
to have household income less than $50,000, to not live in two-parent households, and to smoke. Thirteen pregnant females (0.2%) were removed from analysis. The Ontario Ministry of Health approved the study and gave access to the dataset. Sixteen respondents (0.3%) indicated that they consumed less than a half portion daily of each of the last three food groups of the FFQ (vegetables and fruit, meat and alternatives, and grain products), and their FFQs were deemed unusable. Data on household demographics, income, health behaviors, self-rated health, anthropometric measure-
1783.e2
November 2005 Volume 105 Number 11
ments, and dietary intake were collected. Low income was based on household income, family size, and urban/ rural area of residence (14). Adiposity was estimated using age- and sex-specific body mass index percentiles (15) and stratified into risk for underweight or underweight (⬍10th percentile), normal weight (10th percentile to ⬍85th percentile), and risk for overweight or overweight (⬎85th percentile). The 84-item, semiquantitative FFQ included all food groups and added fats and sweets. Nutrient analysis included protein, carbohydrate, fat, dietary fiber, vitamin A, vitamin C, niacin, riboflavin, thi-
Table 2. Energy, nutrient, and fiber intakes of Ontario adolescent males and females on low-carbohydrate-density (⬍45%) and non–lowcarbohydrate-density (ⱖ45%) diets Males
Females
Carbohydrate (% of Energy)
Carbohydrate (% of Energy)
Nutrients
<45%
Protein (g) Carbohydrate (g) Fat (g) Cholesterol (mg) Dietary fiber (g) Fiber (per 1,000 kcal) Vitamin A (REb) Vitamin C (mg) Niacin (NEc) Riboflavin (mg) Thiamin (mg) Iron (mg) Calcium (mg) Energy (kcal/kg) Energy (J)
4™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™ mean⫾SEM a ™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™ 3 130.1⫾2.08 114.0⫾1.01 98.8⫾1.44 86.7⫾0.78 318.1⫾5.50 410.0⫾4.02 238⫾3.83 312.4⫾2.94 151.3⫾2.7 127.4⫾1.4 109.7⫾1.84 92.2⫾0.92 538.8⫾12.0 441.0⫾5.68 400.1⫾9.36 309.6⫾3.64 17.3⫾0.44 23.1⫾0.36 13.4⫾0.29 18.1⫾0.26 5.4⫾0.08 7.3⫾0.93 5.7⫾0.07 7.51⫾0.08 1,598.6⫾33.7 1,642.7⫾22.2 1,383⫾29.2 1,457⫾20.95 131.8⫾3.8 243.5⫾4.1 114.4⫾2.83 214.7⫾3.33 51.2⫾0.90 45.6⫾0.47 38.5⫾0.62 34.3⫾0.33 2.9⫾0.05 2.9⫾0.03 2.3⫾0.04 2.3⫾0.02 2.0⫾0.03 2.3⫾0.03 1.5⫾0.02 1.7⫾0.01 16.1⫾0.29 17.8⫾0.20 12.1⫾0.20 13.5⫾0.15 1,557⫾25.3 1,611⫾16.26 1,253.5⫾20.80 1,281.3⫾13.81 51.5⫾1.09 52.3⫾0.62 43.7⫾0.87 45.0⫾0.48 3,107⫾51.52 3,170⫾30.76 2,301⫾35.86 2,372⫾21.45
>45%
<45%
>45%
a
SEM⫽standard error of the mean. RE⫽retinol equivalents. NE⫽niacin equivalents.
b c
amin, iron, calcium, and cholesterol. This FFQ, validated for use with those 19 years of age and older (16), was considered reasonable to use among teenagers because it contained a wide variety of foods, including those considered typical for them. The study also focused on the relationship between carbohydrate density and dietary components and not on absolute intakes. Macronutrient density refers to the proportion of energy contributed by a macronutrient. A macronutrient with high or low density would be one that exceeded or fell short of, respectively, its acceptable macronutrient distribution range upper or lower limit. The acceptable macronutrient distribution range for carbohydrate (45% to 65% of energy) was used to assess adolescents’ diets according to their carbohydrate density. A low-carbohydrate-density diet was considered one with ⬍45% of carbohydrate contributing to energy. A non–low-carbohydrate-density diet was one with carbohydrate density ⱖ45%. Acceptable macronutrient distribution ranges are based on the maintenance of energy balance through adequate energy intake and physical activity (13). Because a complex sampling design was used, survey weights were used to calculate point estimates. Design effects calculated specifically for the full Ontario sample were used to adjust variance estimates (17). Study sample characteristics were described using frequencies and proportions. 2 was used to compare food group intakes among low-carbohydrate-density and non– low-carbohydrate-density groups. To compare low-carbohydrate-density and non–low-carbohydrate-density status with personal and health-related variables, univariate logistic regression was used to determine odds ratios and 95% confidence intervals. Statistical analyses were carried out using SAS (version 8.4, 2001; SAS Inc, Cary, NC).
RESULTS The study population was evenly divided between sexes and those 12 to 15 and 16 to 18 years of age. Income was classified as low for 11.1% of the study population. Most subjects lived in two-parent families (83.2%), were Canadian-born (88.9%), were not recent smokers (84.9%) or alcohol consumers (74.4%), were at least moderately active (68.5%), and rated their health status positively (95.8%) (Table 1). Mean carbohydrate density was low (⬍45%) for 27.6% of males and 24.1% of females. These individuals formed the low-carbohydrate-density group, and virtually all (99%) also displayed fat intakes, as percentage of energy, above the acceptable macronutrient distribution range upper limit of 35% (a low-carbohydrate-density/high-fatdensity diet). Carbohydrate density (low vs nonlow) was strongly related to consumption of (a) several food groups at recommended levels according to Canada’s Food Guide to Healthy Eating (18) and (b) sweets, mixed dishes, and added fat based on median servings (Table 4, available online at www.adajournal.org). The proportions with at least five servings of vegetables and fruit (P⬍.001) and three servings of sweets (P⬍.001) were much lower among those with low-carbohydrate-density diets. Consumption of at least two servings of meat and alternatives (P⬍.001), one serving of mixed dishes for females (P⬍.001), and two servings of added fat (P⬍.05 for females, P⬍.001 for males) were higher among those with low-carbohydrate-density diets. Those consuming low-carbohydrate-density diets displayed markedly lower intakes of dietary fiber and vitamin C, whereas intakes of protein, fat, and cholesterol
November 2005 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1783.e3
Table 3. Characteristics associated with a low-carbohydrate-density diet in Ontario adolescents Carbohydrate (% of Energy) Characteristics (weighted %)a Sociodemographic and personal factors Sex Female Male Age group, y 12–15 16–18 Birthplace Other than Canada Canada Household income status Low income ⬍$50,000 But not low income ⱖ$50,000 (above average) Family type Two-parent household One-parent household Other Adiposity At risk of underweight or underweight Normal weight At risk of overweight or overweight Health Behaviors Alcohol use (monthly) No Yes Smoking (“at the present time”) No Yes Physical activity Active Moderately active Inactive Health Self-reported health Good to excellent Fair to poor
<45%
>45%
OR (95% CI)b
46.6 53.4
51.2 48.8
1.00 1.20 (1.00-1.44)
50.5 49.5
53.9 46.1
1.00 1.15 (0.95-1.38)
7.3 92.7
12.4 87.6
1.00 1.78 (1.27-2.50)***
10.9 29.5 59.6
11.2 33.3 55.6
1.00 0.86 (0.69-1.05) 1.16 (0.96-1.39)
84.8 10.6 4.6
82.7 12.3 5.0
1.00 0.84 (0.63-1.13) 0.92 (0.59-1.41)
6.5 72.0 21.4
6.1 76.2 17.7
1.00 0.80 (0.66-0.98)* 1.27 (1.02-1.57)*
68.8 31.2
76.2 23.8
1.00 1.46 (1.21-1.75)***
80.6 19.4
86.4 13.6
1.00 1.53 (1.23-1.90)***
47.1 20.9 32.0
46.0 22.4 31.6
1.00 0.92 (0.73-1.16) 1.02 (0.83-1.25)
94.5 5.5
96.2 3.8
1.00 1.47 (1.01-2.14)*
a
Data are given as percentage of the carbohydrate group having the characteristic. Odds ratio (OR) refers to the odds of having the characteristic in those with a low-carbohydrate-density diet compared with those with a non–low-carbohydrate-density diet. A 95% confidence interval (CI) that includes 1.00 indicates that there is no evidence of an association between the characteristic and low-carbohydrate-density (⬍45%) or non–lowcarbohydrate-density (ⱖ45%) diets. *Significantly different from reference category by univariate logistic regression, P⬍.05. ***Significantly different from reference category by univariate logistic regression, P⬍.001. b
were higher, compared with those with non–low-carbohydrate-density diets (Table 2). Energy intakes were similar in both groups among males and females, respectively. Personal and health-related characteristics and their relationship with carbohydrate density are presented in Table 3. Being Canadian-born, overweight status, alcohol use, smoking, and poorer health were associated with low-carbohydrate-density diets. Being of normal weight was related inversely to low-carbohydrate-density diets.
1783.e4
November 2005 Volume 105 Number 11
Age group, income, family type, and activity level were unrelated to low-carbohydrate-density diets (all P⬎.05). DISCUSSION This study suggests that the diets of adolescents with low-carbohydrate-density are distinguished by a reduced consumption of fruit and vegetables and an increased intake of meat and alternatives, mixed dishes, and added fats. The adverse effects of these food choices were rela-
Table 4. Proportion of adolescent males and females who met food group recommendations according to percentage of energy as carbohydrate ⬍45% or ⱖ45% Percentage Who Met Food Group Recommendationsb
Food groups (weighted %)
Recommended number of servingsa
Grain products (ⱖ5 servings) Vegetables and fruit (ⱖ5 servings) Meat and alternatives (ⱖ2 servings) Milk products (ⱖ3 servings, ⱕ16 y) (ⱖ2 servings, ⱖ17 y) Mixed dishesc (ⱖ1 servings)d Desserts/sugar (ⱖ3 servings)d Added fats (ⱖ2 servings)d
5-12 5-10 2-3 3-4 (12-16 y) 2-4 (17-18 y) ... ... ...
Males
Females
Carbohydrate Intake (Percent of Energy)
Carbohydrate Intake (Percent of Energy)
<45%
>45%
<45%
>45%
33.3 35.8*** 84.1*** 72.3
34.8 60.0 62.2 69.8
21.5 22.5*** 62.3*** 52.9
19.9 53.8 35.3 51.7
49.8 49.5*** 59.4***
43.7 73 45.1
33.9*** 33.1*** 47.9*
23.4 52.7 41.3
a
Based on Canada’s Food Guide for Healthy Eating (18). Data are given as percentage of carbohydrate intake group having an intake of food groups at or above recommended levels or median number of servings. c Dishes containing more than one food group (eg, pizza, burgers with buns, spaghetti with tomato and meat sauce). d ⱖMedian number of servings. *Based on 2 testing, statistically significant at P⬍.05. ***Based on 2 testing, statistically significant at P⬍.001. b
tively lower intakes of dietary fiber and vitamin C and higher intakes of cholesterol and fat. These results are consistent with those previously reported for youth in the United States (19). In addition to problems associated with insufficiency of nutrients, there is concern that higherfat diets in adolescents may foster an increased risk of obesity and coronary heart disease development (13). Knowledge of the characteristics of those with lowcarbohydrate-density diets can help in the development of effective interventions. Being foreign-born was strongly related to non–low-carbohydrate-density intake. Ethnicity can influence taste and home availability of fruit and vegetables, both of which are highly correlated with their consumption (20). White race has previously been related to adolescents’ poorer consumption of sufficient fruit and vegetables (21). A low-carbohydrate-density diet was more likely to be associated with risk of overweight and poorer health behaviors. The greater likelihood of overweight vs underweight status among the low-carbohydrate-density diet group may be related to the underweight adopting weight-control behaviors that include a greater intake of fruit and vegetables (22). Smoking and alcohol use have been associated with adolescents’ inadequate fruit and vegetable intake (23,24). In this study, the association of low-carbohydrate-density diets with greater smoking and alcohol use is moderate but is consistent with earlier findings that health-compromising behaviors of adolescents tend to cluster (2528). A univariate relationship between the low-carbohydrate-density diet and poorer self-rated health was evident, and it would be useful to explore this relationship in a multivariable model in future studies. Strengths of the 1990 Ontario Health Survey are that it includes a large province-wide study population; a considerable range of sociodemographic, behavioral, and health-related variables; and an extensive FFQ. Limita-
tions should be noted. Data in this study are cross-sectional, so it is not possible to draw causal links. The diets of FFQ responders may have differed from those of nonresponders. The FFQ was validated for use with adults and caution should be used interpreting the data of adolescents. Underreporting or overreporting of nutrients are assumed to have occurred in proportion to under- or overreporting of energy. Therefore, macronutrients expressed as a percentage of energy would be expected to be accurate (13). Although under- or overreporting would affect absolute estimates of nutrients, this study focused on the nutrient intake differences between those with low-carbohydrate-density diets compared with others. CONCLUSIONS Approximately one quarter of Ontario youth in 1990 consumed low-carbohydrate-density diets; these diets were related to low fruit and vegetable and high meat and alternative consumption. These patterns were associated with higher total fat and cholesterol and lower vitamin C and fiber intakes. Knowledge of the nutritional consequences of consumption patterns is important in the development of strategies to promote adolescent nutrition. From a public health perspective, these strategies may be an early entrée into reducing adolescents’ risk of chronic disease. References 1. Must A, Jacques PF, Dallal GE, Bejama CJ, Dietz WH. Long-term morbidity and mortality of overweight adolescents. N Engl J Med. 1992;327:13501355. 2. Weaver CM. The growing years and prevention of osteoporosis later in life. Proc Nutr Soc. 2000;59:303306.
November 2005 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1783.e5
3. Liu RH. Health benefits of fruit and vegetables are from additive and synergistic combinations of phytochemicals. Am J Clin Nutr. 2003;78(suppl 3):S517S520. 4. Ness AR, Powles JW. Fruit and vegetables, and cardiovacular disease: A review. Int J Epidemiol. 1997; 26:1-13. 5. Van Duyn MAS, Pivonka E. Overview of the health benefits of fruit and vegetable consumption for the dietetics professional: Selected literature. J Am Diet Assoc. 2000;100:1511-1521. 6. Krauss RM, Eckel RH, Howard B, Appel LJ, Daniels SR, Deckelbaum RJ, Erdman JW Jr, Kris-Etherton P, Goldberg IJ, Kotchen TA, Lichtenstein AH, Mitch WE, Mullis R, Robinson K, Wylie-Rosett J, St Jeor S, Suttie J, Tribble DL, Bazzarre TL. AHA dietary guidelines. revision 2000: A statement for healthcare professionals from the Nutrition Committee of the American Heart Association. Circulation. 2000;102: 2284-2299. 7. Gilliland FD, Berhane KT, Gauderman WJ, McConnell R, Peters J. Children’s lung function and antioxidant vitamin, fruit, juice and vegetable intake. Am J Epidemiol. 2003;158:576-584. 8. Lien N, Lytle LA, Klepp K-I. Stability in consumption of fruits, vegetables and sugary foods in a cohort from age 14 to age 21. Prev Med. 2001;33:217-226. 9. King AJC, Boyce WF, King MA. Healthy eating, dieting and dental hygiene. Trends in the health of Canadian youth. Health Canada. 1999;65–73. 10. Starkey LJ, Johnson-Down L, Gray-Donald K. Food habits of Canadians: Comparison of intakes in adults and adolescents to Canada’s Food Guide to Healthy Eating. Can J Diet Prac Res. 2001;62:61-69. 11. Kelder SH, Perry CL, Klepp K-I, Lytle LL. Longitudinal tracking of adolescent smoking, physical activity, and food choice behaviors. Am J Public Health. 1994;84:1121-1126. 12. Institute of Medicine. Dietary Reference Intakes: Applications in Dietary Assessment. Washington, DC: National Academy Press; 2001. 13. Standing Committee on the Scientific Evaluation of Dietary Reference Intakes. Dietary Reference intakes for Energy, Carbohydrates, Fiber, Fat, Protein and Amino Acids (Macronutrients). Available at: http://books.nap.edu/books/0309085373/html/1.html. 2002. Accessed June 23, 2003. 14. Ontario Ministry of Health. Ontario Health Survey 1990: User’s Guide. Vol. 1. Documentation. Toronto: Ontario Ministry of Health; 1992. 15. Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegel KM, Guo SS, Wei R, Mei Z, Curtin LR, Roche AF, Johnson CL. CDC Growth Charts: United States (Revised). US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics; 2000. 16. Bright-See E, Catlin G, Godin G. Assessment of the
1783.e6
November 2005 Volume 105 Number 11
17. 18. 19.
20.
21.
22. 23. 24.
25.
26.
27.
28.
relative validity of the Ontario Health Survey food frequency questionnaire. J Can Diet Assoc. 1994;55: 33-38. Ontario Ministry of Health. Ontario Health Survey 1990: User’s Guide. Vol. 2. Microdata Manual. Toronto, Canada: Ontario Ministry of Health; 1992. Health Canada. Canada’s Food Guide to Healthy Eating. Minister of Public Works and Government Services Canada; 1997. Beltsville Human Nutrition Research Centre. Continuing Survey of Food Intakes by Individuals (CFII) and 1994-96 Diet Health Knowledge Survey. US Department of Agriculture; 1998. Cited by: Standing Committee on the Scientific Evaluation of Dietary Reference Intakes. Dietary Reference intakes for Energy, Carbohydrates, Fiber, Fat, Protein and Amino Acids (Macronutrients). Tables K-3 and K-6. Available at: http://books.nap.edu/books/0309085373/ html/1.html. 2002. Accessed June 23, 2003. Neumark-Sztainer D, Wall M, Perry C, Story M. Correlates of fruit and vegetable intake among adolescents. Findings from Project EAT. Prev Med. 2003; 37:198-208. Neumark-Sztainer D, Story M, Hannan PJ, Croll J. Overweight status and eating patterns among adolescents: Where do youths stand in comparison with the Healthy People 2010 Objectives? Am J Public Health. 2002;92:844-851. Pesa JA, Turner LW. Fruit and vegetable intake and weight-control behaviors among U.S. youth. Am J Health Behav. 2001;25:3-9. Wilson DB, Nietert PJ. Patterns of fruit, vegetable and milk consumption among smoking and nonsmoking female teens. Am J Prev Med. 2002;22:240-246. Neumark-Sztainer D, Story M, Resnick MD, Blum RW. Correlates of inadequate fruit and vegetable consumption among adolescents. Prev Med. 1996;25: 497-505. Palaniappan U, Starkey LJ, O’Loughlin J, GrayDonald K. Fruit and vegetable consumption is lower and saturated fat intake is higher among Canadians reporting smoking. J Nutr. 2001;131:1952-1958. Neumark-Sztainer D, Story M, Toporoff E, Hines JH, Resnick MD, Blum RW. Covariations of eating behaviors with other health-related behaviors among adolescents. J Adolesc Health. 1997;6:450-458. Burke V, Milligan B, Beilin LJ, Dunbar D, Spencer M, Balde E, Gracey MP. Clustering of health-related behaviors among 18-year-old Australians. Prev Med. 1997;26:724-733. Boyle MH, Szatmari P, Offord DR, Merikangus K. Substance Abuse Among Adolescents and Young Adults: Prevalence, Socio-Demographic Correlates, Associated Problems and Familial Aggregation. Ontario Health Survey 1990, Ontario, Ministry of Health:1993:1-30. Working Paper No.2.
Adolescents’ Low-Carbohydrate-Density Diets Are Related to Poorer Dietary Intakes LINDA S. GREENE-FINESTONE, PhD, RD; M. KAREN CAMPBELL, PhD; SUSAN E. EVERS, PhD, RD; IRIS A. GUTMANIS, PhD
Web site exclusive! Editor’s note: Table 4 that accompanies this article is available online at www.adajournal.org. ABSTRACT This study was undertaken to assess how low-carbohydrate-density diets below the acceptable macronutrient distribution range relate to food and micronutrient intake and sociodemographic and health-related characteristics. The multistage stratified cluster design in the 1990 Ontario Health Survey was used. There were 5,194 subjects, 12 to 18 years of age, in sampled households. Dietary data were collected via a food frequency questionnaire. Lowcarbohydrate-density diets were consumed by 27.6% of males and 24.1% of females. Low-carbohydrate-density diets were related (P⬍.05) to reduced sufficiency of vegetables and fruit and higher consumption of meat and alternatives and added fats. The low-carbohydrate-density diet resulted in intakes lower in vitamin C and fiber and higher in cholesterol and total fat. The low-carbohydrate-density diet was directly associated with being Canadian-born (odds ratio [OR]⫽1.78, 95% confidence interval [CI]⫽1.27 to 2.50), overweight status (OR⫽1.27, 95%
L. S. Greene-Finestone is nutrition advisor in the Office of Nutrition Policy and Promotion, Health Canada, Ottawa, Ontario, Canada, and was previously enrolled in a doctoral program in the Department of Epidemiology and Biostatistics, University of Western Ontario, London, Ontario, Canada. M. K. Campbell is chair of the Department of Epidemiology and Biostatistics, University of Western Ontario, London, Ontario, Canada. S. E. Evers is a professor in the Department of Family Relations and Applied Nutrition, University of Guelph, MacDonald Institute, Guelph, Ontario, Canada. I. A. Gutmanis is director, Evaluation and Research, Specialized Geriatric Services, St Joseph’s Health Care, Parkwood Hospital, London, Ontario, Canada; at the time of the study, she was director of the Southwest Region Health Information Partnership, London, Ontario, Canada. Address correspondence to: Linda S. Greene-Finestone, PhD, RD, Office of Nutrition Policy and Promotion, Health Canada, Qualicum Tower A, 2936 Baseline Rd, 3rd Floor, Ottawa, Ontario, Canada, K1A 0K9 A.L. 3303D. E-mail: [email protected] Copyright © 2005 by the American Dietetic Association. 0002-8223/05/10511-0008$30.00/0 doi: 10.1016/j.jada.2005.08.014
© 2005 by the American Dietetic Association
CI⫽1.02 to 1.57), smoking (OR⫽1.53, 95% CI⫽1.23 to 1.90), alcohol use (OR⫽1.46, 95% CI⫽1.21 to 1.75), and poorer self-rated health (OR⫽1.47, 95% CI⫽1.01 to 2.14). Use of the acceptable macronutrient distribution range identified adolescents with low-carbohydrate-density diets whose food choices and nutrient intake may impact negatively on short- and long-term health. J Am Diet Assoc. 2005;105:1783.e1-1783.e6.
T
he eating behaviors of adolescents affect their current health and may impact on adult-onset chronic disease (1,2). Fruit, vegetables, and grains are particularly implicated with decreased risk of diseases such as cancer, cardiovascular disease, and stroke (3-7). Teenagers’ poor consumption of fruit and vegetables (8-10) is of concern because eating behaviors tend to consolidate by adolescence and follow a similar pattern into young adulthood (8,11). The most recent nutrition recommendations of the United States and Canada are the Dietary Reference Intakes (12). Of the Dietary Reference Intakes, the acceptable macronutrient distribution range recommendations can be a general diet evaluation guide (13). The acceptable macronutrient distribution ranges were estimated to account for the roles of macronutrients in chronic disease development and in achievement of micronutrient sufficiency. Given the long-term consequences of eating patterns formed during adolescence and the roles of acceptable macronutrient distribution ranges in promoting health, this study sought to assess how carbohydrate intakes below the acceptable macronutrient distribution ranges relate to (a) food and micronutrient intake and (b) sociodemographic and health-related factors. METHODS The 1990 Ontario Health Survey (14) was population based, targeting subjects in private dwellings using a two-stage stratified cluster design to sample households. Detailed descriptions of the sampling methodology have been published elsewhere (14). The response rate for the first stage, during which information was gathered from one knowledgeable household member, was 87.5%. The response rates for the second stage were 75% and 68% for 12- to 15- and 16- to 24-year-old males, respectively, and 80% and 76% for 12- to 15- and 16- to 24-year-old females, respectively. Among adolescent respondents, 83% of males and 88% of females completed the food frequency questionnaire (FFQ). Respondents with incomplete FFQs were eliminated and were more likely (P⬍.05) to be male,
Journal of the AMERICAN DIETETIC ASSOCIATION
1783.e1
Table 1. Characteristics of the 1990 Ontario Health Survey population Characteristic Sociodemographic and personal Sex Male Female Age group 12–15 16–18 Household income statusb Low income ⬍$50,000 But not low income $50,000 (above average) Family type Two-parent household One-parent household Other Birthplace Canada Other Adiposityc At risk of underweight or underweight Normal weight At risk of overweight or overweight Health behavior Smoking “at the present time”d No Yes Alcohol use (monthly over the past year)e No Yes Physical Activity Indexf Active Moderate Inactive Health Good to excellent Fair to poor
Unweighteda (n)
Weighted (N)
% of population
2,459 2,735
411,800 411,800
50.0 50.0
3,008 2,186
436,900 386,700
53.1 47.0
559 1,614 2,331
77,700 226,200 369,600
11.1 32.3 56.6
4,380 609 205
685,300 97,700 40,600
83.2 11.9 4.9
4,861 312
729,400 90,900
88.9 11.1
276 3,496 870
45,800 557,500 138,600
6.2 75.1 18.7
4,322 775
684,600 121,700
84.9 15.1
3,669 1,287
582,700 200,900
74.4 25.6
2,322 1,052 1,495
357,500 170,100 245,400
46.3 22.0 31.8
4,902 254
784,100 34,700
95.8 4.2
a The frequency of missing data (either not applicable or not stated) was as follows: 0% for age, sex; 0% to ⬍2% for birthplace, weight change, smoking, self-rated health, family type; 5% to ⬍10% for alcohol use, Physical Activity Index, adiposity; 15% for household income status. b Household income status is a three-level variable derived by the Ontario Health Survey. The level “low income” is based on household income, family size, and urban/rural area of residence. c Based on body mass index and compared with age- and sex-specific norms (15). d Smoking refers to smoking “at the present time.” “No”⫽not at all, “Yes”⫽daily or occasional smoking. e Alcohol use refers to monthly drinking in the past year. “No”⫽none to less than once a month, “Yes”⫽once or twice a month or more. f Physical Activity Index is a three-level variable derived by the Ontario Health Survey. Average daily energy expenditure (kcal//kg/day): 3 kcal/kg/day⫽active, 1.5–2.9⫽moderate, and ⬍1.5⫽inactive.
to have household income less than $50,000, to not live in two-parent households, and to smoke. Thirteen pregnant females (0.2%) were removed from analysis. The Ontario Ministry of Health approved the study and gave access to the dataset. Sixteen respondents (0.3%) indicated that they consumed less than a half portion daily of each of the last three food groups of the FFQ (vegetables and fruit, meat and alternatives, and grain products), and their FFQs were deemed unusable. Data on household demographics, income, health behaviors, self-rated health, anthropometric measure-
1783.e2
November 2005 Volume 105 Number 11
ments, and dietary intake were collected. Low income was based on household income, family size, and urban/ rural area of residence (14). Adiposity was estimated using age- and sex-specific body mass index percentiles (15) and stratified into risk for underweight or underweight (⬍10th percentile), normal weight (10th percentile to ⬍85th percentile), and risk for overweight or overweight (⬎85th percentile). The 84-item, semiquantitative FFQ included all food groups and added fats and sweets. Nutrient analysis included protein, carbohydrate, fat, dietary fiber, vitamin A, vitamin C, niacin, riboflavin, thi-
Table 2. Energy, nutrient, and fiber intakes of Ontario adolescent males and females on low-carbohydrate-density (⬍45%) and non–lowcarbohydrate-density (ⱖ45%) diets Males
Females
Carbohydrate (% of Energy)
Carbohydrate (% of Energy)
Nutrients
<45%
Protein (g) Carbohydrate (g) Fat (g) Cholesterol (mg) Dietary fiber (g) Fiber (per 1,000 kcal) Vitamin A (REb) Vitamin C (mg) Niacin (NEc) Riboflavin (mg) Thiamin (mg) Iron (mg) Calcium (mg) Energy (kcal/kg) Energy (J)
4™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™ mean⫾SEM a ™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™ 3 130.1⫾2.08 114.0⫾1.01 98.8⫾1.44 86.7⫾0.78 318.1⫾5.50 410.0⫾4.02 238⫾3.83 312.4⫾2.94 151.3⫾2.7 127.4⫾1.4 109.7⫾1.84 92.2⫾0.92 538.8⫾12.0 441.0⫾5.68 400.1⫾9.36 309.6⫾3.64 17.3⫾0.44 23.1⫾0.36 13.4⫾0.29 18.1⫾0.26 5.4⫾0.08 7.3⫾0.93 5.7⫾0.07 7.51⫾0.08 1,598.6⫾33.7 1,642.7⫾22.2 1,383⫾29.2 1,457⫾20.95 131.8⫾3.8 243.5⫾4.1 114.4⫾2.83 214.7⫾3.33 51.2⫾0.90 45.6⫾0.47 38.5⫾0.62 34.3⫾0.33 2.9⫾0.05 2.9⫾0.03 2.3⫾0.04 2.3⫾0.02 2.0⫾0.03 2.3⫾0.03 1.5⫾0.02 1.7⫾0.01 16.1⫾0.29 17.8⫾0.20 12.1⫾0.20 13.5⫾0.15 1,557⫾25.3 1,611⫾16.26 1,253.5⫾20.80 1,281.3⫾13.81 51.5⫾1.09 52.3⫾0.62 43.7⫾0.87 45.0⫾0.48 3,107⫾51.52 3,170⫾30.76 2,301⫾35.86 2,372⫾21.45
>45%
<45%
>45%
a
SEM⫽standard error of the mean. RE⫽retinol equivalents. NE⫽niacin equivalents.
b c
amin, iron, calcium, and cholesterol. This FFQ, validated for use with those 19 years of age and older (16), was considered reasonable to use among teenagers because it contained a wide variety of foods, including those considered typical for them. The study also focused on the relationship between carbohydrate density and dietary components and not on absolute intakes. Macronutrient density refers to the proportion of energy contributed by a macronutrient. A macronutrient with high or low density would be one that exceeded or fell short of, respectively, its acceptable macronutrient distribution range upper or lower limit. The acceptable macronutrient distribution range for carbohydrate (45% to 65% of energy) was used to assess adolescents’ diets according to their carbohydrate density. A low-carbohydrate-density diet was considered one with ⬍45% of carbohydrate contributing to energy. A non–low-carbohydrate-density diet was one with carbohydrate density ⱖ45%. Acceptable macronutrient distribution ranges are based on the maintenance of energy balance through adequate energy intake and physical activity (13). Because a complex sampling design was used, survey weights were used to calculate point estimates. Design effects calculated specifically for the full Ontario sample were used to adjust variance estimates (17). Study sample characteristics were described using frequencies and proportions. 2 was used to compare food group intakes among low-carbohydrate-density and non– low-carbohydrate-density groups. To compare low-carbohydrate-density and non–low-carbohydrate-density status with personal and health-related variables, univariate logistic regression was used to determine odds ratios and 95% confidence intervals. Statistical analyses were carried out using SAS (version 8.4, 2001; SAS Inc, Cary, NC).
RESULTS The study population was evenly divided between sexes and those 12 to 15 and 16 to 18 years of age. Income was classified as low for 11.1% of the study population. Most subjects lived in two-parent families (83.2%), were Canadian-born (88.9%), were not recent smokers (84.9%) or alcohol consumers (74.4%), were at least moderately active (68.5%), and rated their health status positively (95.8%) (Table 1). Mean carbohydrate density was low (⬍45%) for 27.6% of males and 24.1% of females. These individuals formed the low-carbohydrate-density group, and virtually all (99%) also displayed fat intakes, as percentage of energy, above the acceptable macronutrient distribution range upper limit of 35% (a low-carbohydrate-density/high-fatdensity diet). Carbohydrate density (low vs nonlow) was strongly related to consumption of (a) several food groups at recommended levels according to Canada’s Food Guide to Healthy Eating (18) and (b) sweets, mixed dishes, and added fat based on median servings (Table 4, available online at www.adajournal.org). The proportions with at least five servings of vegetables and fruit (P⬍.001) and three servings of sweets (P⬍.001) were much lower among those with low-carbohydrate-density diets. Consumption of at least two servings of meat and alternatives (P⬍.001), one serving of mixed dishes for females (P⬍.001), and two servings of added fat (P⬍.05 for females, P⬍.001 for males) were higher among those with low-carbohydrate-density diets. Those consuming low-carbohydrate-density diets displayed markedly lower intakes of dietary fiber and vitamin C, whereas intakes of protein, fat, and cholesterol
November 2005 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1783.e3
Table 3. Characteristics associated with a low-carbohydrate-density diet in Ontario adolescents Carbohydrate (% of Energy) Characteristics (weighted %)a Sociodemographic and personal factors Sex Female Male Age group, y 12–15 16–18 Birthplace Other than Canada Canada Household income status Low income ⬍$50,000 But not low income ⱖ$50,000 (above average) Family type Two-parent household One-parent household Other Adiposity At risk of underweight or underweight Normal weight At risk of overweight or overweight Health Behaviors Alcohol use (monthly) No Yes Smoking (“at the present time”) No Yes Physical activity Active Moderately active Inactive Health Self-reported health Good to excellent Fair to poor
<45%
>45%
OR (95% CI)b
46.6 53.4
51.2 48.8
1.00 1.20 (1.00-1.44)
50.5 49.5
53.9 46.1
1.00 1.15 (0.95-1.38)
7.3 92.7
12.4 87.6
1.00 1.78 (1.27-2.50)***
10.9 29.5 59.6
11.2 33.3 55.6
1.00 0.86 (0.69-1.05) 1.16 (0.96-1.39)
84.8 10.6 4.6
82.7 12.3 5.0
1.00 0.84 (0.63-1.13) 0.92 (0.59-1.41)
6.5 72.0 21.4
6.1 76.2 17.7
1.00 0.80 (0.66-0.98)* 1.27 (1.02-1.57)*
68.8 31.2
76.2 23.8
1.00 1.46 (1.21-1.75)***
80.6 19.4
86.4 13.6
1.00 1.53 (1.23-1.90)***
47.1 20.9 32.0
46.0 22.4 31.6
1.00 0.92 (0.73-1.16) 1.02 (0.83-1.25)
94.5 5.5
96.2 3.8
1.00 1.47 (1.01-2.14)*
a
Data are given as percentage of the carbohydrate group having the characteristic. Odds ratio (OR) refers to the odds of having the characteristic in those with a low-carbohydrate-density diet compared with those with a non–low-carbohydrate-density diet. A 95% confidence interval (CI) that includes 1.00 indicates that there is no evidence of an association between the characteristic and low-carbohydrate-density (⬍45%) or non–lowcarbohydrate-density (ⱖ45%) diets. *Significantly different from reference category by univariate logistic regression, P⬍.05. ***Significantly different from reference category by univariate logistic regression, P⬍.001. b
were higher, compared with those with non–low-carbohydrate-density diets (Table 2). Energy intakes were similar in both groups among males and females, respectively. Personal and health-related characteristics and their relationship with carbohydrate density are presented in Table 3. Being Canadian-born, overweight status, alcohol use, smoking, and poorer health were associated with low-carbohydrate-density diets. Being of normal weight was related inversely to low-carbohydrate-density diets.
1783.e4
November 2005 Volume 105 Number 11
Age group, income, family type, and activity level were unrelated to low-carbohydrate-density diets (all P⬎.05). DISCUSSION This study suggests that the diets of adolescents with low-carbohydrate-density are distinguished by a reduced consumption of fruit and vegetables and an increased intake of meat and alternatives, mixed dishes, and added fats. The adverse effects of these food choices were rela-
Table 4. Proportion of adolescent males and females who met food group recommendations according to percentage of energy as carbohydrate ⬍45% or ⱖ45% Percentage Who Met Food Group Recommendationsb
Food groups (weighted %)
Recommended number of servingsa
Grain products (ⱖ5 servings) Vegetables and fruit (ⱖ5 servings) Meat and alternatives (ⱖ2 servings) Milk products (ⱖ3 servings, ⱕ16 y) (ⱖ2 servings, ⱖ17 y) Mixed dishesc (ⱖ1 servings)d Desserts/sugar (ⱖ3 servings)d Added fats (ⱖ2 servings)d
5-12 5-10 2-3 3-4 (12-16 y) 2-4 (17-18 y) ... ... ...
Males
Females
Carbohydrate Intake (Percent of Energy)
Carbohydrate Intake (Percent of Energy)
<45%
>45%
<45%
>45%
33.3 35.8*** 84.1*** 72.3
34.8 60.0 62.2 69.8
21.5 22.5*** 62.3*** 52.9
19.9 53.8 35.3 51.7
49.8 49.5*** 59.4***
43.7 73 45.1
33.9*** 33.1*** 47.9*
23.4 52.7 41.3
a
Based on Canada’s Food Guide for Healthy Eating (18). Data are given as percentage of carbohydrate intake group having an intake of food groups at or above recommended levels or median number of servings. c Dishes containing more than one food group (eg, pizza, burgers with buns, spaghetti with tomato and meat sauce). d ⱖMedian number of servings. *Based on 2 testing, statistically significant at P⬍.05. ***Based on 2 testing, statistically significant at P⬍.001. b
tively lower intakes of dietary fiber and vitamin C and higher intakes of cholesterol and fat. These results are consistent with those previously reported for youth in the United States (19). In addition to problems associated with insufficiency of nutrients, there is concern that higherfat diets in adolescents may foster an increased risk of obesity and coronary heart disease development (13). Knowledge of the characteristics of those with lowcarbohydrate-density diets can help in the development of effective interventions. Being foreign-born was strongly related to non–low-carbohydrate-density intake. Ethnicity can influence taste and home availability of fruit and vegetables, both of which are highly correlated with their consumption (20). White race has previously been related to adolescents’ poorer consumption of sufficient fruit and vegetables (21). A low-carbohydrate-density diet was more likely to be associated with risk of overweight and poorer health behaviors. The greater likelihood of overweight vs underweight status among the low-carbohydrate-density diet group may be related to the underweight adopting weight-control behaviors that include a greater intake of fruit and vegetables (22). Smoking and alcohol use have been associated with adolescents’ inadequate fruit and vegetable intake (23,24). In this study, the association of low-carbohydrate-density diets with greater smoking and alcohol use is moderate but is consistent with earlier findings that health-compromising behaviors of adolescents tend to cluster (2528). A univariate relationship between the low-carbohydrate-density diet and poorer self-rated health was evident, and it would be useful to explore this relationship in a multivariable model in future studies. Strengths of the 1990 Ontario Health Survey are that it includes a large province-wide study population; a considerable range of sociodemographic, behavioral, and health-related variables; and an extensive FFQ. Limita-
tions should be noted. Data in this study are cross-sectional, so it is not possible to draw causal links. The diets of FFQ responders may have differed from those of nonresponders. The FFQ was validated for use with adults and caution should be used interpreting the data of adolescents. Underreporting or overreporting of nutrients are assumed to have occurred in proportion to under- or overreporting of energy. Therefore, macronutrients expressed as a percentage of energy would be expected to be accurate (13). Although under- or overreporting would affect absolute estimates of nutrients, this study focused on the nutrient intake differences between those with low-carbohydrate-density diets compared with others. CONCLUSIONS Approximately one quarter of Ontario youth in 1990 consumed low-carbohydrate-density diets; these diets were related to low fruit and vegetable and high meat and alternative consumption. These patterns were associated with higher total fat and cholesterol and lower vitamin C and fiber intakes. Knowledge of the nutritional consequences of consumption patterns is important in the development of strategies to promote adolescent nutrition. From a public health perspective, these strategies may be an early entrée into reducing adolescents’ risk of chronic disease. References 1. Must A, Jacques PF, Dallal GE, Bejama CJ, Dietz WH. Long-term morbidity and mortality of overweight adolescents. N Engl J Med. 1992;327:13501355. 2. Weaver CM. The growing years and prevention of osteoporosis later in life. Proc Nutr Soc. 2000;59:303306.
November 2005 ● Journal of the AMERICAN DIETETIC ASSOCIATION
1783.e5
3. Liu RH. Health benefits of fruit and vegetables are from additive and synergistic combinations of phytochemicals. Am J Clin Nutr. 2003;78(suppl 3):S517S520. 4. Ness AR, Powles JW. Fruit and vegetables, and cardiovacular disease: A review. Int J Epidemiol. 1997; 26:1-13. 5. Van Duyn MAS, Pivonka E. Overview of the health benefits of fruit and vegetable consumption for the dietetics professional: Selected literature. J Am Diet Assoc. 2000;100:1511-1521. 6. Krauss RM, Eckel RH, Howard B, Appel LJ, Daniels SR, Deckelbaum RJ, Erdman JW Jr, Kris-Etherton P, Goldberg IJ, Kotchen TA, Lichtenstein AH, Mitch WE, Mullis R, Robinson K, Wylie-Rosett J, St Jeor S, Suttie J, Tribble DL, Bazzarre TL. AHA dietary guidelines. revision 2000: A statement for healthcare professionals from the Nutrition Committee of the American Heart Association. Circulation. 2000;102: 2284-2299. 7. Gilliland FD, Berhane KT, Gauderman WJ, McConnell R, Peters J. Children’s lung function and antioxidant vitamin, fruit, juice and vegetable intake. Am J Epidemiol. 2003;158:576-584. 8. Lien N, Lytle LA, Klepp K-I. Stability in consumption of fruits, vegetables and sugary foods in a cohort from age 14 to age 21. Prev Med. 2001;33:217-226. 9. King AJC, Boyce WF, King MA. Healthy eating, dieting and dental hygiene. Trends in the health of Canadian youth. Health Canada. 1999;65–73. 10. Starkey LJ, Johnson-Down L, Gray-Donald K. Food habits of Canadians: Comparison of intakes in adults and adolescents to Canada’s Food Guide to Healthy Eating. Can J Diet Prac Res. 2001;62:61-69. 11. Kelder SH, Perry CL, Klepp K-I, Lytle LL. Longitudinal tracking of adolescent smoking, physical activity, and food choice behaviors. Am J Public Health. 1994;84:1121-1126. 12. Institute of Medicine. Dietary Reference Intakes: Applications in Dietary Assessment. Washington, DC: National Academy Press; 2001. 13. Standing Committee on the Scientific Evaluation of Dietary Reference Intakes. Dietary Reference intakes for Energy, Carbohydrates, Fiber, Fat, Protein and Amino Acids (Macronutrients). Available at: http://books.nap.edu/books/0309085373/html/1.html. 2002. Accessed June 23, 2003. 14. Ontario Ministry of Health. Ontario Health Survey 1990: User’s Guide. Vol. 1. Documentation. Toronto: Ontario Ministry of Health; 1992. 15. Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegel KM, Guo SS, Wei R, Mei Z, Curtin LR, Roche AF, Johnson CL. CDC Growth Charts: United States (Revised). US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics; 2000. 16. Bright-See E, Catlin G, Godin G. Assessment of the
1783.e6
November 2005 Volume 105 Number 11
17. 18. 19.
20.
21.
22. 23. 24.
25.
26.
27.
28.
relative validity of the Ontario Health Survey food frequency questionnaire. J Can Diet Assoc. 1994;55: 33-38. Ontario Ministry of Health. Ontario Health Survey 1990: User’s Guide. Vol. 2. Microdata Manual. Toronto, Canada: Ontario Ministry of Health; 1992. Health Canada. Canada’s Food Guide to Healthy Eating. Minister of Public Works and Government Services Canada; 1997. Beltsville Human Nutrition Research Centre. Continuing Survey of Food Intakes by Individuals (CFII) and 1994-96 Diet Health Knowledge Survey. US Department of Agriculture; 1998. Cited by: Standing Committee on the Scientific Evaluation of Dietary Reference Intakes. Dietary Reference intakes for Energy, Carbohydrates, Fiber, Fat, Protein and Amino Acids (Macronutrients). Tables K-3 and K-6. Available at: http://books.nap.edu/books/0309085373/ html/1.html. 2002. Accessed June 23, 2003. Neumark-Sztainer D, Wall M, Perry C, Story M. Correlates of fruit and vegetable intake among adolescents. Findings from Project EAT. Prev Med. 2003; 37:198-208. Neumark-Sztainer D, Story M, Hannan PJ, Croll J. Overweight status and eating patterns among adolescents: Where do youths stand in comparison with the Healthy People 2010 Objectives? Am J Public Health. 2002;92:844-851. Pesa JA, Turner LW. Fruit and vegetable intake and weight-control behaviors among U.S. youth. Am J Health Behav. 2001;25:3-9. Wilson DB, Nietert PJ. Patterns of fruit, vegetable and milk consumption among smoking and nonsmoking female teens. Am J Prev Med. 2002;22:240-246. Neumark-Sztainer D, Story M, Resnick MD, Blum RW. Correlates of inadequate fruit and vegetable consumption among adolescents. Prev Med. 1996;25: 497-505. Palaniappan U, Starkey LJ, O’Loughlin J, GrayDonald K. Fruit and vegetable consumption is lower and saturated fat intake is higher among Canadians reporting smoking. J Nutr. 2001;131:1952-1958. Neumark-Sztainer D, Story M, Toporoff E, Hines JH, Resnick MD, Blum RW. Covariations of eating behaviors with other health-related behaviors among adolescents. J Adolesc Health. 1997;6:450-458. Burke V, Milligan B, Beilin LJ, Dunbar D, Spencer M, Balde E, Gracey MP. Clustering of health-related behaviors among 18-year-old Australians. Prev Med. 1997;26:724-733. Boyle MH, Szatmari P, Offord DR, Merikangus K. Substance Abuse Among Adolescents and Young Adults: Prevalence, Socio-Demographic Correlates, Associated Problems and Familial Aggregation. Ontario Health Survey 1990, Ontario, Ministry of Health:1993:1-30. Working Paper No.2.