OBJECTIVE: The aim of this study was to characterize the physical activity (PA) and electronic media (EM) use habits of a population-based, ascertained sample of youths with diabetes mellitus (DM).
METHODS: For this investigation, the Search for Diabetes in Youth Case-Control study (age: 10–20 y; 55% female) recruited 384 youths with provider-diagnosed type 1 DM, 90 youths with type 2 DM, and 173 healthy control subjects between 2003 and 2006, in 2 US centers. PA and EM use were assessed with a 3-day recall of activities, in 30-minute time blocks. Adherence to current recommendations was determined as a report of ≥2 blocks of moderate/vigorous PA per day and <4 blocks of EM use per day. Differences in PA and EM use for DM/control groups were assessed with adjustment for age, study site, and race/ethnicity.
RESULTS: Male subjects with type 2 DM reported lower levels of vigorous PA than did control subjects (1.1 vs 2.3 blocks; P < .05). Compliance with the moderate/vigorous PA recommendation among youths with type 2 DM was lower (68.3%), compared with youths with type 1 DM (81.7%; odds ratio: 0.51 [95% confidence interval: 0.26–1.00]; P = .047) and control subjects (80.4%; odds ratio: 0.48 [95% confidence interval: 0.23–1.02]; P = .05). Rates of compliance with EM use recommendations ranged from 29.5% to 49.1%.
CONCLUSION: In this study, large proportions of youths with DM, especially type 2 DM, failed to meet PA and EM use recommendations.
WHAT'S KNOWN ON THIS SUBJECT:
Among children and youths, increased PA and reduced sedentary behavior contribute to glycemic control, improved body composition, and a more-favorable CVD risk profile and thus constitute a cornerstone in the management of DM in pediatric populations.
WHAT THIS STUDY ADDS:
Large proportions of youths with DM fail to meet current PA and EM standards. Youths with type 2 DM engage in less PA and devote more time to EM use than do control subjects and youths with type 1 DM.
Diabetes mellitus (DM) constitutes the third leading chronic disease in the pediatric population, with an estimated prevalence of 1.82 cases per 1000 youths in the United States.1 Pediatric DM is associated with important health and economic burdens.2,3 Youths with DM have a lower life expectancy4 and a reduced quality of life5 and are at increased risk for early cardiovascular disease (CVD) complications.6,7
Glucose level-lowering therapy, diet, and physical activity (PA) constitute the cornerstones of DM management.8 For adults, there is increasing evidence regarding the health benefits of PA, including acute and chronic glucose level-lowering effects, improved insulin sensitivity, modulation of CVD risk, and reductions in depression and all-cause, cancer, and CVD mortality rates.8,9 For children and adolescents, evidence regarding the beneficial effects of PA is much more limited but generally supports the contributions of PA to metabolic control, improved body composition, and a more favorable CVD risk profile.10,–,12 In fact, current guidelines for the management of DM in the pediatric population recommend increasing daily PA and reducing sedentary behavior as important components of disease management.8,13,14 However, very few epidemiological studies have described the PA levels of current generations of children and adolescents with DM.15 Furthermore, estimates of the time devoted to sedentary behaviors such as television watching or overall use of electronic media (EM), which are considered independent risk factors for childhood obesity and cardiometabolic consequences,16,17 are scarce, especially for youths with type 2 DM.
The Search for Diabetes in Youth Case-Control (SEARCH-CC) study was designed to evaluate selected risk factors for childhood DM and included detailed self-report measures of PA and sedentary behaviors. The aim of this study was to characterize the PA and EM use habits of a population-based sample of youths with type 1 and type 2 DM and to contrast them with a group of healthy control subjects.
Data for this analysis were derived from the SEARCH-CC study, an ancillary study to the Search for Diabetes in Youth (SEARCH) study.18 In brief, the SEARCH study is a multicenter, multiethnic study conducting population-based ascertainment of nongestational cases of diagnosed DM in youths <20 years of age.
Recruitment of Youths With DM
Between July 2003 and March 2006, SEARCH study case subjects ≥10 years of age were invited to participate in the SEARCH-CC study at the Colorado and South Carolina centers. DM cases were identified by using a network of health care providers, including pediatric endocrinologists, hospitals, and other providers. The type of DM was established on the basis of provider diagnosis. By using Health Insurance Portability and Accountability Act–compliant procedures, youths with DM identified through the SEARCH study recruiting network were invited to the SEARCH study visit, which involved questionnaires, a brief physical examination, and laboratory measurements. At that time, data unique to the SEARCH-CC study also were collected, including an in-depth PA assessment, a perinatal questionnaire, and data on psychosocial factors.
Recruitment of Youths Without DM
Healthy control individuals were recruited from primary care offices in the same geographic areas. Participating primary care offices provided an initial study brochure, and patients and their parents or guardians were asked to complete a 1-page form that included limited demographic information and an indication of permission for study staff members to contact them regarding participation in the study. All control individuals were confirmed as nondiabetic on the basis of fasting glucose levels obtained at the SEARCH-CC study visit.
PA and EM Use Habits
PA levels were assessed by using the 3-Day Physical Activity Recall (3DPAR), a self-administered questionnaire addressing activities on the previous 3 days of the week, beginning with the most-recent day.19 The instrument was validated previously19,20 and includes a list of ∼80 common, physically demanding or sedentary activities, grouped into broad categories to improve activity recall. For every 30-minute time increment (block), participants reported the main activity performed and then rated its relative intensity (vigorous, moderate, light, or sedentary). Each activity was assigned a metabolic equivalent (MET) by using the Compendium of Physical Activities.21 Data from each day were reduced to the number of blocks of vigorous PA (VPA) (≥6 METs), moderate/vigorous PA (MVPA) (≥3.0 METs), moderate PA (≥3.0 but <6 METs), and television/video watching and an index of overall EM exposure (combined blocks of television/video watching, video gaming, and Internet use). Finally, a 3-day average was calculated for each of these variables.
Youths who reported participating in an average of ≥1 block of VPA per day or ≥2 blocks of MVPA per day were considered to meet PA standards for adolescents.22 For both television viewing and EM use, youths who reported an average of <4 blocks per day were considered to meet the standard, following the recommendations of the American Academy of Pediatrics23 and Healthy People 2010.24
Sociodemographic characteristics and medical history, including family history of DM, parental education, household income, and race/ethnicity, were obtained through questionnaire. Standardized physical examinations were conducted by trained, certified staff members. Height was measured in centimeters by using a stadiometer. Weight was measured in kilograms by using an electronic scale. Height and weight measurements were used to calculate BMI. Age- and gender-specific BMI z scores were derived on the basis of Centers for Disease Control and Prevention national standards, and weight status categories were assigned as overweight for individuals in 85th to 94th percentiles and obese for individuals in ≥95th percentiles.25 Waist circumference was measured at the end of normal expiration at the uppermost border of the ilium, with a standard anthropometric measuring tape, according to the National Health and Nutrition Examination Survey protocol. Blood was drawn after ≥8 hours of fasting, for measurement of glycohemoglobin levels. Hemoglobin A1c levels were measured in whole blood with an automated, nonporous, ion-exchange, high-performance liquid chromatography system (model G-7 [Tosoh Bioscience, Montgomeryville, PA]). Specimens were processed at the site and shipped within 24 hours to the Northwest Lipid Metabolism and Diabetes Research Laboratories (Seattle, WA), which served as the study central laboratory, for analyses.
Before implementation of the protocol, the study was reviewed and approved by the local institutional review boards that had jurisdiction over the local study population. Before the study visit, written informed consent was obtained for case subjects and control subjects according to the guidelines established by the local institutional review boards, from individuals who were ≥18 years of age or from parents or guardians for subjects who were <18 years of age. Written assent was obtained from individuals who were <18 years of age, as governed by local institutional review board instructions.
Final Subject Inclusion and Exclusion
The present analyses included 384 youths with type 1 DM, 90 youths with type 2 DM, and 173 control youths without DM, that is, a total of 647 (71%) of 910 eligible individuals. Exclusions included lack of 3DPAR data (n = 241), systematic errors in the 3DPAR (n = 13), and extreme/implausible PA reports (>10 VPA blocks or >18 MVPA blocks per day; n = 9). To evaluate the potential for selection bias, we compared our sample with excluded subjects with respect to sociodemographic, glycemic control, and anthropometric variables, as well as PA/ sedentary behavior assessed with questions based on the Youth Risk Behavior Surveillance System,26 which were available for all youths. Compared with nonrespondents (176 youths with type 1 DM, 26 youths with type 2 DM, and 39 control subjects), a larger proportion of 3DPAR respondents resided in South Carolina (P < .05), and they showed higher mean hemoglobin A1c levels (8.0 ± 2.3% vs 7.3 ± 2.1%; P = .002). This difference remained significant for youths with type 1 DM (8.7 ± 1.8% vs 8.2 ± 1.8%; P = .004) but not for youths with type 2 DM or control subjects.
Participant characteristics were described by using proportions or means and SDs. Differences between DM type/control groups were determined by using analyses of variance or χ2 tests. Gender- and case status (type 1 DM, type 2 DM, or control)–specific regression models were used to calculate compliance with recommendations and mean levels of PA and EM exposure, with adjustment for potential confounders (age, study site, and race/ethnicity). Analysis of variance was used to determine differences in PA and EM exposure variables between case status and gender groups. Mean VPA and EM exposure levels also were compared between normal-weight and obese youths among subjects with type 1 DM and control subjects. An additional model was used to calculate compliance with the MVPA standard for each case status group, with stratification according to race/ethnicity. All analyses were performed with SAS 9.0 (SAS Institute, Cary, NC). Statistical significance was determined at P < .05.
Overall, youths with type 2 DM were predominantly female, older, and more likely to be of black background, had shorter mean DM duration, lower mean glycohemoglobin levels, and greater mean waist circumference, and came from households with lower parental education and income, compared with youths with type 1 DM (Table 1). The proportion of obese youths was greater for youths with type 2 versus type 1 DM (85.1% vs 12.3%; P < .05) and was lower among youths with type 1 DM versus control subjects (12.3% vs 25.3%; P < .05). There were no striking differences in the prevalence of normal weight, overweight, and obesity according to DM type/control status when the data were stratified according to age groups.
Among male subjects, those with type 2 DM reported fewer blocks of VPA per day than did control subjects (1.1 ± 0.4 vs 2.3 ± 0.3 blocks per day; P < .05) (Table 2), with adjustment for age, site, and race/ethnicity. Rates of compliance with the VPA standard ranged from 32.9% among female subjects with type 2 DM to 61.4% among male control subjects, whereas rates for MVPA ranged from 64.6% among male subjects with type 2 DM to 82.3% among male subjects with type 1 DM (Table 2). Male subjects with type 2 DM showed lower rates of compliance with the VPA standard than control subjects (35.9% vs 61.4%; P < .05) and of compliance with the MVPA standard than male subjects with type 1 DM and control subjects (64.6% vs 82.3% and 81.3%, respectively; all P < .05). Rates of compliance with the EM use recommendations ranged from 29.5% to 49.1%, and rates of compliance with the television recommendations ranged from 46.7% to 58.4%.
To evaluate the role of weight status in this context, we limited the analysis to normal-weight versus obese subjects and excluded subjects with type 2 DM, because of the small number of normal-weight youths in this group (4 of 90 subjects). Among subjects with type 1 DM and subjects without DM, normal-weight youths tended to report higher levels of VPA and lower levels of exposure to EM than their obese counterparts; however these differences did not reach statistical significance (Fig 1).
For both genders combined, youths with type 2 DM were less likely to meet the MVPA standard, compared with youths with type 1 DM (odds ratio [OR]: 0.51 [95% confidence interval [CI]: 0.26–1.00]; P = .047) and control youths (OR: 0.48 [95% CI: 0.23–1.02]; P = .05), with adjustment for age, gender, race/ethnicity, study site, and BMI z score. Regardless of case status, female subjects were less likely than male subjects to meet the VPA standard (OR: 0.62 [95% CI: 0.45–0.86]; P = .004), but male subjects were less likely than female subjects to meet the EM (OR: 0.57 [95% CI: 0.41–0.78]; P < .001) and television (OR: 0.54 [95% CI: 0.37–0.79]; P < .001) use standards.
Of the 3 race/ethnicity groups, black youths showed the lowest prevalence of compliance with the MVPA standard (Fig 2). This difference reached statistical significance in comparison with Hispanic youths with type 1 DM (70% vs 92%; P < .05) and with non-Hispanic white youths with type 2 DM (59% vs 81%; P < .05). Within the black group, youths with type 2 DM showed a lower rate of compliance with the MVPA standard, compared with youths with type 1 DM and control youths (59% vs 70% and 79%, respectively; all P < .05). Regardless of case status, Hispanic and black youths were less likely than non-Hispanic white youths to meet the EM (OR: 0.52 [95% CI: 0.35–0.77]; P < .001) and television (OR: 0.65 [95% CI: 0.44–0.94]; P = .024) use standards.
To the best of our knowledge, this is the first study to describe the PA levels and EM use habits of a large sample of children and adolescents with type 1 and type 2 DM. A large proportion (17.7%–35.4%) of this population of youths with DM failed to meet current recommendations for MVPA, and the majority (50.9%–70.5%) did not meet the EM use recommendations. These findings are of particular concern because PA plays a pivotal role in the management of the disease, by improving acute and chronic glucose control and by reducing the risk of cardiometabolic complications.
We found that youths with type 2 DM engaged in significantly less VPA per day, compared with control subjects, a trend that also was evident (but not statistically significant) in comparison with youths with type 1 DM, at least for male subjects. Analyses assessing PA categorically showed that, with controlling for age, gender, race/ethnicity, study site, and BMI z score, youths with type 2 DM were ∼50% less likely to meet the MVPA standard, compared with youths with type 1 DM (P = .017) and control subjects (P = .05).
Because type 2 DM is still a rare disease in the overall population (0.01 cases per 1000 children 0–9 years of age and 0.42 cases per 1000 youths 10–19 years of age),1 there are few studies with which we can compare our results. Faulkner et al27 used 7-day activity recall among 27 adolescents with type 2 DM and 105 with type 1 DM. Those authors found that youths with type 2 DM reported engaging in less PA than their type 1 counterparts (34.7 vs 33.5 METs per day; P > .05). Rothman et al28 evaluated 103 adolescents with type 2 DM and found that 18% reported no PA, 48% reported exercising 1 to 3 times per week, and 34% reported exercising ≥4 times per week. On the basis of the available evidence, it seems that levels of participation in PA among contemporary youths with type 2 DM are rather low. The present study adds that rates of compliance with current public health recommendations for PA also are low, especially among male subjects with type 2 DM, compared with healthy control subjects or youths with type 1 DM. These results provide important baseline information regarding the PA levels of youths with type 2 DM, which should be contrasted with results from future studies, and indicate that promotion of PA is urgently needed in this group.
As a group, youths with type 1 DM reported higher PA levels than did youths with type 2 DM and very comparable or slightly lower levels than did control subjects in this study. Previous studies assessed PA among youths with type 1 DM and contrasted them with control populations, which yielded mixed results.29,–,31 Most studies reported lower PA estimates among youths with type 1 DM,32 although some reported no differences33,34 or slightly higher PA levels35,36 for subjects with type 1 DM, compared with control subjects. Differences in the methods used to assess PA make comparisons between studies difficult, although it seems that youths with type 1 DM tend to engage in PA at similar levels, compared with their healthy counterparts. However, there is still a need for improvement. In our study, ∼60% of youths with type 1 DM reported <1 block (30 minutes) of VPA per day and 20% reported <2 blocks of MVPA per day. The nature of type 1 DM, in which PA plays a key role in management of the disease, the increasing rates of overweight and obesity in this population (19.5% and 12.3%, respectively, in the present study), and the low rates of compliance with EM use recommendations (∼40%) are all indications that a physically active lifestyle should be promoted intensively in this group. In addition, previous research indicated that regular PA or individually tailored exercise training among youths with type 1 DM is safe, does not result in increased hypoglycemic episodes,12,15,29 and leads to reductions in lipoprotein levels,12,35,37 body fat, and blood pressure12,36,38 and improvements in cardiorespiratory fitness15,29,34,36 and chronic glucose control.38
Our study adds to current literature a characterization of television watching and overall EM use among youths with DM. In general, youths with type 2 DM reported slightly higher levels of television watching and exposure to other forms of EM, such as video games and Internet use, compared with both control subjects and youths with type 1 DM. However, the most striking finding was the very low prevalence of compliance with EM exposure recommendations exhibited by youths with and without DM. Surprisingly, we found very few previous studies assessing television watching or exposure to EM among youths with DM. A study of 538 youths with type 1 DM in Norway showed a strong negative correlation between hours of television watching and hemoglobin A1c levels, even with adjustment for age, BMI, and insulin dose (P < .001), with 46.6% of the sample reporting compliance with the television recommendation of <2 hours per day.17 Sedentary behavior, particularly time devoted to watching television, constitutes an important independent risk factor for childhood obesity39 and is associated with increased cardiometabolic risk16 and lower activity and fitness levels in pediatric populations.40 The fact that large proportions of children and youths with DM fail to meet both PA and EM use recommendations is troubling. Interventions aimed at improving metabolic control and prevention of CVD complications among youths with DM not only should provide opportunities for PA but also should include strategies to limit exposure to EM.
We found that black youths with DM showed lower prevalence of compliance with the MVPA recommendation, compared with both Hispanic and non-Hispanic white youths. In addition, regardless of case status, minority youths were nearly 50% less likely to meet EM exposure recommendations, compared with non-Hispanic white youths. This is consistent with general population data suggesting that minority youths engage in less PA and devote more time to sedentary behaviors than do non-Hispanic white youths.41
We also compared PA/EM use estimates across extremes of the BMI continuum (normal-weight versus obese youths). Compared with their normal-weight counterparts, obese youths with type 1 DM reported lower VPA (30%) and higher EM use (15%) levels. In addition, obese youths with type 1 DM showed lower VPA levels (40%) in comparison with obese control subjects. A previous cross-sectional study of 19143 pediatric patients with type 1 DM in Germany showed significantly higher BMI z scores among youths who reported no regular PA, compared with those who reported at least 1 or 2 sessions per week.36 Moreover, controlled PA interventions among youths with type 1 DM were shown to decrease the proportion of body fat by 6% to 8% and to increase lean body mass significantly.12,35,37 This study and previous investigations indicate that obese youths with type 1 DM seem to engage in less PA and more EM use than normal-weight youths with type 1 DM and should be considered a priority group for interventions.
The results of the present study must be interpreted in light of some limitations. First, our PA and EM use estimates were based on self-reports and were prone to information bias. However, the activity log we used is suitable for large-scale epidemiological studies, has proven reliability, and has been validated against criterion measures such as accelerometry among healthy adolescent girls,19 although not among youths with DM. Second, our study might have been underpowered to detect differences in activity/sedentary behaviors when multiple stratifications were made, especially among youths with type 2 DM. Third, we were unable to examine the potential effects of assessment days (weekdays versus weekends) on the PA and EM use estimates. In addition, although 3DPAR respondents did not differ in terms of sociodemographic, anthropometric, or PA/EM use characteristics, compared with nonrespondents, the fact that 3DPAR respondents had significantly lower hemoglobin A1c levels indicates that the possibility of selection bias, although low, cannot be excluded. Also, use of BMI as a marker of adiposity in pediatric clinical studies has its limitations. Finally, the cross-sectional nature of the data precludes any etiologic association between PA/sedentary behavior and DM or obesity risk.
This study provides comprehensive characterization of PA and EM use among a population-based sample of youths with DM. The results indicate that large proportions of youths with DM fail to meet current public health and clinical recommendations. In addition, youths with type 2 DM engage in less PA and devote more time to EM use than do control subjects and youths with type 1 DM. Also, male subjects, minority youths, and obese youths exhibit poorer activity and/ or sedentary behavior profiles and should be specifically targeted in future interventions. These findings are of concern because children and youths with DM are already at increased risk for the early development of cardiometabolic complications, and a large proportion of them might be missing the health benefits associated with an active lifestyle.
The SEARCH-CC study was funded by the National Institutes of Health and the National Institute of Diabetes and Digestive and Kidney Disease. This research was also supported by the Paffenbarger-Blair Fund for Epidemiologic Research on Physical Activity, from the American College of Sports Medicine Foundation. Dr Lobelo was funded by the American College of Sports Medicine (ACSM) Foundation.
We acknowledge Andrea M. Anderson and the SEARCH and SEARCH-CC study staff members, as well as the participating patients and their families, for collaboration in the study.
- Accepted January 28, 2010.
- Address correspondence to Angela D. Liese, PhD, MPH, University of South Carolina, Arnold School of Public Health, Department of Epidemiology and Biostatistics, 800 Sumter St, Columbia, SC 29208. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
Funded by the National Institutes of Health (NIH).
- PA =
- physical activity •
- EM =
- electronic media •
- DM =
- diabetes mellitus •
- VPA =
- vigorous physical activity •
- MVPA =
- moderate/vigorous physical activity •
- CVD =
- cardiovascular disease •
- OR =
- odds ratio •
- 3DPAR =
- 3-Day Physical Activity Recall •
- MET =
- metabolic equivalent •
- SEARCH =
- Search for Diabetes in Youth •
- SEARCH-CC =
- Search for Diabetes in YouthCase-Control •
- CI =
- confidence interval
- Liese AD,
- D'Agostino RB Jr.,
- Hamman RF,
- et al
- 2.↵American Diabetes Association. Economic costs of diabetes in the US in 2007. Diabetes Care. 2008;31(3):596–615
- 4.↵National Diabetes Data Group. Diabetes in America. Bethesda, MD: National Institute of Diabetes and Digestive and Kidney Diseases; 1995
- Rodriguez BL,
- Fujimoto WY,
- Mayer-Davis EJ,
- et al
- 8.↵American Diabetes Association. Standards of medical care in diabetes: 2008. Diabetes Care. 2008;31(suppl 1):S12–S54
- McGavock J,
- Sellers E,
- Dean H
- Kavey RE,
- Allada V,
- Daniels SR,
- et al
- Herbst A,
- Kordonouri O,
- Schwab KO,
- et al
- Margeirsdottir HD,
- Larsen JR,
- Brunborg C,
- et al
- 23.↵American Academy of Pediatrics, Committee on Public Education. Children, adolescents, and television. Pediatrics. 2001;107(2):423–426
- 24.↵US Department of Health and Human Services. Healthy People 2010. 2nd ed.Washington, DC: US Government Printing Office; 2000
- 25.↵National Center for Chronic Disease Prevention and Health Promotion. About BMI for children and teens. Available at: www.cdc.gov/healthyweight/assessing/bmi/childrens_bmi/about_childrens_bmi.html. Accessed February 27, 2009
- Faulkner MS,
- Quinn L,
- Rimmer JH,
- et al
- Rothman RL,
- Mulvaney S,
- Elasy TA,
- et al
- Raile K,
- Kapellen T,
- Schweiger A,
- et al
- Massin MM,
- Lebrethon MC,
- Rocour D,
- et al
- Lehmann R,
- Kaplan V,
- Bingisser R,
- et al
- Zeitler P,
- Epstein L,
- Grey M,
- et al
- Copyright © 2010 by the American Academy of Pediatrics