PLOTNIKOFF, RONALD C.1; TAYLOR, LORIAN M.2; WILSON, PHILIP M.3; COURNEYA, KERRY S.4; SIGAL, RONALD J.5; BIRKETT, NICHOLAS6; RAINE, KIM7; SVENSON, LAWRENCE W.8
The prevention and management of diabetes is a growing public heath concern in Canada (4,24) and worldwide (5,29). The U.S. Centers for Disease Control recently estimated that in the United States, the prevalence of diabetes in people over age 20 is 8.7%, with a higher incidence (18.3%) in people over age 60 (5), and that between 90 and 95% of diabetes cases are type 2 (5). The prevalence in Canada is slightly lower (5 and 13% for the respective age groups) (13). The worldwide prevalence of diabetes is projected to increase by 48% by the year 2030 (4). This global rise is associated with the dramatic increase in the prevalence of obesity, the adoption of a more sedentary lifestyle, and increasing population sizes in the aging population and high-risk ethnic groups (4,29).
The long-term complications of diabetes, including microvascular and macrovascular disease and neuropathy, can be delayed or even prevented with proper management (12). Much research now supports the beneficial role of physical activity in the prevention and management of diabetes (2,4,6,9,21). New clinical practice guidelines issued by the American (21) and Canadian Diabetes Associations (4) advise regular moderate activity, such as brisk walking and biking, to be performed for a minimum of 150 min accumulated over at least three nonconsecutive days of the week, although a higher level (≥ 240 min·wk−1) of accumulated physical activity is preferred (4,21). Given the potential health benefits from regular, moderate physical activity, it is concerning that the majority of Canadian adults with diabetes remain either sedentary or insufficiently active. Indeed, 60% of Canadians aged 35-64 yr and 76% over age 65 yr are sedentary/insufficiently active (12), which is higher than in the general population (53 and 58%, respectively) (23).
The determinants of physical activity participation in people with diabetes need to be identified before interventions can be targeted appropriately. Only one study (a retrospective cohort of 481 U.S. individuals aged 18-44 yr) was identified that specifically explored demographic and health-related correlates of physical activity in people with type 1 diabetes (14). Although combined gender results were not reported in this study, for males, higher activity levels were observed for those who were younger, smoked less, consumed more alcohol, and had fewer diabetes-related vascular complications (14). In contrast, only one factor (a lower level of diabetes-related complications) was associated with being more active in females (14). One further U.S. study reported no association between age, gender, and race with exercise behavior (no other factors were examined) in a small subsample of 83 adults with type 1 diabetes (22).
Four studies to date have reported on associations of physical activity levels with demographic and health factors in adults with type 2 diabetes. Briefly, Hays (11) investigated 260 older adults (> 55 yr); Nothwehr (16) surveyed 733 individuals aged 50-62 yr; Nelson (15) reported results of a subsample of 1480 adults with type 2 diabetes from the nationally representative National Health and Nutrition Examination Survey (NHANES III); and Spangler (22) published results from a convenience sample of 322 participants. The results from these four methodologically diverse U.S. studies are inconsistent (11,15,16,22). Older age (11,15) and being female (15,16) were reported to be negatively associated with physical activity; however, in the other studies, age (16,22) and sex (11,22) had no significant relationship with physical activity. No significant associations were found for ethnicity in three studies (11,15,16) However, Spangler identified Caucasians were more active compared with blacks in the population they examined (22). In the two studies that examined education level, Hays reported that people with no high school education were more likely to be inactive (11), whereas Nelson did not show any significant relationship between education and physical activity (15). When examining income, Nelson identified that a low income was negatively associated with physical activity (15), whereas Hays did not reveal a significant association between these two factors (11). Further, no significant relationship between body mass index and physical activity was reported in the two studies that examined this potential correlate (11,15). The lack of consistency among the study results and the limited factors investigated underscores the need to further examine the correlates of physical activity in adults with type 2 diabetes.
To address this issue, we conducted the Alberta Longitudinal Exercise and Diabetes Research Advancement (ALEXANDRA) study. The ALEXANDRA study is a longitudinal study examining the social cognitive determinants of physical activity of individuals with type 1 and type 2 diabetes over an 18-month period. Here we present results on the baseline measures of demographic and health factors associated with physical activity levels in this population. We believe that the knowledge gained by identifying demographic and health factors associated with physical activity in the adult diabetes population will guide us in designing effective population-based programs.
Sample and procedure.
Study participants (18 yr and older) were recruited using two strategies. All participants completed identical baseline self-administered study instruments during May 2002 and provided written informed consent.
The first recruitment strategy identified members from the Canadian Diabetes Association registry, limited to members living in the province of Alberta. Postage-paid return questionnaires (N = 4609) and consent forms were mailed with the association's newsletter to all members, requesting completion by those members with diabetes. A follow-up postcard reminder was sent 12 d after the initial mailing.
The second recruitment strategy used a random digit-dialing (RDD) method developed by a university-based population research lab (PRL) to obtain a final list of 601 individuals with diabetes living in the province of Alberta. This second strategy allowed us to 1) augment the overall study sample, and 2) examine, in part, the representativeness of the registry respondents (first recruitment strategy) by comparing results with this randomly selected sample (i.e., RDD protocol). A quota sample of 300 individuals with diabetes was recruited from the RDD contacted households (list 1). (There were 1601 initial refusals from contacted eligible individuals.) In addition, households were permitted to nominate a family member or friend with diabetes who did not live with them. These additional names (list 2, quota sample of 301) were also telephoned by PRL. (There were 70 initial refusals from these eligible individuals.) Once contact by telephone was established with the individuals with diabetes in both RDD protocols (N = 601), a postage-paid return questionnaire and consent form were mailed. Up to two telephone reminder calls were made to nonrespondents 10 d after the mail-out.
To avoid any possible respondent duplication between the recruitment strategies, participants were instructed to sign up only once, in the event they were already contacted by another strategy. The study received ethical approval from a university ethics review board.
Demographic factors were measured employing questions based on the Statistics Canada 2001 census (24) and included age, gender (females = 0; males = 1), ethnic origin, marital status, educational level (0 = no university degree; 1 = university degree), gross annual family income, and employment status.
The following self-reported health factors were assessed (19): diabetes type (type 1 or 2 as been told by a doctor or nurse); height and weight to calculate body mass index; level of perceived disability influencing physical activity participation (rated on a six-point scale ranging from "no limitation = 0" to "complete limitation = 5"); daily use of insulin and oral antihyperglycemic medication (with "yes = 0"/"no = 1" response options to these two types of medication); smoking behavior status (currently smoking = 0 or not currently smoking = 1); heart disease status (assessed as having ever been told by a doctor or healthcare professional of having heart disease, with a "yes = 0"/"no = 1" response option); and blood pressure and cholesterol levels (also assessed as having ever been told by a doctor or healthcare professional of having i) high cholesterol and ii) high blood pressure, respectively, with "yes = 0"/"no = 1" response options for each question).
Physical activity was measured using the Godin Leisure-Time Exercise Questionnaire (GLTEQ) (10). Participants were asked to report the average number of times per week over the past month they had engaged in strenuous (heart beats rapidly, sweating), moderate (not exhausting, light perspiration), and mild (minimal effort, no perspiration) physical activity. The GLTEQ was slightly modified (7) to have respondents also specify the average time per session (minimum of 10 min per session) for each of the three levels of intensity.
The modified GLTEQ was used to measure physical activity for this study in two different ways. First, for the study's multiple regression models, minutes of participation in mild, moderate, and vigorous activity were measured on a continuous scale. Participant responses in each of these three activity categories were then converted to MET-minutes by multiplying the weekly minutes of mild activity by 2.5 METs (22), moderate activity by 4.0 METs (3), and weekly minutes of vigorous activity by 7.5 METs (3).
Second, for the study's logistic regression models, participants were categorized as "active" or "inactive" based on public health (17) and diabetes-specific (4,21) guidelines of achieving moderate activity at least 150 min of physical activity per week. The number of minutes was computed by multiplying the frequency and duration of i) weekly minutes of moderate physical activity × 4.0 METs, and ii) weekly minutes of vigorous physical activity × 7.0 METs. The weekly minutes for moderate and vigorous were then summed for a total MET score. One minute of vigorous physical activity is equivalent to 1.875 min of moderate activity (7.5/4.0) based on the average MET levels for strenuous activity (MET level = 7.5) and moderate activity (MET level = 4.0) set by Brown and Bauman (3). This weighting provides more credit for participating in vigorous activity. Individuals who accumulated ≥ 600 MET·min·wk−1 were classified as "adequately active for health benefit," whereas those who did not were classified as "inadequately active" (3). This criterion reflects achieving 150 min of moderate (4.0 METs) or 80 min of vigorous (7.5 METs) activity per week, or any combination thereof (3)."
Procedures and analyses.
Data were initially screened for discrepant responses, missing data, and multicollinearity problems in accordance with previous recommendations (10). Descriptive statistics were calculated for all study variables. Imputation of missing values for categorical variables used the hot deck approach as described by Cox and Cohen (10), using imputation classes based on gender (male and female), diabetes type (type 1 and type 2), and age (18-25, 26-35, 36-45, 46-60, 61-70, or 71-92 yr). Imputation of missing variables for continuous variables used mean value imputation. Only three variables had greater than 5% missing data: heart disease status (15.6%), cholesterol level (8.7%), and income (10.5%).
Regression models were calculated to detect demographic and health factors that were significantly associated with total physical activity. Due to the different etiology and pathology of type 1 and type 2 diabetes (4), the two types of diabetes were examined separately.
A series of simultaneous multiple regression analyses were performed separately for the type 1 and type 2 groups to assess the relationship between demographic and health factors and physical activity (including light, moderate, and vigorous levels). For all analyses, age, body mass index, income, and perceived disability index were treated as continuous measures; all other factors were categorical variables (see coding note in Table 3). The first regression analysis (the demographic model) included only the demographic factors: age, gender, marital status, education, and income. The second analysis (the health factor model) included only the health factors: body mass index, smoking status, perceived disability, heart disease status, cholesterol level, and blood pressure level. For the type 2 diabetes analysis, the use of insulin and oral antihyperglycemic medications were included in the model. Demographic and health factors demonstrating a significant relationship with physical activity from both the demographic and the health factor models were then entered into a third regression model (the combined model). The fourth model (the full model) included all demographic and health factors.
Logistic regression models examined the above demographic and health factors associated with adherence to current physical activity recommendations (4,17,21) (which include moderate and vigorous intensity levels only). Odds ratios (OR) and confidence intervals (CI) were reported for significant (P < 0.05) factors for both type 1 and type 2 groups.
Eight cases were excluded from the analyses, as they did not indicate they had diabetes and their diagnosis could not be confirmed from their medication history. Values for body mass index and MET-minutes with reported standard deviations > 3.29 (22 and 64 cases, respectively) were truncated to 3.29 standard deviations from the mean to reduce the impact of outliers (25). Collinearity diagnostics indicated no important violations among the variables in this study (VIF = 1.06-1.48 and tolerance = 0.68-0.94) (25).
Recruitment strategy 1 obtained completed questionnaires from 1923 individuals (609 type 1; 1307 type 2; 7 missing diabetes type) from the Canadian Diabetes Association Alberta Registry (4609 adults were listed in the registry). The response rate for this sample could not be determined because the association's membership includes an unknown number of friends, family members, healthcare practitioners, and researchers who do not have diabetes. The second recruitment strategy obtained completed questionnaires from 396 of the 600 individuals who agreed to participate in the study (66%: 88 type 1, 307 type 2, and 1 missing diabetes type). Of this group, 206 (52%) were directly contacted in the random digit-dialing (RDD) protocol, whereas 190 (48%) were referred by family/friends in the RDD protocol. There were no statistically significant differences between the recruitment strategies for sex or physical activity levels (assessed by MET-minutes) for the combined diabetes types. Further, there were no statistically significant differences between the recruitment strategies for the type 1 group for age, body mass index, and MET-minutes. Although a marginal difference was reported for the type 2 group on mean age (59.2 ± 12.0 vs 63 ± 12.0 yr) and mean body mass index (31.2 ± 6.1 vs 29.2 ± 5.6) for the RDD and registry recruitment protocols, respectively, MET-minutes scores were not significantly different. Therefore, we combined the recruitment groups for the study analyses.
The demographic characteristics of our study generally reflect Canada's diabetic population in terms of age and sex distributions (13). Demographic health and physical activity characteristics of the combined sample are shown in Table 1 and revealed the following significant (P < 0.05) differences between the type 1 and type 2 groups. Compared with those with type 2 diabetes, people with type 1 diabetes were younger, had a higher educational achievement, and had greater income. As expected, 100% of people with type 1 diabetes were using insulin (compared with only 21.3% in type 2), whereas 67% of people with type 2 diabetes used oral antihyperglycemic medication. People with type 1 diabetes had a lower body mass index and lower perceived disability scores. The prevalence of self-reported heart disease, elevated cholesterol, and hypertension were also lower in those with type 1 diabetes. As the sample was primarily Caucasian in origin (type 1, 92.7%; type 2, 89.9%), ethnicity was not included in the analyses.
People with type 1 diabetes reported higher levels of physical activity (Table 1), although only 36.3% of this group met the current public health recommendations (achieving ≥ 600 MET· min·wk−1) (17). Adherence to recommended physical activity levels was somewhat lower in women and in people with type 2 diabetes, and decreased with age for type 1 individuals (Table 2). Because recommendations for physical activity can be expressed in terms of either MET-minutes or "time on task" (i.e., 150 min·wk−1), we examined the concordance of adherence determined using these two measures (Table 2). There was good agreement in the proportion meeting the recommendations using the two criteria (kappas > 0.80), except for men under age 45 yr, where adherence was better when based on the MET-minutes criterion, perhaps because these younger men undertake higher-intensity exercise than other groups.
Multiple regression models.
The results of the multiple regression models are reported in Table 3, which provides standardized beta coefficients and significance levels.
Type 1 diabetes.
The demographic model found that higher levels of physical activity were associated with younger age (β = −0.21, P < 0.001), male gender (β = 0.07, P < 0.05), being single (β = −0.11, P < 0.01), and higher income (β = 0.15, P < 0.001) after controlling for the other demographic factors. The variance accounted for by the demographic model was 9% (P < 0.001). The health factor model found that higher levels of physical activity were significantly associated with lower levels of perceived disability influence (β = −0.21, P < 0.001), not having elevated cholesterol (β = 0.09, P < 0.05), and not having elevated blood pressure (β = 0.10, P < 0.05) after controlling for the other health factors. The health factor model accounted for 10% (P < 0.001) of the variance in physical activity.
The combined model found that higher levels of physical activity were significantly associated with a lower age (β = −0.12, P < 0.01), being single (β = −0.11, P < 0.01), a higher income (β = 0.11, P < 0.01), and a lower level of perceived disability (β = −0.19, P < 0.001). Being male, cholesterol level, and blood pressure level no longer contributed significantly to the model. The five factors accounted for 14% (P < 0.001) of the variance. The full model also explained 14% (P < 0.001) of the variance, and the same factors were significant between the combined model and the fourth regression model.
Type 2 diabetes.
For participants with type 2 diabetes, the demographic model demonstrated an association between higher physical activity levels with being male (β = 0.16, P < 0.001), higher education level (β = 0.05, P < 0.05), and higher income level (β = 0.08, P < 0.01) after controlling for the other demographic factors. The variance accounted for by these demographic factors was 5% (P < 0.001). The health factor model found that higher levels of physical activity were significantly associated with lower body mass index (β = −0.10, P < 0.001) and lower levels of perceived disability (β = −0.22, P < 0.001) after controlling for the other health factors. The health factor model accounted for 7% (P < 0.001) of the variance in physical activity.
The combined model found that higher levels of physical activity were associated with being male (β = 0.12, P < 0.001), higher education level (β = 0.05, P < 0.05), higher income level (β = 0.06, P < 0.05), lower body mass index (β = −0.09, P < 0.001), and lower level of perceived disability (β = −0.19, P < 0.001). The six variables in the regression accounted for 10% of the variance. The full model also explained 10% (P < 0.001) of the variance and included younger age (β = −0.07, P < 0.05), being male (β = 0.13, P < 0.01), lower BMI (β = −0.11, P < 0.01), and level of perceived disability (β = −0.18, P < 0.001).
Logistic regression models.
The logistic regression analyses in Table 3 demonstrate that in adults with type 1 diabetes, being sufficiently active was significantly associated with younger age (OR = 0.98, 95% CI = 0.96-0.99, P < 0.001), higher income (OR = 1.14, 95% CI = 1.01-1.30, P < 0.05), not smoking (OR = 2.75, 95% CI= 1.36-5.53, P < 0.05), and lower level of perceived disability (OR = 0.72, 95% CI = 0.63-0.82, P < 0.001). The variance accounted for in this model was 22% (Nagelkerke R2 = 0.22). These results contrast with the combined model multiple linear regression results, which did not identify a relationship between smoking and activity level but did identify an association between being single and higher levels of physical activity.
For those with type 2 diabetes, the logistic regression results demonstrate that being sufficiently active was significantly associated with younger age (OR = 0.98, 95% CI = 0.97-0.99, P < 0.001), being male (OR = 1.37, 95% CI = 1.07-1.75, P < 0.05), lower BMI (OR = 0.95, 95% CI = 0.93-0.97, P < 0.001), and lower level of perceived disability (OR = 0.77, 95% CI = 0.71-0.83, P < 0.001). The variance accounted for in this model was 13% (Nagelkerke R2 = 0.13). The results of this logistic regression model parallel those of the full model multiple linear regression results.
The purpose of the present study was to identify demographic and health factors associated with physical activity in adults with type 1 and type 2 diabetes. The results of this study provide one of the more comprehensive examinations on this topic to date, with data presented for both type 1 and type 2 diabetes in a large population-based sample.
Our results indicate that most adults with diabetes do not regularly engage in adequate physical activity to achieve health benefits; the majority of both type 1 and type 2 participants were considered insufficiently active (63.7 and 71.9%, respectively). This is consistent with findings from other research. For example, the Canadian National Population Health Survey reported that approximately 60% of people with diabetes aged 35-64 yr and 76% of those over the age of 65 yr were inactive (12). Considering the numerous health benefits to be gained from regular physical activity for adults with diabetes (4), the above poor participation rates attest to the need for developing effective population health programs to increase physical activity.
The results of our findings for the type 1 group parallel those of the research conducted by Moy (14) in that higher levels of perceived disability (which could be related to diabetes-related complications) were negatively associated with physical activity. Further, we report that older age and smoking behavior (from our logistic regression analyses) were also negatively related with physical activity, which is consistent with Moy's stratified gender results for males (14). Further, higher income levels and being single in our study were significantly associated with higher physical activity levels; the relationship of these factors with physical activity has not been previously studied in people with type 1 diabetes. Finally, we did not find consistent significant associations across our analyses between body mass index, gender, heart disease status, and cholesterol and blood pressure levels with physical activity. Apart from Spangler's (22) small subsample of 88 type 1 individuals, which also found no relationship between gender and activity, these factors have not, to date, been examined in the type 1 population.
Our results support previous studies in adults with type 2 diabetes that identified an association between higher physical activity levels and being male (15,16) and younger age (15,11). However, in other examinations of these factors, gender (11,22) and age (16,22) were found not to be significantly related to physical activity. Our study reported significant modest positive associations in the multiple regression models between higher education and income levels with physical activity, which is consistent with the studies by Hay (for education) (11) and Nelson (for income) (15). In addition, our data revealed significant negative associations between body mass index and physical activity, which was not found in two other studies (11,15). The lack of association between body mass index and physical activity in these two previous studies is perhaps due to the nongeneralization in one study (11) and the classification of BMI as a categorical versus continuous variable in the other study (15). We also reported that daily use of insulin and oral antihyperglycemic medication were both not associated with activity level, which is consistent with the only other study (15) that examined these potential medication correlates. Further, the level of perceived disability showed a significant strong negative relationship with physical activity in the present study, which has not been previously studied with type 2 individuals. Finally, heart disease status, along with cholesterol and blood pressure levels, had no association with physical activity, which, to date, has also not been examined in the literature with this population.
Epidemiological research has identified an array of demographic and health factors that are correlated with adult physical activity participation in the general population (26). Trost et al. (26) summarized these key factors from 338 studies. Age, overweight, and obesity are consistently reported to have a negative association with physical activity; education level, male gender, and income level are consistently reported to have a positive association with physical activity, and a weak negative association is reported for being married and smoking (26). This contrasts with the research presented here because no significant association between body mass index and physical activity was evident in those with type 1 diabetes. The lack of association between these two factors may, in part, be explained by the fact that people with type 1 diabetes who are physically active frequently overeat to avoid hypoglycemia (27,30). Further, the relationship between male gender and physical activity was not overtly present in our type 1 group analyses, which also contradicts the findings for the general population (26). Diabetes causes greater relative increases in cardiovascular risk in women than in men, and this is particularly true in younger, premenopausal women (28). We speculate that many type 1 diabetic women may be particularly conscious of this, and might, therefore, make extra efforts to minimize their cardiovascular risks, including increased efforts to exercise regularly. Our results for those with type 2 diabetes are consistent with correlates identified in the general population (26,20), with the exception of smoking; this correlate did not demonstrate a significant relationship with physical activity in the current study.
The results of our study between the type 1 and type 2 groups were generally similar on most of the characteristics, suggesting that, for the most part, interventions may be generically appropriate to both types. A major difference between the two diabetes types, however, was the significant negative association of body mass index with physical activity in the type 2 group. This result was expected because of the etiology of type 2 diabetes (4). Other contrasting results between the two diabetes types included the stronger association of physical activity with gender (males) in the type 2 group, and martial status (being single), not smoking, and normal cholesterol and blood pressure levels in the type 1 group. These differences should be taken into account when tailoring diabetes programs.
The lack of consistency among the previous studies examining the demographic and health correlates in the diabetes population may be due to the unique and widely diverse methods employed in terms of recruitment and sampling design, sample characteristics, chosen factors to examine, validity and reliability of the employed measures, and analytical procedures. It should be noted that our study did not have significant black and Hispanic populations as reported as in the existing U.S. studies examining demographic and health factors associated with physical activity in the diabetes population (11,15,16,22). Further, none of the existing studies with the type 2 population assessed or adjusted for perceived disability, which was found to be a very strong correlate in the present study. Regardless, our type 2 results most closely reflect the NHANES III findings of the study's 1460 type 2 individuals (15), which is arguably the most methodologically rigorous sampling frame to date to accurately represent a national population of adults with type 2 diabetes. However, our study's numerous regression models to examine the association of potential demographic and health factors and accounting for perceived physical disability are further strengths in light of the existing limited literature in this domain. Further, this appears to be one of the first studies to have examined demographic and health characteristics of physical activity of adults with diabetes outside of the United States.
Our findings also highlight some key considerations for health practitioners. Perceived level of disability that would limit physical activity participation was a strong negative correlate of physical activity in those with type 1 and type 2 diabetes. This finding highlights the need to individualize physical activity programs around the person's specific limitations. For example, low-impact, moderate-intensity activities, such as walking, swimming, or stationary bicycling, which can be done by individuals whose ability to be active is already limited, may be worthy of consideration. Indeed, a significant proportion of those with diabetes in this study (of which the majority report some perceived disability influencing physical activity participation) report participating in one or more of these activity modes (18).
Researchers and practitioners have become increasingly interested in understanding the factors promoting physical activity among people with diabetes (19). The overall findings of this study suggest that specific demographic and health factors are able to account for a statistically significant portion of the variance associated with physical activity in those with type 1 or type 2 diabetes. Although our study investigated a number of potential demographic and health correlates of physical activity, other potential factors (e.g., duration of diabetes, knowledge of physical activity and other lifestyle recommendations) were not examined. Despite this, certain population groups, such as those with a perceived disability, females, those of older ages, those of lower income and education levels, and those with a high body mass index (for those with type 2) may be most in need of physical activity interventions. Healthcare providers may therefore need to reconsider the way they currently offer services to increase the success of their programs and services. Strategies should promote physical activities in community environments that are affordable, equally attractive to females and males, and provide a range of activities appropriate for people with disabilities and who are obese. Strategies should also be tailored to both diabetes types whenever possible. However, given the low overall participation rates, governments, healthcare organizations, and practitioners need to promote participation in regular physical activity in accordance with current guidelines (4) to all those with diabetes. Canada's healthcare system has the potential to reach significant proportions of the population living with diabetes considering the country's health and medical structure, which is accessible and free to all its citizens and residents.
In conclusion, our findings indicate that the majority of people with diabetes in our sample are insufficiently active based on current public health guidelines and require effective programs to promote physical activity. Our study also reveals that demographic and health factors for both type 1 and type 2 adults account for a significant proportion of the variance for physical activity in both diabetes types. Hopefully, the reported significant demographic and health factors will guide such strategies in identifying and targeting segments of the population who are the least physical active. However, it is necessary to continue advancing our understanding of demographic, psychosocial, and environmental factors for the development and testing of tailored physical activity interventions in people with diabetes. Indeed, current research from this cohort will broaden the scope of variables by examining the collected psychosocial and environmental characteristics thought to influence physical activity patterns toward a comprehensive understanding of the factors and mechanisms of physical activity behavior change in this population.
This study was funded by the Alberta Heritage Foundation for Medical Research and the Canadian Diabetes Association. Dr. Ronald Plotnikoff is supported on awards from the Alberta Heritage Foundation for Medical Research (AHFMR) and the Canadian Institutes for Health Research. Lorian Taylor is supported with a Health Studentship Award from AHFMR. Dr. Kerry Courneya is supported by the Canadian Research Chair Program. Dr. Ronald Sigal is supported by the OHRI Research Chair in Lifestyle Research.
1. Ainsworth, B. E., W. L. Haskell, M. C. Whitt, et al. Compendium of physical activities: an update of activity codes and MET intensities. Med. Sci. Sports Exerc.
2. Boule, N. G., E. Haddad, G. P. Kenny, G. A. Wells, and R. J. Sigal. Effects of exercise on glycemic control and body mass in type 2 diabetes mellitus: a meta-analysis of controlled clinical trials. JAMA
3. Brown, W. J., and A. E. Bauman. Comparison of estimates of population levels of physical activity using two measures. Aust. N. Z. J. Public Health
4. Canadian Diabetes Association Clinical Practice Guidelines Expert Committee. Canadian Diabetes Association 2003 Clinical Practice Guidelines for the Prevention and Management of Diabetes in Canada. Can. J. Diabetes
5. Centers for Disease Control: National Diabetes Fact Sheet: National Estimates and General Information on Diabetes in the United States. Atlanta, GA, US Department of Health and Human Services, Centers for Disease Control, pp.1-8, 2003.
6. Church, T. S., Y. J. Cheng, C. P. Earnest, et al. Exercise capacity and body composition as predictors of mortality among men with diabetes. Diabetes Care
7. Courneya, K. S., L. W. Jones, R. E. Rhodes, and C. M. Blanchard. Effects of different combinations of intensity categories on self-reported exercise. Res. Q. Exerc. Sport
8. Cox, B. G., and S. B. Cohen. Methodological Issues for Health Care Surveys. New York: Marcel Dekker, Inc., pp. 1-350, 1985.
9. Diabetes Prevention Program Research Group. Within-trial cost-effectiveness of lifestyle intervention or metformin for the primary prevention of type 2 diabetes. Diabetes Care
10. Godin, G., and R. J. Shephard. A simple method to assess exercise behavior in the community. Can. J. Appl. Sport Sci.
11. Hays, L. M., and D. O. Clark. Correlates of physical activity in a sample of older adults with type 2 diabetes. Diabetes Care
12. Health Canada. Diabetes in Canada: National Statistics and Opportunities for Improved Surveillance, Prevention and Control
. Ottawa, Ontario, Health Canada, pp.1-69, 1999.
13. Health Canada. Responding to the challenge of diabetes in Canada: First report of the national diabetes surveillance system (NDSS)
. Ottawa, Ontario, Health Canada, pp.1-122, 2003.
14. Moy, C. S., T. J. Songer, R. E. LaPorte, et al. Insulin-dependent diabetes mellitus, physical activity, and death. Am. J. Epidemiol.
15. Nelson, K. M., G. Reiber, and E. J. Boyko. Nhanes, III: Diet and exercise among adults with type 2 diabetes: Findings from the third national health and nutrition examination survey (NHANES III). Diabetes Care
16. Nothwehr, F., and T. Stump. Health-promoting behaviors among adults with type 2 diabetes: findings from the Health and Retirement Study. Prev. Med.
17. Pate, R. R., M. Pratt, S. N. Blair, et al. Physical activity and public health. A recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA
18. Plotnikoff, R. C. Physical activity in the management of diabetes: population-based perspectives and strategies. Can. J. Diabetes
19. Plotnikoff, R. C., S. Brez, and S. B. Hotz. Exercise behavior in a community sample with diabetes: Understanding the determinants of exercise behavioral change. Diabetes Educ.
20. Plotnikoff, R. C., A. Mayhew, N. Birkett, C. A. Loucaides, and G. Fodor. Age, gender, and urban-rural differences in the correlates of physical activity. Prev. Med.
21. Sigal, R. J., G. P. Kenny, D. H. Wasserman, and C. Castaneda-Sceppa. Physical activity/exercise and type 2 diabetes. Diabetes Care
22. Spangler, J. G., and J. C. Konen. Predicting exercise and smoking behaviors in diabetic and hypertensive patients. Age, race, sex, and psychological factors. Arch. Fam. Med.
23. Statistics Canada. Physical Activity, By Age Group and Sex, Household Population Aged 12 and Over
. Ottawa, Ontario, Statistics Canada, 2000-2001, CANSIM Table 105-0033.
24. Statistics Canada. Census 2001-2B
. Ottawa, Ontario, Health Canada, pp.1-32, 2001.
25. Tabachnick, B. G., and L. S. Fidell. In: Using Multivariate Statistics
. Needham Heights, MA: Allyn and Bacon, pp. 56-176, 2001.
26. Trost, S. G., N. Owen, A. E. Bauman, J. F. Sallis, and W. Brown. Correlates of adults' participation in physical activity: review and update. Med. Sci. Sports Exerc.
27. Wallberg-Henriksson, H., R. Gunnarsson, J. Henriksson, P. DeR. Fronzo, J. Felig, et al. Increased peripheral insulin sensitivity and muscle mitochondrial enzymes but unchanged blood glucose control in type I diabetics after physical training. Diabetes
28. Wingard, D. L., and E. Barrett-Connor. Heart disease and diabetes. In: Diabetes in America
, 2nd ed. M. I. Harris (Eds.). Bethesda, MD: National Institutes of Health, pp. 429-448, 1995.
29. World Health Organization. Diabetes Estimates and Projections
. World Health Organization. 2003. Fact Sheet Number 236.
30. Zinman, B., S. Zuniga-Guajardo, and D. Kelly. Comparison of the acute and long-term effects of exercise on glucose control in type I diabetes. Diabetes Care