Dietary intake among young adolescents in Ontario: associations with vegetarian status and attitude toward health

Preventive Medicine 40 (2005) 105 – 111 www.elsevier.com/locate/ypmed

Dietary intake among young adolescents in Ontario: associations with vegetarian status and attitude toward health Linda S. Greene-Finestone, Ph.D., R.D., a,* M. Karen Campbell, Ph.D., a Iris A. Gutmanis, Ph.D., a,b and Susan E. Evers, Ph.D., R.D. c a

Department of Epidemiology and Biostatistics, University of Western Ontario, London, Ontario, Canada b Southwest Region Health Information Partnership, London, Ontario, Canada c Department of Family Relations and Applied Nutrition, University of Guelph, Guelph, Ontario, Canada Available online 20 July 2004

Abstract Background. This study aimed to describe the dietary intakes of adolescent vegetarians and omnivores and determine if, and how, attitude toward personal health related to food consumption. Methods. Among grade 9 students in Ontario, Canada (n = 630), vegetarian status and estimated consumption of foods and food groups were determined by food frequency questionnaire. Personal health was self-categorized as very important (the ‘‘health conscious’’) or somewhat/not important (the ‘‘non-health conscious’’). Results. The prevalence of vegetarianism was 6.5% (CI = 4.6 – 8.4%) among females and 1.0% (CI = 0.2 – 1.8%) among males. Healthconscious omnivores consumed more grain, vegetables and fruit, and milk product than non-health-conscious omnivores ( P < 0.05). Healthconscious vegetarians ingested more grain products, vegetables and fruit, and meat and alternatives than non-health-conscious vegetarians ( P < 0.05). Among non-health-conscious vegetarians, none consumed two daily servings of meat and alternatives compared to 60.5% of nonhealth-conscious omnivores ( P < 0.001). Among health-conscious vegetarians, milk product consumption was lower than that of healthconscious omnivores ( P = 0.015). Conclusions. Large proportions of both vegetarian and omnivore adolescents consumed suboptimal diets. Health consciousness had value as an indicator of dietary adequacy and may be useful as a rudimentary screen for problematic dietary consumption patterns. D 2004 The Institute For Cancer Prevention and Elsevier Inc. All rights reserved. Keywords: Adolescent; Vegetarian; Omnivore; Prevalence; Food intake; Attitude to health

Introduction Interest in various vegetarian diets appears to be increasing in popularity [1]. Estimates of adolescent vegetarianism (including semivegetarian, lacto-ovo vegetarian, and vegan types) from population-based surveys outside of Canada have ranged from 2.4% to 6.5% for males. For female adolescents, vegetarian estimations range from 4.0% to up to 25.0% for those in Umea˚, Sweden [2 – 4]. Based on Canadian teenagers’ self-reported vegetarian status or red meat avoidance, estimates are 1.7 –2.4% for males and 3.3 –

* Corresponding author. Department of Physical Medicine and Rehabilitation, Elisabeth-Bruye`re Health Centre, 43 Bruye`re Street., Ottawa, Ontario, Canada KIN 5C8. Fax: +1-613-562-6312. E-mail address: [email protected] (L.S. Greene-Finestone).

8.8% for females [5,6], with upper levels associated with increased age. The diets of adolescents are of importance because they can have short- and long-term influence on physical and social well-being They are related to growth as well as the development of lifelong eating habits and health promotion [7]. Vegetarian diets, when appropriately planned, can be healthful choices for teenagers [1]. Studies [2– 4,8] comparing dietary intakes of adolescent vegetarians to those of omnivores have described positive food consumption patterns such as higher ingestion of fruits and vegetables, lower fast food intake, decreased energy intake from fat, as well as the promotion of good lifelong eating habits. However, potential negative aspects include concern over intakes of specific nutrients such as B12 as well as associations between vegetarianism and both healthy and disordered weight control behaviors [8– 10].

0091-7435/$ - see front matter D 2004 The Institute For Cancer Prevention and Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2004.05.009

106

L.S. Greene-Finestone et al. / Preventive Medicine 40 (2005) 105–111

Vegetarians and omnivores are not homogeneous groups. Whether teenage vegetarianism is a positive or negative state, from a nutritional perspective, may be related to their motive for becoming vegetarian. For adolescents, motivations for following a vegetarian diet include weight control, compassion for animals, and taste of meat, and these appear to be more popular than health-related reasons [2,9]. It has been observed that health-conscious college students have higher nutrient intakes [11]. Similarly, attitude toward personal health may influence the dietary intake of vegetarians. In Canada, there has been no systematic evaluation of the diets of vegetarian and omnivore adolescents. The objective of this study was to describe the dietary intakes of vegetarian and omnivore teens and determine if, and how, their attitude toward personal health related to their food consumption.

Methods The Opening Doors Student Self-Report Questionnaire [12] was approved by the Research Ethics Review Board of the Centre for Addiction and Mental Health, which is affiliated with the University of Toronto. This questionnaire was used to survey students in the Hamilton – Wentworth and Niagara regions of Ontario, Canada, in 1999. In high schools that agreed to take part, students in the entire ninth grade were eligible to participate, after parental consent. Surveys were removed from analysis if patterns of consumption were conspicuously unlikely or if data on all major issues of interest—body mass index (BMI), dieting, the food frequency questionnaire (FFQ), and vegetarian status—were missing. Food consumption was estimated using a brief 19-item food frequency questionnaire (FFQ), which included all food groups as well as sweets and high-fat snack foods. Its five frequency categories were ‘‘never or rarely’’, ‘‘monthly but not weekly’’, ‘‘weekly but not daily’’, ‘‘once a day’’, and ‘‘at least twice daily’’. Food item intakes from each group were summed. Items eaten weekly were counted as 0.5 of a daily serving (an average of 1– 6 times/7 days). Those eaten less than weekly were not considered to contribute to daily servings. Because the FFQ’s highest category of intake was ‘‘at least twice daily’’, the number of estimated daily servings within each food group was capped at two in order that the number of daily servings not be underestimated in those eating many servings from a single food item. Vegetarian status was assessed using reported consumption or exclusion of red meats, poultry, fish, eggs, and milk products. For the purpose of this study, the vegetarians (VEG) included those identified as vegan, lacto-ovo, and semivegetarian. Individuals classified as vegans were those whose diets excluded all foods of animal origin. Lacto-ovo vegetarians included eggs and dairy products and avoided red meat, poultry, and fish. Semivegetarians included eggs and dairy products, fish, and/or poultry, but avoided red

meat. Those consuming red meat at least monthly were considered omnivores (OMN). Sociodemographic and personal variables included sex, age, and attitude toward personal health. Participants were asked to rate how much they valued their health. The response ‘‘My health is very important’’ indicated a high value being placed on personal health and this group was called health conscious. The response ‘‘My health is somewhat important’’ or ‘‘My health is not important’’ indicated that personal health was not highly valued and this group was called non-health conscious. The dietary diversity score (DDS) was used to assess participants’ success in consuming foods from a variety of food groups. Each of the four food groups (fruits and vegetables, grain products, milk products, and meat and alternatives) contributed 1 point for a maximum possible DDS score of 4. The DDS has been used previously in U.S. studies [13,14] with fruits and vegetables grouped separately, reflecting the five food groups of the food guide pyramid [15]. Consumption of a minimum of one daily source of dietary fiber (non-starch polysaccharide) was also determined. Sources chosen were those generally containing at least 1 g of fiber per serving. Data were analyzed using SAS version 8.4 [16]. Frequency analyses were used to report VEG and OMN prevalence. Chi-square analyses, Fisher’s Exact tests, and t tests were used, where appropriate, to compare vegetarian status with frequency of consumption variables. Because the VEG group was predominantly composed of females, statistical analysis of females only was performed to eliminate possible effect modification of sex. Using the Mantel – Haenszel Method and Breslow – Day test for homogeneity of the odds ratios, the value of the health consciousness variable was explored as a possible effect modifier of the relationship between the frequency of food group consumption and vegetarian status. P values <0.05 are considered to be statistically significant.

Results The response rates of the schools ranged from 19% to 86% (19%, 22%, 24%, 48%, 62%, 66%, and 86%). Data from the three lowest response schools did not differ significantly on nutrition and health-related issues compared to the other schools and all schools were retained for analysis. Intraclass correlation measured the similarity of students within schools relative to that between schools [17]. The intraclass correlations for vegetarian status, selfreported health, and dieting in the past year ranged from 0.0004 to 0.0037, suggesting the effect of clustering within schools was negligible. In all, 4.0% (27/680) of surveys were excluded due to unreliable responses. Discarded surveys were more likely to come from those who reported alcohol, glue, and solvent use and had lower marks ( P < 0.05 for all). Food frequency

L.S. Greene-Finestone et al. / Preventive Medicine 40 (2005) 105–111 Table 1 Description of study participants Characteristics

Total sample, % (n = 630)a

Male, % (n = 286)

Female, % (n = 337)

Age (years) 13 14 15

12.7 (76)a 85.0 (509)a 2.3 (14)

9.5 (26) 87.6 (240) 2.9 (8)

15.2 (49) 82.9 (267) 1.9 (6)

Vegetarian status Lacto-ovo vegetarian Semivegetarian Omnivore

2.5 (16) 1.4 (9) 96.0 (605)a

0.4 (1) 0.7 (2) 98.9 (283)

4.5 (15) 2.1 (7) 93.5 (315)

a Total number for sex and specified age and omnivore groups are greater than the sum for males and females of those groups as some respondents did not indicate their sex.

questionnaires of 10 participants (1.5%) were excluded due to poor completion (>50% blank responses). Data on the consumption of all foods of animal origin listed in the food frequency questionnaire was necessary to determine vegetarian status. Complete data were available to categorize the vegetarian status of 630 respondents. There were no important sex, health, diet, or social adjustment differences

107

between those with and without complete flesh food consumption data except that alcohol use was higher among the latter ( P < 0.02). Characteristics of the study sample are given in Table 1. The VEG group consisted of 25 adolescents (4.0%, CI = 2.5 – 5.5%), of whom 16 (2.5%, CI = 1.3 –3.7%) were lactoovo vegetarian and 9 (1.4%, CI = 0.5 –2.3%) were semivegetarian. No vegans were identified. There were 605 OMN in the study. The prevalence of VEG was 6.5% (CI = 4.6– 8.4%) among females and 1.0% (CI = 0.2 –1.8%) among males. The frequency of food item consumption of the total group was determined and results are given in Table 2. White breads and cereals were a more popular daily choice than whole grain bread and cereals (64.5% vs. 44.1%). Consumption of fiber-containing foods by the total sample was fairly low. Sources of fiber such as whole grain breads or cereals, vegetables and fruit, and beans were consumed less than daily by 56%, 28.3%, and 93.9% of participants, respectively. Low-fat milk was a much more popular daily choice than full-fat milk (74.0% vs. 21.1%) with few individuals consuming soy milk on a daily basis (2.1%). From the meat and alternatives group, favored foods were

Table 2 Frequency of consumption of selected food items by adolescents Food (row percentages)

Grain products Whole grain breads or cereals White breads or cereals Vegetables and fruit Vegetables and fruit Fruit juice Milk products Milk—full fat/homogenized Milk—soy Milk—1%, 2%, or skim Cheese or yogurt Meat and alternatives Meat hamburger or hotdog Other types of beef, pork, or lamb Chicken or turkey Fish Eggs Vegetarian burger, hotdog, or other meat substitutes Beans (baked, chickpeas, kidney beans, lentils, tofu) Peanut butter Sweets and snack foods Soft drinks or other drinks containing sugar Candy, cake, pies, doughnuts French fries, potato chips, tortillas

Frequency of consumption

Total (n)

Never or rarely, % (n)

Monthly but not weekly, % (n)

Weekly but not daily, % (n)

Once a day, % (n)

At least twice a day, % (n)

11.6 (72) 4.9 (30)

16.6 (103) 7.6 (47)

27.8 (173) 23.1 (142)

30.4 (189) 44.2 (272)

13.7 (85) 20.3 (125)

622 616

5.1 (32) 4.3 (27)

2.7 (17) 4.6 (29)

20.5 (129) 13.6 (85)

36.9 (232) 36.5 (228)

34.8 (219) 41.0 (256)

629 625

56.4 91.0 10.4 8.1

(347) (549) (64) (50)

10.2 3.3 3.9 14.9

(63) (20) (24) (92)

12.2 (75) 3.5 (21) 11.6 (71) 36.2 (224)

11.2 (69) 1.3 (8) 29.0 (178) 30.9 (191)

9.9 0.8 45.0 10.0

(61) (8) (276) (62)

615 603 613 619

6.0 12.1 6.4 41.4 18.8 72.8

(38) (76) (40) (258) (117) (450)

23.2 22.5 19.8 31.7 30.1 13.6

(146) (142) (123) (198) (187) (84)

55.9 50.6 60.6 23.7 43.4 10.5

(352) (319) (377) (148) (270) (65)

12.7 (80) 13.7 (86) 11.9 (74) 2.9 (18) 6.6 (41) 2.1 (13)

2.2 1.1 1.3 0.3 1.1 1.0

(14) (7) (8) (2) (7) (6)

630 630 622 624 622 618

38.8 (241)

30.0 (186)

25.1 (156)

4.2 (26)

1.9 (12)

621

20.1 (124)

26.3 (162)

35.7 (220)

14.1 (87)

3.7 (23)

616

2.7 (17)

5.9 (37)

31.5 (197)

33.1 (207)

26.8 (168)

626

3.9 (24) 2.2 (14)

7.0 (43) 11.5 (72)

41.2 (255) 52.4 (328)

30.9 (191) 25.7 (161)

17.1 (106) 8.2 (51)

619 626

108

L.S. Greene-Finestone et al. / Preventive Medicine 40 (2005) 105–111

red meats and poultry. Approximately three fourths of participants did not eat fish weekly and half did not eat eggs weekly. Daily frequency of consumption of sweets was high, particularly for sugar-containing drinks (59.9%). Table 3 explores the impact of attitude toward health on food group consumption among males and females. Because most vegetarians were female, the analysis was repeated using only females. Patterns for the entire male and female sample were very similar to those for females only. The frequency of food group consumption (minimum of two servings daily) is described among VEG and OMN who reported themselves as health conscious and those who were not health conscious. Among the total group of the health conscious, less than half as many VEG consumed at least two daily servings of milk products compared to OMN ( P = 0.015). Among the non-health-conscious VEG, frequency of consumption of the grain and particularly the meat and alternatives food groups were of concern. A striking observation among the non-health-conscious VEG was the complete lack of twoserving frequency of consumption of meat and alternatives ( P < 0.001). The Breslow – Day test of homogeneity of odds ratios related to the latter suggests that the meat and alternatives intakes of health-conscious and non-healthconscious VEG and OMN were too different for the issue of the importance of personal health to be neglected in analysis. Further detail on this was explored to determine the extent of the dearth of meat and alternatives in the nonhealth-conscious VEG. Only 2/11 (18.2%) of them consumed even a single serving from this food group compared to 122/136 (89.7%) of non-health-conscious OMN. The very low frequency of grain product consumption among non-health-conscious VEG was of also interest, although it fell short of statistical significance ( P = 0.098). As well, the variety of food group intake (DDS) was lowest for nonhealth-conscious VEG.

Table 3 also provided an opportunity to compare the diets of OMN and VEG who were health conscious with those who are not. Health-conscious teenagers generally reported higher frequencies and greater diversity of food group consumption than non-health-conscious teens. Health-conscious OMN reported a significantly higher frequency of consumption from the grain, vegetables and fruit, and milk products food groups compared to non-health-conscious OMN. For healthconscious VEG, consumption of grain products, vegetables and fruit, and meat and alternatives were significantly higher than among non-health-conscious VEG. The daily consumption of food sources of fiber among health-conscious and non-health-conscious VEG and OMN is described in Table 4. Consumption of a variety of sources of fiber was highest among health-conscious VEG and lowest among non-health-conscious VEG. Daily consumption of a fiber-containing food was higher among the healthconscious compared to non-health conscious. Of note was the apparent effort of the health-conscious VEG to incorporate specialty items and other plant-based choices from the meat and alternatives group in their daily diets. Among health-conscious VEG, daily consumption of vegetarian meat substitutes and legumes was common (both 66.7%). Peanut butter was consumed by 83.3% on a daily basis. In contrast, consumption of these foods by nonhealth-conscious VEG was 0%, 9.1%, and 20.0%, respectively (all P < 0.05).

Discussion In this sample, the overall prevalence of vegetarianism among young adolescents was fairly low at 4.0%, although there was a large discrepancy between estimates for males (1.1%) and females (6.5%). The estimate for females appears reasonable and consistent with that obtained by

Table 3 Proportion of omnivore and vegetarian adolescents who state that ‘‘My health is very important’’ and ‘‘My health is somewhat/not important’’ consuming at least two daily servings of selected food groups ‘‘My health is very important’’ Omnivores (n = 433), % (n)

Vegetarians (n = 12), % (n)

Major food groups (consumption of food group z twice daily) Grain products 48.8 (204/418)§ 58.3 (7/12)b Vegetables and fruit 74.4 (318/427)§ 91.7 (11/12)b § Milk products 68.6 (275/401) 30.0 (3/10) Meat and alternativesz 68.0 (274/403) 63.6 (7/11)b Diet diversity score 3.6 F 0.66 (365)§ 3.2 F 0.97(9) (mean F 1 SD)** Currently on a dietz 10.1 (43/428) 25.0 (3/12) §

‘‘My health is somewhat/not important’’ P value

Omnivores (n = 136), % (n)

Vegetarians (n = 11), % (n)

P value

0.515* 0.309y 0.015y 0.751y 0.077

37.3 60.3 57.1 60.5 3.4

10.0 45.5 50.0 0 2.6

0.098y 0.357y 0.726* <0.001y 0.022

0.120y

7.4 (10/136)

(50/134) (82/136) (72/126) (78/129) F 0.89 (117)

P < 0.05 for omnivores who state ‘‘My health is very important’’ vs. ‘‘My health is somewhat/not important’’. P < 0.05 for vegetarians who state ‘‘My health is very important’’ vs. ‘‘My health is somewhat/not important’’. y Fisher’s Exact test. z Breslow – Day test for homogeneity of odds ratio < 0.05. * Chi-square. ** t test. b

(1/10) (5/11) (4/8) (0/10) F 1.06(8)

63.6 (7/11)

<0.001y

L.S. Greene-Finestone et al. / Preventive Medicine 40 (2005) 105–111

109

Table 4 Proportion of omnivore and vegetarian adolescents who state that ‘‘My health is very important’’ and ‘‘My health is somewhat/not important’’ consuming daily sources of fiber Daily consumption of food item/nutrient

Sources of fiber Legumes Vegetables and fruit Whole wheat bread and cereals Daily fiber source

‘‘My health is very important’’ Omnivores (n = 433), % (n)

‘‘My health is somewhat/not important’’ Vegetarians (n = 12), % (n)

Omnivores (n = 136), % (n)

Vegetarians (n = 11), % (n)

34.3 (146/426) 75.2 (325/432) 48.5 (207/427)

66.7 (8/12) 91.7 (11/12) 58.3 (7/12)

21.5 (29/135) 65.4 (89/136) 34.1 (46/135)

9.1 (1/11) 63.6 (7/11) 18.2 (2/11)

84.7 (355/419)

100.0 (12/12)

75.4 (101/134)

72.7 (8/11)

the Canadian Eating Trends Study [6] in which vegetarianism increased with age from 3.3% for 10 – 14 year olds to 8.2% for 15 – 18 year olds. In another Ontario sample, 8.8% of female adolescents reported being vegetarian [5], but there is some evidence that up to a third of self-reported vegetarian teens may regularly be consuming red meat [18]. The prevalence for males reported here was somewhat lower than the 1.7– 2.5% for males previously reported [5,6]. Approximately 2/3 of VEG were lacto-ovo vegetarians and the remainder, semivegetarians. Both VEG subgroups were largely female. Most studies have reported a greater prevalence of semivegetarianism compared to lacto-ovo vegetarianism [2,3]. It is possible that the criterion for semivegetarian classification in this study was different from that of previous population studies. There may be a tendency for those who are semivegetarians to occasionally (e.g., once a month) eat red meat. In this study, such a person would be classified as an omnivore. Only those reporting red meat consumption as ‘‘rarely or never’’ were classified as semivegetarians. Among all subjects in this sample, consumption of lowfat milk was more popular than that of full-fat milk. Negative findings were the high daily intakes of sugarcontaining beverages and foods. A high intake of such foods has previously been observed among Ontario schoolchildren [19]. In light of the precipitous rise in obesity among Canadian children and young teenagers in the past 20 years [20], the latter is an area that may be targeted for change. Some teenagers may have consumed an inadequate diet. This is reflected by the fact that large proportions of both VEG and OMN did not consume even two servings of each food group daily (Table 3). Yet, degree of health consciousness appeared to strongly differentiate between the diet quality of both the VEG and OMN groups. Generally, the frequencies of food group consumption among health-conscious VEG and OMN were superior to those of their nonhealth-conscious counterparts in terms of quality and variety. As well, frequencies of consumption were similar among health-conscious VEG and OMN and among nonhealth-conscious VEG and OMN. Two major exceptions were in the milk products and meat and alternatives groups. In the milk products group, the frequency of consumption of two daily servings was lowest among both health-

conscious and non-health-conscious VEG. This finding suggests that some of the health-conscious VEG may have perceived milk to be unhealthy. In other studies comparing the consumption of milk products or calcium intake among VEG and OMN (that did not distinguish between degrees of health consciousness), results have been mixed. VEG have demonstrated higher intakes [2], lower intakes [8], and similar intakes [3,21]. Consumption of milk products is a consequential issue as they have been reported to be the major source of calcium for adolescent vegetarians and omnivores [22,23]. If consumption is low, alternative sources of calcium are required and supplementation may be necessary to meet their recently increased Dietary Reference Intake of 1,300 mg/day [24]. Milk products are also rich sources of calcium and vitamin D and important sources of protein and vitamins A and B12. Large proportions of both VEG and OMN did not consume two servings of meat and alternatives daily. This was particularly an issue for the non-health-conscious VEG; none consumed two servings daily and only 18% consumed a single daily serving. Poor frequency of consumption of meat and alternatives would adversely affect protein, iron, and zinc intakes unless compensation occurred through sources in other food groups. Perry et al. [3] reported higher iron consumption in vegetarians compared to omnivores in a large school-based sample. Further, mean protein and zinc intakes of VEG and OMN were similar and met Dietary Reference Intake levels [25]. Most other data on intake of these nutrients in adolescent vegetarians and omnivores derive from convenience samples and results are varied [21,26 –28]. Consumption of grain products and vegetables or fruit did not reach even two daily servings for large proportions of the health-conscious and non-health-conscious VEG and OMN groups. Poor intake of fruits and vegetables has been previously observed among adolescents [29,30]. These food groups have been observed to be primary sources of energy, protein, fiber, iron, zinc and vitamins C, thiamine, niacin, and riboflavin for vegetarians and omnivores [22]. They are also important sources of magnesium, folic acid, and vitamin A. Although higher fiber intake among adolescent vegetarians has been reported [3,21], in this study, consumption of a daily fiber source was higher for the health-

110

L.S. Greene-Finestone et al. / Preventive Medicine 40 (2005) 105–111

conscious and similar among the non-health-conscious OMN and VEG. We can speculate about why the frequency of food group consumption by non-health-conscious VEG are the least satisfactory for almost all food groups. In this study, far more non-health-conscious VEG were currently on a diet to lose weight. Further research is necessary to determine if there is a relationship between non-health-conscious vegetarianism, weight loss dieting, and unhealthy food restriction. This study presents some direction for strategies for improving the diets of adolescents. A substantial proportion of teenagers (21 – 28%) consumed sources of fiber on a weekly basis. These teenagers are not totally opposed to consuming these foods, and with education and encouragement could be persuaded to increase their frequency of consumption to a daily basis. More creative strategies would be required for the remainder of the group. A further initiative would be to shift away from beverages and foods that are high in sugars. Higher soft drink consumption has been related to greater energy intake and likelihood of obesity in children and adolescents [31,32]. For this reason, it has been suggested that the substantial increase in soft drink consumption has, in part, contributed to the rise in obesity among youth today [33]. A strength of this survey is that it is school-based with entire ninth grades eligible to participate in consenting schools. It offers a recent, large sample of young adolescents. Other studies have examined the dietary intakes of VEG and OMN but to our knowledge, this is the first to explore differences in consumption patterns based on the importance individuals place on their health. Limitations include the uneven response rate within schools that could potentially affect the survey’s external generalizability. Although responses from low and higher response schools did not differ significantly on issues studied, the impact of nonresponse bias is unknown because characteristics of the respondents and nonrespondents could not be compared. The FFQ offered estimates of dietary intake at the level of foods or food groupings. Because it was not comprehensive and there were assumptions around portion sizes, it was prudent not to use FFQ information to determine intake at the level of nutrients. Although FFQs are recommended for large surveys, it must be acknowledged that they cannot account for the consumption of all foods. Results should be confirmed using multiple 24-hour recalls or a FFQ that has been validated with adolescents against actual intakes as determined by other methods or repeat measures. The sample of vegetarians was limited and was further reduced by stratification based on health consciousness. Because of this, the representativeness of the groups is uncertain and caution should be used in interpreting the results. The small size of the vegetarian sample did not allow for the study of sex as an interaction with frequency of food group consumption. In a recent large study exploring similar issues, sex by vegetarian status (and even race by vegetarian status) interactions were not observed when vegetarian status was similarly defined

[3]. The study’s results should be confirmed using a larger sample size. Despite the small size, however, most differences between the health conscious and non-health conscious were large enough to reach statistical significance. This survey is cross-sectional in design. Therefore, it was not possible to infer causal links among the variables.

Conclusion This study presents an estimate of the prevalence of vegetarian and omnivore dietary styles in a young adolescent sample. The finding that 4% of the sample are vegetarian (6.5% of females and 1.0% of males) suggests that this is not a rampant phenomenon in this population. Large proportions of both vegetarians and omnivores groups consumed suboptimal diets. In the quest for nutritional adequacy, adherents to both types of diets can benefit from additional knowledge and planning. Although the eating behaviors of the vegetarians reflected those of the omnivores with respect to the influence of health consciousness, results should be confirmed using a larger sample of vegetarians. The non-health-conscious group in particular should be targeted for nutrition education strategies. The notion of health consciousness may be useful as a rudimentary screen for problematic dietary consumption patterns in adolescents in clinical and public health programs.

References [1] American Dietetic Association, Dietitians of Canada. Position of the American dietetic association and dietitians of Canada: vegetarian diets. J Am Diet Assoc 2003;64(2):62 – 81. [2] Worsley A, Skrzypiec G. Teenage vegetarianism: prevalence, social and cognitive contexts. Appetite 1998;30(2):151 – 70. [3] Perry C, McGuire MT, Neumark-Sztainer D, Story M. Adolescent vegetarians. How well do their dietary patterns meet the healthy people 2010 objectives? Arch Pediatr Adolesc Med 2002;156(2): 431 – 7. [4] Larsson CL, Klock KS, Astrom AN, Haugejorden O, Johansson G. Prevalence and food habits of vegetarian adolescents. Public Health Nutr 2001;4(5):1005 – 14. [5] Cohen B, Evers S, Manske S, Bercovitz K, Edward HG. Smoking, physical activity and breakfast consumption among secondary school students in a Southwestern Ontario community. Can J Public Health 2003;94(1):41 – 4. [6] Beef Information Centre, Agriculture and Agri-Food Canada. Canadian Eating Trends Study; 1998. [7] Neumark-Sztainer D, Story M, Perry C, Casey MA. Factors influencing food choices of adolescents: Findings from focus—group discussions with adolescents. J Am Diet Assoc 1999;99(1):929 – 37. [8] Neumark-Sztainer D, Story M, Resnick MD, Blum RW. Adolescent vegetarians, a behavioural profile of a school-based population in minnesota. Arch Pediatr Adolesc Med 1997;151:833. [9] Perry CL, McGuire MT, Neumark-Sztainer D, Story M. Characteristics of vegetarian adolescents in a multiethnic urban population. J Adolesc Health 2001;29:406 – 16. [10] Worsley A, Skrzypiec G. Teenage vegetarianism: beauty or the beast? Nutr Res 1997;17:391 – 407. [11] Horacek TM, Betts NM. Students cluster into 4 groups according to

L.S. Greene-Finestone et al. / Preventive Medicine 40 (2005) 105–111

[12]

[13]

[14]

[15]

[16] [17] [18]

[19]

[20] [21]

[22]

[23]

the factors influencing their dietary intake. J Am Diet Assoc 1998;98(12):1464 – 7. DeWit DJ, Braun K, Ellis K, Prentice L, Silverman G, Smythe C, et al. In: Addiction Research Foundation C, editor. The evaluation of opening doors: an in-school drug prevention program for at-risk youth in their transition years. Toronto: Addiction Research Foundation; 1997. Kant AK, Schatzkin A, Harris TB, Zeigler RG, Block G. Dietary diversity and subsequent mortality in the first national health and nutrition examination survey epidemiologic follow-up study. Am J Clin Nutr 1994;57(3):434 – 40. Kant AK, Schatzkin A, Zeigler RG. Dietary diversity and subsequent cause-specific mortality in the NHANES I epidemiologic follow-up study. J Am Coll Nutr 1995;14(3):233 – 8. US Department of Agriculture HNIS. The food guide pyramid. Home and Garden Bulletin, vol. 252. Hyattsville (MD): US Department of Agriculture; 1992. SAS. Cary, NC; 1999. Ridout MS, Demetrio CGB, Firth D. Estimating intraclass correlation for binary data. Biometrics 1999;55:137 – 48. Greene-Finestone LS, Campbell MK, Evers SE, Gutmanis IA. (Abstract) Vegetarian status of adolescents: consistency between two selfreport measures. Can J Diet Pract Res 2003;64(Suppl 2):S109. Evers S, Taylor J, Manske S, Midgett C. Eating and smoking behaviours of school children in Southwestern Ontario and Charlottetown, PEI. Can J Public Health 2001;92(6):433 – 6. Tremblay MS, Willms JD. Secular trends in the body mass index of Canadian children. Can Med Assoc J 2000;163(11):1429 – 33. Donovan UM, Gibson RS. Iron and zinc status of young women aged 14 to 19 years consuming vegetarian and omnivorous diets. J Am Coll Nutr 1995;14:463 – 72. Donovan UM, Gibson RS. Dietary intakes of adolescent females consuming vegetarian, semi-vegetarian and omnivorous diets. J Adolesc Health 1996;18:292 – 300. Hedley M, Chambers LW, Tomasik HH, Randall E, Woolcott D, Brown H, et al. Nutrition Report. Working Paper No. 10, p. 1 – 86;

[24]

[25]

[26]

[27] [28]

[29]

[30]

[31]

[32]

[33]

111

1995. Ontario: Ministry of Health. Ontario Health Survey 1990, Working Papers. Standing Committee on the Scientific evaluation of Dietary Reference Intakes FaNBIoM. Dietary reference intakes for calcium, phosphorus, magnesium, vitamin D and fluoride. Washington (DC): National Academy Press; 1997. Standing Committee on the Scientific evaluation of Dietary Reference Intakes FaNBIoM. Dietary reference intakes for A. vitamin, K. vitamin, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium and zinc. Washington (DC): National Academy Press; 2001. McEndree LS, Kies CV, Fox HM. Iron intake and nutritional status of lacto-ovo-vegetarian and omnivore students eating in a lacto-ovovegetarian food service. Nutr Rep Int 1983;27:199 – 206. Helman AD, Danton-Hill I. Vitamin and iron status in new vegetarians. Am J Clin Nutr 1987;45:785 – 9. Treuherz J. Possible inter-relationship between zinc and dietary fibre in a group of lacto-ovo vegetarian adolescents. J Plant Foods 1982;4:89 – 93. King AJC, Boyce WF, King MA. Trends in the health of Canadian youth. Canada: Health Canada, 1999. p. 1 – 110 [Health Behaviour in School-Aged Children—A World Health Organization CrossI National Study]. Neumark-Sztainer D, Story M, Resnick MD, Blum RW. Correlates of inadequate fruit and vegetable consumption among adolescents. Prev Med 1996;25(5):497 – 505. Harnack L, Stang J, Story M. Soft drink consumption among US children and adolescents: nutritional consequences. J Am Diet Assoc 1999;99:436 – 41. Ludwig DS, Peterson KE, Gortmaker SL. Relation between consumption of sugar-sweetened drinks and childhood obesity: a prospective, observational analysis. Lancet 2001;357:505 – 8. St-Onge M-P, Keller KL, Heymsfield SB. Changes in childhood food consumption patterns: a cause for concern in light of increasing body weights. Am J Clin Nutr 2003;78:1068 – 73.