Changes in dietary and physical activity risk factors for type 2 diabetes in Alberta youth between 2005 and 2008.
Forbes, Laura E. ; Fraser, Shawn N. ; Downs, Shauna M. 等
Unhealthy dietary and physical activity patterns are prevalent
among Canadian youth (1,2) and are implicated in increasing rates of
early onset type 2 diabetes that have been observed in Canada. Type 2
diabetes, which until the 1970s was considered an adult disease, is
increasing in prevalence in youth, with one clinic in Manitoba, Canada
reporting a 15-fold increase in the number of children with type 2
diabetes between 1987 and 2005. (3) Lifestyle risk factors for type 2
diabetes for adults include overweight and obesity, (4) high dietary
glycemic index (GI), high dietary glycemic load (GL), (5) low fibre, (6)
low magnesium intake, (7) low vegetable and fruit consumption, (8) high
fat intake, (9) and low physical activity levels. (10) These risk
factors represent important targets for health promotion and disease
prevention.
Schools have been suggested as an ideal location for lifestyle
interventions targeting adolescents, as children spend much of their
time in school. Growing concerns in the mid-2000s about the health of
Alberta youth spawned the creation and implementation of many school
health initiatives. In 2005, the Alberta Coalition for Healthy School
Communities hosted an inaugural conference bringing together school
communities interested in promoting health, (11) and in 2007, the
Alberta Healthy School Community Wellness Fund was initiated to support
healthy school programming. (12) In 2006, the Alberta Government
introduced the Daily Physical Activity Initiative, which required grades
1-9 students to accumulate 30 minutes of physical activity each day.
(13) In 2007, the Edmonton Public School Board created a policy stating
the importance of creating healthy food and physical activity promoting
environments. (14) Although the efficacy of individual school
initiatives has been assessed previously, (15,16) the province-wide
effect of this variety of initiatives has not been investigated.
Therefore, the purpose of this study was to examine differences in the
prevalence of lifestyle risk factors for type 2 diabetes between two
province-wide samples of Alberta adolescents in grades 7 to 10, one
surveyed in 2005 and the other in 2008.
METHODS
Participants and procedure
In 2005 and 2008, students in grades 7-10 (aged 11-16) in Alberta
were recruited to participate in the Web Survey of Physical Activity and
Nutrition (Web-SPAN). The methods for study have been described
previously. (17,18) This survey was a self-administered, web-based
questionnaire that assessed dietary and physical activity patterns and
social and environmental determinants of these behaviours. Recruitment
protocols differed slightly between the two data collection periods. In
2005, all Alberta School Boards were invited to participate and 7
schools per board were randomly selected for the study. In 2008, schools
were randomly selected from a list of all Alberta schools. In both 2005
and 2008, information about the study was mailed to selected school
boards and schools. Participation was subsequently requested by
telephone. Active parental consent and student assent were obtained. The
Web-SPAN was conducted during school time and took approximately 45
minutes to complete. In 2005, the school board, school and student
response rates were 75%, 37% and 27%, respectively and in 2008, response
rates were 64%, 40% and 44%. The student response rate was significantly
higher in 2008 due to the participation of several schools whose
administration invested extra effort into recruitment for this study.
Ethical approval was obtained from the University of Alberta.
Measures
Demographic and Anthropometric Characteristics Median neighbourhood
income of schools was determined by matching 2006 Canadian census data
to each school using postal codes. (19) Students entered self-reported
height and weight into the Web-SPAN and BMI (kg/[m.sup.2]) was
calculated.
Dietary Intake
Dietary intake of participants was measured using a validated
web-based 24-hour recall tool: the Food Behaviour Questionnaire. (20) No
changes were made to this tool between 2005 and 2008. Students recorded
all food and beverages consumed in the previous day, choosing from a
list of over 800 foods. Cues were provided for portion sizes and
students were prompted to record foods often missed. Dietary intake data
were analyzed using the ESHA Food Processor (version 7.9, ESHA Research,
Salem, OR) and the 2001 Canadian Nutrient File (21) database. Food group
intake according to Canada's Food Guide (22) was calculated.
To calculate dietary GI and GL, each carbohydrate-containing food
in the Web-SPAN database was assigned a GI score based on published
values. Daily dietary GI and GL were calculated by a method described
previously. (23)
Physical Activity
Physical activity was assessed using an online version of the
Physical Activity Questionnaire for Older Children (PAQ-C). (24)
Physical activity levels were determined based on self-reported
activities performed in the last 7 days. The PAQ-C scores range from 1
to 5 with 1 being least active and 5 being most active. (24)
Defining Risk Factors
Major dietary and physical activity risk factors for diabetes and
cutoffs for identifying those at risk were determined using literature
searches. Evidence was drawn from large epidemiological studies of
diabetes risk. (4-10,18,25,26) Wherever possible, studies of children
and youth were evaluated, (18,25,26) but these were limited in number.
Dichotomous variables (at risk or not at risk) for each risk factor were
constructed based on levels of dietary intake that were significantly
related to risk of developing diabetes. Studies referenced and cut-offs
used are found in Table 1. Risk factors identified included overweight,
obesity, high GI, high GL, low fibre, low magnesium, low vegetable and
fruit intake, high fat intake and low physical activity levels.
Statistical analysis
Predetermined criteria from the 2005 survey were used to identify
and remove outliers from the database. Those removed from analysis
included students who signed onto but did not complete any part of the
survey (2005: n=9, 2008: n=26), and extreme outliers of total energy
intake (2005: n=24, 2008: n=33), physical activity (2005: n=14, 2008:
n=19) and BMI (2005: n=34, 2008: n=33). The total number of outliers
removed was 36 in 2005 and 111 in 2008, resulting in final sample sizes
of 4,936 in 2005 and 5,119 in 2008.
Characteristics of participants were assessed using descriptive
statistics and differences between boys and girls were determined using
independent sample t-tests. The frequency of each diabetes risk factor
was assessed using descriptive statistics. Differences in the proportion
of participants reporting risk factors between 2005 and 2008 were
determined using chi-square tests. As a follow-up, specific risk factors
were tested as dependent variables in a series of logistic regression
models to determine whether differences in risk factor presence between
2005 and 2008 remained after accounting for age and median income
quintiles.
The total number of risk factors was calculated for each
participant. For the risk factors of overweight and obesity, overweight
individuals were considered to have one risk factor and obese
individuals were considered to have two risk factors due to their weight
status (i.e., they were considered both overweight and obese).
Intraclass correlations (ICC) were calculated using Restricted Maximum
Likelihood Estimation (REML) for the total number of risk factors to
examine whether school clustering was present. ICCs of 0.040 for the
entire sample, 0.029 for 2005 and 0.032 for 2008 were calculated,
indicating that a very small level of clustering was present. Therefore,
linear mixed-modelling (LMM) was used to account for clustering. The
intercepts-only model showed that school clustering had a modest effect
on the number of risk factors (p<0.001), therefore clustering was
included in the final model. To reduce collinearity, the continuous
variables of age and income were centred about the grand mean. This was
not done for sex and year. Median neighbourhood income was considered a
level two factor compared to year, sex, and age which were level one
factors. All statistical analyses were performed using SPSS statistical
software (Version 19.0.0.1; SPSS Inc, Chicago).
RESULTS
In total, 4,936 students from 136 schools within 44 school boards
participated in this study in 2005 and 5,119 students from 109 schools
in 37 school boards participated in 2008 (Table 2a). The mean age was
13.6 years in 2005 and lower at 13.3 years in 2008 (p<0.01). In 2005
and 2008, 54% and 55% of participants were girls, respectively,
representing a slight over-sampling of female students. Median school
neighbourhood income was higher in 2008 compared to 2005 at $74,797 and
$71,739, respectively (p<0.010) (Table 2a). Median school
neighbourhood income in 2005 was comparable to median income of Alberta
families in 2005 ($71,000). (27) Rates of overweight were similar to
those reported by the CCHS in 2004; obese youth were under-represented.
(28)
Differences in mean values for each of the risk factor variables
between 2005 and 2008 were assessed using independent sample t-tests.
Mean BMI, total energy intake, % energy from fat, GI and GL were
significantly lower in 2008 compared to 2005 and % energy from
carbohydrate, % energy from protein, fibre and vegetable and fruit
intakes were higher in 2008 (Table 2a).
When demographic variables were separated by sex, several
significant differences between the genders were observed (Table 2b). At
both time points, boys had greater BMIs, total energy, % fat, % protein,
GI, GL, fibre, magnesium and vegetable and fruit intakes and had higher
physical activity scores compared to girls. Girls had a greater % of
energy coming from carbohydrate.
When risk factor cut-offs were applied (Table 3a), there was a
lower proportion of youth with the risk factors of overweight, obesity,
high GI diets, high GL diets, high fat diets, low fibre diets and low
vegetable and fruit intake in 2008 compared to 2005. No differences
existed in the proportion of youth consuming low magnesium diets or
having low levels of physical activity. When the results were separated
by sex (Table 3b), a lower proportion of boys had the risk factors of
overweight, obesity, high GI and high GL in 2008 compared to 2005. A
lower proportion of girls had the risk factors of high GI, high GL, low
fibre, low magnesium and high fat in 2008 compared to 2005. When the
risk factors were tested to determine whether these differences between
2005 and 2008 remained after accounting for age and median income
quintiles, results showed that all differences persisted with the
exception of magnesium (Table 4), which was no longer different between
2005 and 2008 after accounting for age and income. Contrary to the
results of chi square tests, year and sex were significantly related to
physical activity score in the logistic regression model, with girls,
and those surveyed in 2005 being more likely to have low physical
activity levels. Further, compared to the highest income category,
participants in the bottom 3 income quintiles were at higher risk of
inactivity.
The mean total number of risk factors (Figure 1) was significantly
higher in 2005 compared to 2008 (3.1 vs. 2.9, p<0.001). Girls
reported having significantly fewer risk factors than boys in both 2005
and 2008 (2005: boys 3.2, girls 3.0; 2008: boys 3.0, girls 2.8;
p<0.001 for both years). When a REML LMM was applied to determine
whether the differences observed persisted after accounting for
differences in age, sex and income, this pattern was unchanged
(p<0.001). However, each year of age was associated with an increase
in the mean number of risk factors. School area income was also related
to the total number of risk factors, with a higher number of risk
factors being associated with a lower mean income.
DISCUSSION
This study examined the presence of diabetes risk factors in large
samples of Alberta adolescents in 2005 and 2008 and found that the
prevalence of many of these risk factors was high. However, independent
of demographic differences between the samples, the prevalence of
several diabetes risk factors decreased from 2005 to 2008. Food and
nutrient intakes observed were similar to intakes observed in other
North American studies of youth. (29,30) We found that dietary risk
factors for type 2 diabetes are highly prevalent in Alberta adolescents.
The most commonly reported dietary risk factors were high GI and GL
diets, low fibre intakes and low magnesium intakes - all of which
indicate that the diets of youth are low in vegetables, fruit and whole
grains. These results are concerning as low GI, high fibre foods help to
prevent obesity, (31) and contain many beneficial vitamins, minerals and
antioxidants; (32) all of which may help to decrease diabetes risk.
Over 50% of youth in the current study were considered inactive
which is consistent with the findings of other studies of Canadian youth
using the PAQ-C. (33) Low levels of physical activity are concerning as
physical activity decreases diabetes risk both by helping to regulate
body weight and body composition and by increasing insulin sensitivity,
(34) and is critical to the prevention of many other chronic diseases.
Students in low SES-area schools were more likely to be physically
inactive and have a greater number of risk factors. Studies of the
effects of SES on Canadian youth consistently show strong effects of SES
on health behaviours. (35) In the current study, SES was measured using
postal codes; an imprecise method which does not capture individual
variation in SES. However, the link between SES and health behaviours
was identified even when using an imprecise measure, suggesting that SES
may be of greater importance than demonstrated in this study. Previous
studies show that school community SES affects the health behaviours of
youth independently of individual SES, likely due to obesogenic
environments of low SES neighbourhoods. (35)
Independent of differences in demographic characteristics, boys
were less likely to have the risk factors of overweight, obesity, high
GI and high GL, and girls were less likely to have the risk factors of
high GI, high GL, low fibre and high fat in 2008 compared to 2005. In
addition, the total number of risk factors decreased for both boys and
girls between 2005 and 2008. This suggests that the diets of youth in
Alberta have shown modest improvement. Physical activity levels of youth
did not change during this time period. This lack of change in physical
activity levels may confirm previous reports that school physical
activity programs were poorly implemented, (13) or that the PAQ-C was
unable to detect the changes. Although the cause(s) of these dietary
changes over time cannot be established from this study due to the lack
of a control group, several changes happened between 2005 and 2008 in
Alberta that may have contributed to this trend. During this time, many
school boards adopted school food policies, schools became engaged in
programs that took a Comprehensive School Health approach, (11-14) and
the problem of obesity in childhood and adolescence was widely
publicized; all of which may have contributed to the changes observed.
[FIGURE 1 OMITTED]
Canadian studies that examined the effects of school policies and
comprehensive school health initiatives have shown significant effects
on the health behaviours of youth. One study assessed the effect of
school food policy changes in Prince Edward Island on the dietary
choices of youth in grades 5-6 by measuring dietary intake before
(2001/2002) and one year after policy implementation (2007). This study
showed that youth consumed approximately one fewer serving of
low-nutrient dense food per day post intervention. (16) An Alberta study
evaluated the effects of the APPLE Schools program (a comprehensive
school health intervention). Students in APPLE schools ate more
vegetables, fewer calories, were more physically active and were less
likely to be obese in 2010 compared to 2008. (15) The magnitudes of the
changes observed in the current study are comparable to changes seen in
the policy study. It is plausible that changes in the current study were
the result of changes in health policies and programs, but as
participation in these programs was not measured in this study,
definitive conclusions cannot be made. Other explanations for this
difference could include the effect of the growing awareness of obesity
in Alberta adolescents on their health behaviours. It is also possible
that these differences were caused by increasing interest in adolescent
health which led to increased response bias in 2008. The higher student
response rate and higher SES seen in 2008 may be evidence of this.
Further study is needed to determine whether positive lifestyle changes
occurred in Canadian youth. Studies that track possible causes of these
changes would allow Canadian school administrators and public policy
makers to pinpoint effective interventions for more extensive
implementation across Canada.
Strengths of this study include the large samples and use of
validated tools to assess dietary intake and physical activity. Few
Canadian studies of this size have evaluated the lifestyles of youth or
followed changes over time. Although sampling techniques differed
slightly between 2005 and 2008, both appeared to be fairly
representative of Alberta youth. The convenience sampling method used is
subject to volunteer bias, which may have affected study results.
Schools and students interested in health may have been more likely to
participate and improvements in dietary patterns may have been more
likely to happen in this population. The survey did not assess ethnicity
or family history of diabetes, which are important diabetes risk
factors. However this study aimed to examine modifiable lifestyle risk
factors to determine opportunities for prevention. Finally, cut-offs for
diabetes risk were based on values from adult literature and it is not
known how appropriate these cut-offs are for youth.
In conclusion, this study showed that current dietary intake and
physical activity patterns among Alberta youth could increase risk of
future type 2 diabetes, however, dietary behaviours showed improvement
between 2005 and 2008. Physical activity levels were low at both time
points and showed no improvements. Although the cause of these changes
could not be determined from this study, it is possible that increased
awareness and the implementation of school health programs and policies
contributed to this change. Continued monitoring of adolescent dietary
patterns is warranted. Monitoring of exposure to programs and policies
is needed to pinpoint the causes of these changes.
Acknowledgements: The authors thank our funders and all the
students, parents and teachers who participated in this study.
Funding Sources: Canadian Institutes of Health Research/Heart and
Stroke Foundation, Alberta Innovates: Health Solutions, and the Alberta
Centre for Child, Family and Community Research.
Conflict of Interest: None to declare.
REFERENCES
(1.) Active Healthy Kids Canada. 2013 Active Healthy Kids Canada
Report Card on Physical Activity for Children and Youth. Toronto, ON:
Active Healthy Kids Canada, 2013.
(2.) Health Canada. Do Canadian Adolescents Meet Their Nutrient
Requirements Through Food Intake Alone? Ottawa, ON: Health Canada, 2012.
(3.) Toth G. Type 2 diabetes occurs in Aboriginal youth in
Manitoba? Diabetes Q 2005; Autumn:10.
(4.) Hu FB, Manson JE, Stampfer MJ, Colditz G, Liu S, Solomon CG,
et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in
women. N Engl J Med 2001;345(11):790-97.
(5.) Barclay AW, Petocz P, McMillan-Price J, Flood VM, Prvan T,
Mitchell P, et al. Glycemic index, glycemic load, and chronic disease
risk--a meta-analysis of observational studies. Am J Clin Nutr
2008;87(3):627-37.
(6.) Schulze MB, Liu S, Rimm EB, Manson JE, Willett WC, Hu FB.
Glycemic index, glycemic load, and dietary fiber intake and incidence of
type 2 diabetes in younger and middle-aged women. Am J Clin Nutr
2004;80(2):348-56.
(7.) Lopez-Ridaura R, Willett WC, Rimm EB, Liu S, Stampfer MJ,
Manson JE, et al. Magnesium intake and risk of type 2 diabetes in men
and women. Diabetes Care 2004;27(1):134-40.
(8.) Liu S, Serdula M, Janket SJ, Cook NR, Sesso HD, Willett WC, et
al. A prospective study of fruit and vegetable intake and the risk of
type 2 diabetes in women. Diabetes Care 2004;27(12):2993-96.
(9.) van Dam RM, Willett WC, Rimm EB, Stampfer MJ, Hu FB. Dietary
fat and meat intake in relation to risk of type 2 diabetes in men.
Diabetes Care 2002;25(3):417-24.
(10.) Hsia J, Wu L, Allen C, Oberman A, Lawson WE, Torrens J, et
al. Physical activity and diabetes risk in postmenopausal women. Am J
Prev Med 2005;28(1):19-25.
(11.) Alberta coalition for healthy school communities historical
timeline. 2009. Available at: http://www.achsc.org/history.htm (Accessed
May 1, 2013).
(12.) Alberta healthy school community wellness fund website. 2012.
Available at: http://www.publichealth.ualberta.ca/collaborators_partners/community_eng agement/ahscwf.aspx (Accessed May 1, 2013).
(13.) Alberta Education. Daily physical activity survey report
2007. Edmonton, AB: Alberta Education, 2008.
(14.) Edmonton public school board health and wellness of staff and
students policy. 2007. Available at:
http://www.epsb.ca/policy/gbe.bp.shtml (Accessed May 1, 2013).
(15.) Fung C, Kuhle S, Lu C, Purcell M, Schwartz M, Storey K, et
al. From "best practice" to "next practice": The
effectiveness of school-based health promotion in improving healthy
eating and physical activity and preventing childhood obesity. Int
JBehav Nutr Phys Act 2012;9:27.
(16.) Mullally ML, Taylor JP, Kuhle S, Bryanton J, Hernandez KJ,
MacLellan DL, et al. A province-wide school nutrition policy and food
consumption in elementary school children in Prince Edward Island. Can J
Public Health 2010;101(1):40-43.
(17.) Storey KE, Forbes LE, Fraser SN, Spence JC, Plotnikoff RC,
Raine KD, et al. Diet quality, nutrition and physical activity among
adolescents: The web-SPAN (web-survey of physical activity and
nutrition) project. Public Health Nutr 2009;12(11):2009-17.
(18.) Plotnikoff RC, Karunamuni N, Spence JC, Storey K, Forbes L,
Raine K, et al. Chronic disease-related lifestyle risk factors in a
sample of Canadian adolescents. J Adolesc Health 2009;44(6):606-9.
(19.) Sin DD, Svenson LW, Man SF. Do area-based markers of poverty
accurately measure personal poverty? Can J Public Health
2001;92(3):184-87.
(20.) Hanning RM, Royall D, Toews JE, Blashill L, Wegener J,
Driezen P. Web-based food behaviour questionnaire: Validation with
grades six to eight students. Can J Diet Pract Res 2009;70(4):172-78.
(21.) Canadian nutrient file [electronic version]. 2001. Available
at: http://www.hcsc.gc.ca/food-aliment/ns-sc/nr-rn/surveillance/cnf-fcen/e_index.html (Accessed May 1, 2013).
(22.) Eating Well With Canada's Food Guide. 2007. Available
at: http://www.hcsc.gc.ca/fn-an/alt_formats/hpfb-dgpsa/pdf/food-guide-aliment/view_eatwell_ vue_bienmang-eng.pdf (Accessed May 1, 2013).
(23.) Forbes LE, Storey KE, Fraser SN, Spence JC, Plotnikoff RC,
Raine KD, et al. Dietary patterns associated with glycemic index and
glycemic load among alberta adolescents. Appl Physiol Nutr Metab
2009;34(4):648-58.
(24.) Kowalski KC, Crocker ORE, Donen RM. The physical activity
questionnaire for older children (PAQ-C) and adolescents (PAQ-A) manual.
2004.
(25.) Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a
standard definition for child overweight and obesity worldwide:
International survey. BMJ 2000;320(7244):1240-43.
(26.) Morrison JA, Glueck CJ, Horn PS, Schreiber GB, Wang P.
Pre-teen insulin resistance predicts weight gain, impaired fasting
glucose, and type 2 diabetes at age 18-19 y: A 10-y prospective study of
black and white girls. Am J Clin Nutr 2008;88(3):778-88.
(27.) Statistics Canada. Census families include couple families,
with or without children, and lone-parent families CANSIM, table
111-0009. 2010.
(28.) Shields M. Overweight and obesity among children and youth.
Health Rep 2006;17(3):27-42.
(29.) Statistics Canada. Canadian community health survey (CCHS
3.1), CANSIM table 1050449. 2005.
(30.) U.S. Department of Agriculture, Agricultural Research
Service. Nutrient intakes from food: Mean amounts consumed per
individual, by race/ethnicity and age, one day, 2005-2006. 2008.
(31.) Lindstrom J, Peltonen M, Eriksson JG, Louheranta A, Fogelholm
M, Uusitupa M, et al. High-fibre, low-fat diet predicts long-term weight
loss and decreased type 2 diabetes risk: The Finnish diabetes prevention
study. Diabetologia 2006;49(5):912-20.
(32.) Zhou K, Su L, Yu LL. Phytochemicals and antioxidant
properties in wheat bran. J AgricFood Chem 2004;52(20):6108-14.
(33.) Carter LM, Whiting SJ, Drinkwater DT, Zello GA, Faulkner RA,
Bailey DA. Self-reported calcium intake and bone mineral content in
children and adolescents. J Am Coll Nutr 2001;20(5):502-9.
(34.) O'Gorman DJ, Karlsson HK, McQuaid S, Yousif O, Rahman Y,
Gasparro D, et al. Exercise training increases insulin-stimulated
glucose disposal and GLUT4 (SLC2A4) protein content in patients with
type 2 diabetes. Diabetologia 2006;49(12):2983-92.
(35.) Janssen I, Boyce WF, Simpson K, Pickett W. Influence of
individual- and area-level measures of socioeconomic status on obesity,
unhealthy eating, and physical inactivity in Canadian adolescents. Am J
Clin Nutr 2006;83(1):139-45.
Received: May 7, 2013
Accepted: October 29, 2013
Laura E. Forbes, PhD, [1] Shawn N. Fraser, PhD, [2] Shauna M.
Downs, MSc, [3] Kate E. Storey, PhD, [4] Ronald C. Plotnikoff, PhD, [5]
Kim D. Raine, PhD, [6] John C. Spence, PhD, [7] Rhona M. Hanning, PhD,
[8] Linda J. McCargar, PhD [9]
Author Affiliations
[1.] Department of Family Relations and Applied Nutrition,
University of Guelph, Guelph, ON
[2.] Faculty of Health Disciplines, Athabasca University,
Athabasca, AB
[3.] Menzies Centre for Health Policy, School of Public Health,
University of Sydney, Sydney, Australia
[4.] Department of Public Health Sciences, School of Public Health,
University of Alberta, Edmonton, AB
[5.] Priority Research Centre in Physical Activity and Nutrition,
Advanced Technology Centre, Callaghan, NSW
[6.] Centre for Health Promotion Studies, School of Public Health,
University of Alberta, Edmonton, AB
[7.] Faculty of Physical Education and Recreation, University of
Alberta, Edmonton, AB
[8.] School of Public Health and Health Systems, University of
Waterloo, Waterloo, ON
[9.] Department of Agricultural, Food and Nutritional Science,
University of Alberta, Edmonton, AB
Correspondence: Linda J. McCargar, Department of Agricultural, Food
and Nutritional Science, 4-126 Li Ka Shing Centre for Health Research,
University of Alberta, Edmonton, AB T6G 2E1, Tel: 780-492-9287, Fax:
780-492-2011, E-mail: Linda.McCargar@ualberta.ca
Table 1. Summary of Cut-offs Used for Assessing Diabetes
Risk Factors
Variable Cut-off = [up arrow] risk Reference
Overweight International Obesity Task (25)(26)(4)
and obesity Force (IOTF) Cut-offs for
Overweight and Obesity
Glycemic index >58 (5)
Glycemic load >142 (5)
Fibre <14.75 g (6)
Magnesium <222 mg/day females
<280 mg/day males (7)
Vegetables <2 servings/day
and fruit (>1.99 servings = not at risk) (8)
Fat intake >39% of calories (9)
Physical activity <3 score from the PAQ-C (18)(10)
Table 2a. Characteristics of Participants in 2005 and 2008
Variable 2005 2008
n = 4887 n = 5119
Age (years) 13.6 [+ or -] 1.2 13.3 [+ or -] 1.5
([dagger])
Income ($) 71,739 [+ or -] 19,635 74,797 [+ or -] 21,737
([dagger])
BMI (kg/[m.sup.2]) 20.9 [+ or -] 6.3 20.2 [+ or -] 4.1
([double dagger]) ([dagger])
Total energy 2077 [+ or -] 1101 1971 [+ or -] 1054
intake (kcal) ([dagger])
Carbohydrate 54.7 [+ or -] 11.4 55.2 [+ or -]11.4 *
(%kcal)
Fat (%kcal) 31.6 [+ or -] 9.2 30.5 [+ or -]9.3
([dagger])
Protein (%kcal) 15.3 [+ or -] 5.3 16.0 [+ or -]5.4
([dagger])
Glycemic index 55.4 [+ or -] 7.3 54.7 [+ or -] 7.4
([dagger])
Glycemic load 147 [+ or -] 81 140 [+ or -] 82
([dagger])
Fibre (g) 14.5 [+ or -] 9.8 15.2 [+ or -] 10.0
([dagger])
Magnesium (mg) 277 [+ or -] 158 283 [+ or -] 160
Vegetables and 5.1 [+ or -] 4.1 5.3 [+ or -] 4.4
fruit (svg/d) ([dagger])
PAQ-C Physical 2.9 [+ or -] 0.7 2.9 [+ or -] 0.7
Activity Score
* p < 0.05.
([dagger]) p < 0.01 for 2005 vs. 2008.
([double dagger]) BMI calculation was based on
self-reported height and weight.
Table 2b. Characteristics of Participants in 2005 and 2008
by Sex
Variable Males
2005 2008
n = 2264 n = 2313
Age (years) 13.6 [+ or -] 1.2 13.4 [+ or -]
1.6 ([dagger])
Income ($) 72,381 [+ or -] 20,167 75,307 [+ or -]
21,651 ([dagger])
BMI (kg/ 21.5 [+ or -] 7.8 20.4 [+ or -]
[m.sup.2]) 4.2 ([dagger])
Total energy 2398 [+ or -]1204 2297 [+ or -]
intake (kcal) 1204 ([dagger])
Carbohydrate 53.4 [+ or -] 11.4 53.8 [+ or -] 11.7
(%kcal)
Fat (%kcal) 32.2 [+ or -] 9.1 31.3 [+ or -]
9.4 ([dagger])
Protein (%kcal) 15.8 [+ or -] 5.4 16.4 [+ or -]
5.6 ([dagger])
Glycemic index 56.1 [+ or -] 7.4 55.2 [+ or -]
7.8 ([dagger])
Glycemic load 168 [+ or -] 90.0 162 [+ or -] 97 *
Fibre (g) 16.0 [+ or -] 10.8 16.6 [+ or -] 11.3
Magnesium (mg) 313 [+ or -] 172 320 [+ or -] 181
Vegetables and 5.3 [+ or -] 4.4 5.6 [+ or -] 5.0 *
fruit (svg/d)
PAQ-C 3.0 [+ or -] 0.7 3.0 [+ or -] 0.7
Variable Females
2005 2008
n = 2623 n = 2806
Age (years) 13.6 [+ or -] 1.2 13.3 [+ or -]1.5 ([double
dagger]) ([dagger])
Income ($) 71,250 [+ or -] 74,361 [+ or -]
19,160 ([double 21,785 ([dagger])
dagger])
BMI (kg/ 20.5 [+ or -] 20.1 [+ or -]4.0 ([double
[m.sup.2]) 4.6 ([section]) dagger]) ([dagger])
Total energy 1801 [+ or -] 1705 [+ or -]825
intake (kcal) 913 ([section]) ([section]) ([dagger])
Carbohydrate 55.8 [+ or -] 56.3 [+ or -]
(%kcal) 11.2 ([section]) 11.0 ([section])
Fat (%kcal) 31.0 [+ or -] 29.8 [+ or -] 9.2
9.2 ([section]) ([section]) ([dagger])
Protein (%kcal) 14.9 [+ or -] 15.7 [+ or -] 5.3
5.2 ([section]) ([section]) ([dagger])
Glycemic index 54.9 [+ or -] 54.3 [+ or -]7.1
7.1 ([section]) ([section]) ([dagger])
Glycemic load 128 [+ or -] 121 [+ or -]62
68 ([section]) ([section]) ([dagger])
Fibre (g) 13.2 [+ or -] 14.0 [+ or -]8.6
8.6 ([section]) ([section]) ([dagger])
Magnesium (mg) 247 [+ or -] 254 [+ or -]
138 ([section]) 134 ([section])
Vegetables and 4.8 [+ or -] 5.0 [+ or -]
fruit (svg/d) 3.9 ([section]) 3.9 ([section]) *
PAQ-C 2.8 [+ or -] 2.8 [+ or -]
0.6 ([section]) 0.7 ([section])
* p<0.05.
([dagger]) p<0.01 for 2005 vs. 2008.
([double dagger]) p<0.05, ([section]) p<0.01 for boys vs. girls.
Table 3a. Proportion of Participants With Diabetes Risk
Factors Between 2005 and 2008
Risk Factor 2005 2008
% (n) % (n)
Overweight 21.1 (870) 18.3 (706) ([dagger])
Obese 6.1 (252) 4.4 (170) ([dagger])
High GI 35.8 (1760) 31.7 (1620) ([dagger])
High GL 44.8 (2205) 40.1 (2062) ([dagger])
Low fibre 59.1 (2903) 56.4 (2882) ([dagger])
Low magnesium 48.8 (2373) 47.0 (2389)
Low vegetable and fruit 23.0 (1129) 21.2 (1084) *
High fat 19.5 (959) 17.4 (889) ([dagger])
Low physical activity 56.8 (2497) 55.9 (2278)
* p<0.05.
([dagger]) p<0.01 for 2005 vs. 2008.
Table 3b. Proportion of Participants With Diabetes Risk
Factors Between 2005 and 2008 by Sex
Risk Factor Males
2005 2008
% (n) % (n)
Overweight 26.6 (530) 21.1 (390) ([dagger])
Obese 7.8 (155) 4.8 (89) ([dagger])
High GI 39.0 (878) 35.2 (808) ([dagger])
High GL 56.3 (1269) 52.5 (1212)*
Low fibre 52.8 (1190) 50.4 (1155)
Low magnesium 47.9 (1079) 47.6 (1091)
Low vegetable 22.8 (514) 21.3 (487)
and fruit
High fat 21.8 (491) 19.8 (454)
Low physical 49.7 (991) 49.4 (881)
activity
Risk Factor Females
2005 2008
% (n) % (n)
Overweight 15.9 (340) ([double dagger]) 15.8 (316)
Obese 4.5 (97) ([double dagger]) 4.1 (81)
High GI 33.0 (863) ([double dagger]) 28.7 (801) ([dagger])
([double dagger])
High GL 35.1 (918) ([double dagger]) 30.0 (840) ([dagger])
([double dagger])
Low fibre 64.2 (1679) ([double dagger]) 61.2 (1711) *
([double dagger])
Low magnesium 49.5 (1294) 46.5 (1298) *
Low vegetable 23.0 (602) 21.1 (589)
and fruit
High fat 17.6 (459) ([double dagger]) 15.4 (430) *
([double dagger])
Low physical 62.9 (1485) ([double dagger]) 61.1 (1386)
activity ([double dagger])
* p<0.05.
([dagger]) p<0.01 for 2005 vs. 2008.
([double dagger]) p<0.01 for boys vs. girls.
Table 4. Influence of Differences in Age, WebSPAN Survey
Year, Sex, School Income and School Clustering on the
Total Number of Risk Factors
Effect Estimate Std. Error
Fixed effects
Intercept 3.19 ([dagger]) 0.04
Age 0.07 ([dagger]) 0.02
WebSPAN year -0.27 ([dagger]) 0.05
Sex -0.18 ([dagger]) 0.04
School income -0.0000026 * 0.0000012
([double dagger])
Random effects
Between-school variance 0.05 ([dagger]) 0.01
Within-school variance 2.13 ([dagger]) 0.04
95% Confidence Interval
Effect Lower Bound Upper Bound
Fixed effects
Intercept 3.12 3.28
Age 0.04 0.10
WebSPAN year -0.39 -0.18
Sex -0.25 -0.11
School income -0.0000050 -0.00000022
([double dagger])
Random effects
Between-school variance 0.03 0.09
Within-school variance 2.05 2.20
* p<0.05.
([dagger]) p<0.01.
([double dagger]) Effect of school income was determined
using income quintiles. Quintile ranges were: Q1:
46,288-56,287; Q2: 56,288-66,074; Q3: 66,075-76,809;
Q4 76,810-88,882; Q5 88,883-138,258.
