Abstract
Objectives
To characterize, over a two-year period, the proportion of youth with type 2 diabetes (T2D) enrolled in the Treatment Options for Type 2 Diabetes in Adolescents and Youth (TODAY) study that reported ever at least trying smoking cigarettes and/or drinking alcohol.
Study design
Longitudinal data were examined for participants with T2D ages 10 to 18 years old at baseline. Youth psychosocial, parent/family, environmental, and biological correlates of trying health risk behaviors were tested via cross-sectional multivariate models at each time point. Longitudinal models were explored for selected factors.
Results
Data were obtained from the TODAY study’s ethnically diverse participants at baseline (N=644), 6-month (N=616), and 24-month (N=543) assessments. Percent of youth ever trying only smoking remained stable at 4%, only drinking alcohol increased from 17% to 26%, and both smoking and drinking increased from 10% to 18% over the two-year period. Factors related to trying health risk behaviors were older age, male sex, non-Hispanic White race-ethnicity, lower grades, more depressive symptoms and stressful life events. Depressive symptoms, stressful life events, and BMI Z-score (the latter with smoking only) were related to engagement in health risk behaviors over time.
Conclusions
Youth with T2D who are already at risk for health complications and who reported engaging in activities that further increase the likelihood of life-threatening morbidities were characterized. Although most correlates of trying these risk behaviors are non-modifiable, intervention efforts may need to focus on potentially modifiable factors, such as depressive symptoms and lower grades.
Keywords: Type 2 Diabetes, Alcohol, Smoking, Health Risk, Youth
Youth with type 2 diabetes (T2D) may be at risk to develop comorbid complications, including hypertension, hyperlipidemia, and cancer, at an early age.1,2 Progression of complications could be exacerbated by duration of diabetes and the obesity that contributed to the initial diagnosis. Engaging in health risk behaviors such as smoking cigarettes and drinking alcohol may further jeopardize their health. Early initiation of smoking and drinking in youth, even in some cases if “just trying,” has been linked to later nicotine/alcohol use disorders and other health conditions.3–8 Published reports of smoking and alcohol use among youth with T2D are rare.9 Knowledge of the correlates of these health risk behaviors is limited to predominantly physically healthy youth and youth with type 1 diabetes (T1D). Demographic variables, e.g., age, sex, race/ethnicity, and socioeconomic status, have been related to youth health risk behaviors.9–12 Among psychosocial variables, lower grades, depressive symptomatology, stressful life events, and environmental factors are associated with youth substance use.10–13 Biological correlates of substance use risk behaviors are elevated body mass index (BMI) in youth with T1D14 and physically healthy youth15,16 and compromised metabolic control (HbA1c) that is associated with smoking14,17 and drinking18 among youth with T1D.
The primary aim of the present study was to characterize youth with T2D reporting ever at least trying smoking cigarettes and/or drinking alcohol over a two-year period. The youth were enrolled in Treatment Options for type 2 Diabetes in Adolescents and Youth (TODAY) Study Group, a multicenter, randomized clinical trial.2 The secondary aim was to test possible correlates of health risk behaviors. Hypotheses were that ever at least trying health risk behaviors would be related to older age, male sex, non-Hispanic White race/ethnicity, lower socioeconomic status, lower grades, more depressive symptoms and stressful life events, less perceived neighborhood safety, and higher BMI and HbA1c. Depressive symptoms and stressful life events were expected to predict both health risk behaviors across time, and higher BMI Z-scores were expected to be related to a greater likelihood of trying smoking.
Methods
A total of 644 of the 699 participants in the TODAY cohort provided data for this study; 52 were excluded because they belonged to “other” race-ethnicity. The multiple subcategories for this ethnic group were too heterogeneous to combine and the numbers for those belonging to any one specific racial/ethnic group within the ‘other’ category were too small to model. Three participants were excluded because of missing baseline outcome data.
The study design has been reported in detail, including containing a CONSORT flow chart.2,19 Briefly, the study group included 15 clinical centers, a data coordinating center and central laboratory. At baseline (July 2004 to February 2009), eligible individuals were 10–17 years of age with T2D diagnosed less than two years at time of randomization and with a BMI ≥85th percentile for age and sex. Participants were randomized to one of three treatment groups: (1) metformin alone, (2) metformin plus rosiglitazone, and (3) metformin plus an intensive lifestyle program. Outcome data were collected at baseline, 6 months, 12 months, and annually thereafter. The protocol was approved by an External Evaluation Committee convened by the NIDDK and the Institutional Review Boards for the Protection of Human Subjects of each institution. All participants provided informed consent; minor children confirmed assent. A Data and Safety Monitoring Board convened by NIDDK reviewed progress, safety, and interim outcome analyses.
For the present report, data were obtained at baseline, 6 and 24 months and included health risk behavior outcomes, baseline demographic information, youth psychosocial variables, a parent/family factor, an environmental factor, and biologic variables (BMI, HbA1c). Age- and sex-specific BMI Z-scores were calculated using CDC 2000 criteria.20 Except for baseline demographic data, all measures were collected at each assessment point.
Materials developed and used for the TODAY standard diabetes education program and the intensive lifestyle intervention program are available at https://today.bsc.gwu.edu/.
Outcome: Health risk behaviors
Youth were asked two questions: “Have you ever tried cigarette smoking, even one or two puffs?” (Yes/No), and “Have you ever had even one drink of alcohol?” (Yes/No). Once a participant responded positively to ever engaging in either risk behavior, all later time points reflected the affirmative response.
Predictors: Baseline demographics
Youth reported age (date of birth), sex, and race/ethnicity; a parent/guardian provided household annual income categorized as <$25,000, $25,000–$49,999, or ≥ $50,000.
Predictors: Youth psychosocial factors
Participants reported their academic performance (ie, grades earned during the last year [mostly As, Bs, Cs, or Ds and Fs]). Depressive symptoms were assessed with either the Children’s Depression Inventory (CDI)21 for participants ≥ 15 years or the Beck Depression Inventory II (BDI-II)22 for those ≥ 16 years. Scores were normalized to produce depressive symptomatology z-scores.
Predictors: Parent/family and environmental factors
Number of potentially stressful life events over the past year was determined by six items: change of household composition, serious illness or death of a close family member, moving, parent/guardian losing a job, parent/guardian getting a new job, and legal trouble or incarceration of a close family member. All positive responses were given a value of 1 and summed for a score of 0 to 6. As an environmental measure of perceived personal safety, participants reported the number of days in the last month they did not go to school because of feeling unsafe (0, 1, 2–3, 4–5, ≥ 6).
Predictors: Biologic factors
Height and weight were collected at all visits and used to calculate BMI. Height measurements were taken with a clinical stadiometer, and weight was measured in duplicate using a Seca scale (model 882, Seca USA, Hanover, MA) with a third measurement obtained if the first measurements differed >0.2 kg. BMI was calculated as weight (kg)/height2 (m2). Glycated hemoglobin (HbA1c) concentration was measured using a dedicated HPLC method (TOSOH Biosciences Inc., South San Francisco, CA) certified by the National Glycohemoglobin Standardization Program. At baseline all participants were on oral medications and not insulin. At later assessments, participants who had failed to maintain glycemic control had transitioned to insulin.
Statistical Analyses
Descriptive statistics, including means and standard deviations for continuous variables and frequency and percent for discrete variables, were used to describe the cohort at each assessment point. Only the three racial/ethnic categories with adequate sample sizes were evaluated (Hispanic, non-Hispanic Black, non-Hispanic White). A single health risk behavior outcome variable combined both smoking and drinking risk behaviors into four categories: (1) never smoked or drank; (2) only tried smoking; (3) only tried drinking; or (4) tried both smoking and drinking. The composite variable was used to examine initial associations between risk behaviors and variables of interest using a logistic model with cumulative logit. Treatment assignment was a priori included in a similar multivariate model in addition to variables significantly associated (p<0.05) with the risk behaviors for at least one assessment point in univariate analysis. The optimal multivariate logistic model was selected by evaluation of variable p-values and Akakie Information Criterion, excluding treatment. Longitudinal data were analyzed using a repeated measures logistic model with cumulative logit to account for the multiple observations per participant. The composite risk behavior (full behavior characterization) was modeled as a function of depressive symptom z-scores, stressful life events, and BMI Z-scores separately and jointly adjusting for age, sex, race/ethnicity, and treatment group. Similar analyses were done for outcome variables representing smoking only (Smoke, No Smoke) and drinking only (Drink, No Drink). The Statistical Analysis Software package (SAS, version 9.3, 2008, SAS Institute Inc., Cary, NC) was used for all analyses.
Results
Data were obtained from 644 participants of the TODAY study at baseline, 616 participants at the 6-month assessment, and 543 participants at the 24-month assessment (Table I).2 At baseline, participants were on average 14.2 years old (SD=2.1; range=10–18); 64.27% female; 22% Non-Hispanic White, 43% Hispanic, and 35% Non-Hispanic Black. Mean BMI-for-age percentile was 97.8 (SD=3.2; range=70.4–100.0) with the majority classified as obese (n=576; 89.4%) and a much smaller percentage as overweight (n=60; 9.3%) with only 1.3% (n=8) in the normal weight category. Mean HbA1c was 6.05% (SD=0.84; range=4.3–10.9%), and all participants were on oral medications and not insulin at baseline. No difference existed between participants included and not included in the analysis at 6 months. At 24 months, those included in the analysis were on average younger (13.8 ± 2.0 years old vs. 14.7 ± 2.0 years old; p-value<.0001) and engaged in less risky behavior at baseline (9.8% Smoke & Drink, 3.3% Smoke Only, 16.3% Drink Only vs. 13.6 Smoke & Drink, 6.8% Smoke Only, 21.4% Drink Only; p-value=0.0138) compared with those not in the analysis.
Table 1.
Anthropometric, Laboratory, and Survey Descriptive Statistics at Baseline, 6 Months and 24 Months
| Baseline (N=644) Mean ± SD or % |
6 Months (N=616) Mean ± SD or % |
24 Months (N=543) Mean ± SD or % |
|
|---|---|---|---|
| BMI Z-Score | 2.23 ± 0.46 | 2.20 ± 0.51 | 2.16 ± 0.55 |
| HbA1c* | 6.05 ± 0.77 | 6.42 ± 1.48 | 7.34 ± 2.39 |
| Depression Z-Score | −0.50 ± 0.84 | −0.66 ± 0.74 | −0.68 ± 0.76 |
| Grades in school in last year | |||
| Mostly As | 21.0% | 24.5% | 22.7% |
| Mostly Bs | 43.3% | 41.4% | 42.2% |
| Mostly Cs | 28.7% | 27.0% | 27.3% |
| Mostly Ds and Fs | 7.0% | 7.1% | 7.8% |
| Number of days did not go to school because felt unsafe | |||
| 0 days | 89.2% | 92.6% | 92.5% |
| 1 day | 4.4% | 2.5% | 1.7% |
| 2 or 3 days | 2.7% | 3.4% | 3.3% |
| 4 or 5 days | 2.2% | 0.7% | 1.4% |
| 6 or more days | 1.6% | 0.8% | 1.2% |
| Stressful Life Events | 1.46 ± 1.27 | 1.00 ± 1.01 | 0.96 ± 1.07 |
| Risk Behaviors | |||
| Smoke & Drink | 10.4% | 13.5% | 17.9% |
| Smoke Only | 3.9% | 4.1% | 3.9% |
| Drink Only | 17.1% | 20.8% | 26.2% |
| Neither Smoke nor Drink | 68.6% | 61.7% | 52.1% |
| Smoking | 14.3% | 17.5% | 21.7% |
| Drinking | 27.5% | 34.3% | 44.0% |
HbA1c at baseline was prior to any participants beginning insulin; 8.6% (N=53) were on insulin at 6 months and 30.2% (N=164) were on insulin at 24 months.
For establishing prevalence estimates in this cohort, at baseline 10.4% had ever tried both smoking cigarettes and drinking alcohol, which increased over time (13.5% at 6 months, 17.9% at 24 months). Percent trying smoking only was similar across time (ranging 3.9–4.1%), and percent trying only alcohol increased (17.1% at baseline, 20.8% at 6 months, 26.2% at 24 months). The majority reported not ever trying smoking cigarettes or drinking alcohol at baseline (68.6%), 6 months (61.7%), and 24 months (52.1%).
Analysis of covariates was undertaken examining smoking only, drinking only, and a composite variable of both smoking and drinking (Tables II–IV; Tables II and III available at www.jpeds.com). In general, the modeling results did not differ among the three methods for grouping outcomes, but exceptions are noted in Results and Discussion sections. Cross-sectional analyses of associations for health risk behaviors using univariate (ie, unadjusted) models were conducted for youth psychosocial, parent/family, environmental, and biological factors. P-values for the association of independent factors separately with health risk behaviors were examined at each time point. Of all of the variables, only three (household income, perceived personal safety, and HbA1c) were not statistically significant at any time point (p<0.05) and were dropped from further analyses.
Table 2.
Overall Multivariate Model P-values and Means ± SD or % for Smoking at Each Time Point
| Baseline | 6 Months | 24 Months | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Smoke | No Smoke | Univariate Model p-value |
Multivariate Model p-value** |
Smoke | No Smoke | Univariate Model p-value |
Multivariate Model p-value** |
Smoke | No Smoke | Univariate Model p-value |
Multivariate Model p-value** |
|
|
|
|
|
||||||||||
| Age | 15.3 ± 1.6 | 13.8 ± 2.0 | <.0001 | <.0001 | 15.6 ± 1.7 | 14.2 ± 2.0 | <.0001 | <.0001 | 16.9 ± 1.6 | 15.5 ± 2.0 | <.0001 | <.0001 |
| Sex | 0.4581 | * | 0.1292 | * | 0.0815 | * | ||||||
| Female | 66.7% | 64.3% | 57.6% | 64.3% | 55.6% | 65.7% | ||||||
| Male | 33.3% | 35.7% | 42.4% | 35.7% | 44.4% | 34.3% | ||||||
| Race/Ethnicity | 0.0318 | 0.0480 | 0.3301 | 0.2830 | 0.1406 | 0.0531 | ||||||
| Black | 39.3% | 34.8% | 36.4% | 33.2% | 36.1% | 34.1% | ||||||
| Hispanic | 28.6% | 44.1% | 39.4% | 45.3% | 37.0% | 45.1% | ||||||
| White | 32.1% | 21.2% | 24.2% | 21.5% | 26.9% | 20.8% | ||||||
| Household Income | 0.6036 | * | 0.6107 | * | 0.3825 | * | ||||||
| <$25,000 | 34.2% | 40.0% | 39.8% | 40.7% | 39.4% | 40.9% | ||||||
| $25,000 – $49,999 | 41.1% | 33.3% | 36.4% | 33.1% | 38.3% | 32.5% | ||||||
| ≥$50,000 | 24.7% | 26.8% | 23.9% | 26.2% | 22.3% | 26.6% | ||||||
| BMI Z-score | 2.3 ± 0.5 | 2.2 ± 0.5 | 0.0106 | 0.1211 | 2.3 ± 0.5 | 2.2 ± 0.5 | 0.0010 | 0.0404 | 2.2 ± 0.6 | 2.1 ± 0.5 | 0.1079 | 0.5439 |
| HbA1c# | 6.0 ± 0.7 | 6.0 ± 0.8 | 0.8370 | * | 6.4 ± 1.4 | 6.4 ± 1.5 | 0.9115 | * | 7.5 ± 2.5 | 7.2 ± 2.3 | 0.4038 | * |
| Depression Z-Score | −0.3 ± 1.0 | −0.6 ± 0.8 | 0.0201 | 0.0169 | −0.6 ± 0.8 | −0.7 ± 0.7 | 0.0059 | 0.0561 | −0.6 ± 0.8 | −0.7 ± 0.7 | 0.1265 | 0.1866 |
| Grades | 0.0014 | 0.4404 | 0.0004 | 0.0274 | 0.0328 | 0.3578 | ||||||
| Mostly As | 11.9% | 22.5% | 9.2% | 27.7% | 13.5% | 25.0% | ||||||
| Mostly Bs | 40.5% | 43.8% | 44.9% | 40.6% | 43.8% | 41.8% | ||||||
| Mostly Cs | 34.5% | 27.7% | 33.7% | 25.6% | 34.4% | 25.5% | ||||||
| Mostly Ds & Fs | 13.1% | 6.1% | 12.2% | 6.0% | 8.3% | 7.7% | ||||||
| # of Days Missed School for Safety | 0.8653 | * | 0.7581 | * | 0.6285 | * | ||||||
| 0 Days | 88.1% | 90.0% | 92.7% | 92.9% | 91.7% | 92.7% | ||||||
| 1 Day | 7.1% | 3.7% | 2.1% | 2.6% | 0% | 2.3% | ||||||
| 2–3 Days | 2.4% | 2.7% | 3.1% | 3.4% | 4.2% | 3.1% | ||||||
| 4–5 Days | 2.4% | 2.0% | 0% | 0.9% | 1.0% | 1.3% | ||||||
| 6+ Days | 0% | 1.6% | 2.1% | 0.2% | 3.1% | 0.5% | ||||||
| Stress Score | 2.0 ± 1.4 | 1.4 ± 1.2 | <.0001 | 0.0023 | 1.1 ± 1.0 | 0.9 ± 1.0 | 0.0462 | 0.7092 | 1.1 ± 1.1 | 0.9 ± 1.0 | 0.0066 | 0.1151 |
HbA1c at baseline was prior to any participants beginning insulin; 8.6% (N=53) were on insulin at 6 months and 30.2% (N=164) were on insulin at 24 months.
Variable not included in the multivariate model.
Multivariate model adjusted for treatment assignment.
Table 4.
Overall Univariate and Multivariate Model P-values and Means ± SD or % for Full Characterization of Health Risk Behaviors at Each Time Point
| Baseline | 6 Months | 24 Months | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Smoke & Drink |
Smoke Only |
Drink Only | Neither | Univariate Model p-value |
Multivariate Model** p-value |
Smoke & Drink |
Smoke Only |
Drink Only | Neither | Univariate Model p-value |
Multivariate Model** p-value |
Smoke & Drink |
Smoke Only |
Drink Only | Neither | Univariate Model p-value |
Multivariate Model** p-value |
|
|
|
|
|
||||||||||||||||
| Age | 15.5 ± 1.4 | 14.7 ± 1.9 | 14.7 ± 1.8 | 13.5 ± 2.0 | <.0001 | <.0001 | 15.9 ± 1.5 | 14.7 ± 1.8 | 15.1 ± 1.9 | 13.9 ± 2.0 | <.0001 | <.0001 | 17.1 ± 1.5 | 16.1 ± 1.7 | 16.2 ± 1.8 | 15.1 ± 1.9 | <.0001 | <.0001 |
| Sex | 0.2352 | 0.1467 | 0.0202 | 0.7920 | 0.0025 | 0.0256 | ||||||||||||
| Female | 66.7% | 66.7% | 61.1% | 65.1% | 56.6% | 60.9% | 57.4% | 66.7% | 54.4% | 61.1% | 56.7% | 70.2% | ||||||
| Male | 33.3% | 33.3% | 38.9% | 34.9% | 43.4% | 39.1% | 42.6% | 33.3% | 45.6% | 38.9% | 43.3% | 29.8% | ||||||
| Race/Ethnicity | 0.0443 | 0.0328 | 0.3363 | 0.3291 | 0.0633 | 0.0221 | ||||||||||||
| Black | 38.3% | 41.7% | 33.3% | 35.1% | 34.2% | 43.5% | 31.1% | 33.9% | 33.3% | 50.0% | 32.1% | 35.1% | ||||||
| Hispanic | 28.3% | 29.2% | 42.6% | 44.5% | 40.8% | 34.8% | 45.9% | 45.1% | 36.7% | 38.9% | 42.5% | 46.4% | ||||||
| White | 33.3% | 29.2% | 24.1% | 20.4% | 25.0% | 21.7% | 23.0% | 21.0% | 30.0% | 11.1% | 25.4% | 18.5% | ||||||
| Household Income | 0.7303 | * | 0.3709 | * | 0.3000 | * | ||||||||||||
| <$25,000 | 33.3% | 36.8% | 38.1% | 40.4% | 39.7% | 40.0% | 35.6% | 42.3% | 38.5% | 43.8% | 34.2% | 44.2% | ||||||
| $25,000 – $49,999 | 38.9% | 47.4% | 29.9% | 34.2% | 35.3% | 40.0% | 31.7% | 33.5% | 38.5% | 37.5% | 31.6% | 32.9% | ||||||
| ≥$50,000 | 27.8% | 15.8% | 32.0% | 25.4% | 25.0% | 20.0% | 32.7% | 24.1% | 23.1% | 18.8% | 34.2% | 22.9% | ||||||
| BMI Z-score | 2.3 ± 0.5 | 2.4 ± 0.5 | 2.2 ± 0.5 | 2.2 ± 0.5 | 0.0623 | 0.2588 | 2.3 ± 0.5 | 2.4 ± 0.5 | 2.2 ± 0.6 | 2.2 ± 0.5 | 0.0205 | 0.1012 | 2.2 ± 0.6 | 2.2 ± 0.5 | 2.1 ± 0.6 | 2.2 ± 0.5 | 0.3451 | 0.7024 |
| HbA1c# | 5.9 ± 0.8 | 6.2 ± 0.7 | 6.1 ± 0.7 | 6.0 ± 0.8 | 0.8085 | * | 6.3 ± 1.2 | 6.7 ± 1.9 | 6.5 ± 1.4 | 6.3 ± 1.5 | 0.4670 | * | 7.4 ± 2.3 | 8.2 ± 3.1 | 7.2 ± 2.2 | 7.2 ± 2.4 | 0.5994 | * |
| Depression Z-Score | −0.4 ± 0.9 | −0.2 ± 1.1 | −0.5 ± 0.8 | −0.6 ± 0.8 | 0.0220 | 0.0546 | −0.6 ± 0.8 | −0.4 ± 0.8 | −0.7 ± 0.7 | −0.7 ± 0.7 | 0.0209 | 0.0443 | −0.6 ± 0.7 | −0.6 ± 1.0 | −0.7 ± 0.7 | −0.8 ± 0.7 | 0.0493 | 0.0329 |
| Grades | <.0001 | 0.0492 | 0.0017 | 0.1172 | 0.0144 | 0.1944 | ||||||||||||
| Mostly As | 6.7% | 25.0% | 12.1% | 25.2% | 10.7% | 4.3% | 24.2% | 29.0% | 12.5% | 18.8% | 20.5% | 27.3% | ||||||
| Mostly Bs | 48.3% | 20.8% | 43.0% | 44.0% | 41.3% | 56.5% | 42.5% | 40.0% | 45.0% | 37.5% | 44.7% | 40.4% | ||||||
| Mostly Cs | 35.0% | 33.3% | 37.4% | 25.2% | 37.3% | 21.7% | 30.8% | 23.8% | 35.0% | 31.3% | 27.3% | 24.6% | ||||||
| Mostly Ds & Fs | 10.0% | 20.8% | 7.5% | 5.7% | 10.7% | 17.4% | 2.5% | 7.2% | 7.5% | 12.5% | 7.6% | 7.7% | ||||||
| # of Days Missed School for Safety | 0.6666 | * | 0.6363 | * | 0.2979 | * | ||||||||||||
| 0 Days | 90.0% | 83.3% | 94.4% | 88.9% | 94.5% | 87.0% | 91.7% | 93.3% | 91.3% | 93.8% | 93.1% | 92.6% | ||||||
| 1 Day | 5.0% | 12.5% | 1.9% | 4.2% | 2.7% | 0% | 1.7% | 2.9% | 0% | 0% | 2.3% | 2.3% | ||||||
| 2–3 Days | 3.3% | 0% | 0.9% | 3.2% | 2.7% | 4.3% | 5.0% | 2.9% | 5.0% | 0% | 3.8% | 2.7% | ||||||
| 4–5 Days | 1.7% | 4.2% | 2.8% | 1.7% | 0% | 0% | 1.7% | 0.6% | 1.3% | 0% | 0% | 2.0% | ||||||
| 6+ Days | 0% | 0% | 0% | 2.0% | 0% | 8.7% | 0% | 0.3% | 2.5% | 6.3% | 0.8% | 0.4% | ||||||
| Stress Score | 2.0 ± 1.5 | 1.9 ± 1.2 | 1.4 ± 1.2 | 1.4 ± 1.3 | 0.0012 | 0.0500 | 1.1 ± 1.0 | 1.1 ± 1.0 | 1.1 ± 1.0 | 0.9 ± 1.0 | 0.0089 | 0.1712 | 1.1 ± 1.1 | 1.3 ± 1.4 | 1.0 ± 1.0 | 0.8 ± 1.0 | 0.0013 | 0.0090 |
HbA1c at baseline was prior to any participants beginning insulin; 8.6% (N=53) were on insulin at 6 months and 30.2% (N=164) were on insulin at 24 months.
Variable not included in the multivariate model.
Multivariate model adjusted for treatment assignment.
Table 3.
Overall Multivariate Model P-values and Means ± SD or % for Drinking at Each Time Point
| Baseline | 6 Months | 24 Months | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Drink | No Drink | Univariate Model p-value |
Multivariate Model p-value** |
Drink | No Drink | Univariate Model p-value |
Multivariate Model p-value** |
Drink | No Drink | Univariate Model p-value |
Multivariate Model p-value** |
|
|
|
|
|
||||||||||
| Age | 15.0 ± 1.7 | 13.6 ± 2.0 | <.0001 | <.0001 | 15.4 ± 1.8 | 14.0 ± 2.0 | <.0001 | <.0001 | 16.6 ± 1.7 | 15.2 ± 1.9 | <.0001 | <.0001 |
| Sex | 0.2628 | 0.3422 | 0.0440 | 0.3504 | 0.0024 | 0.0558 | ||||||
| Female | 63.1% | 65.2% | 57.1% | 66.3% | 55.8% | 69.6% | ||||||
| Male | 36.9% | 34.8% | 42.9% | 33.7% | 44.2% | 30.4% | ||||||
| Race/Ethnicity | 0.1597 | * | 0.3776 | * | 0.0529 | * | ||||||
| Black | 35.1% | 35.5% | 32.3% | 34.5% | 32.6% | 36.0% | ||||||
| Hispanic | 37.5% | 43.6% | 43.9% | 44.5% | 40.2% | 45.9% | ||||||
| White | 27.4% | 20.9% | 23.7% | 21.0% | 27.2% | 18.0% | ||||||
| Household Income | 0.5235 | * | 0.2017 | * | 0.1267 | * | ||||||
| <$25,000 | 36.4% | 40.3% | 37.2% | 42.2% | 35.9% | 44.1% | ||||||
| $25,000 – $49,999 | 33.1% | 34.8% | 33.1% | 33.9% | 34.4% | 33.2% | ||||||
| ≥$50,000 | 30.5% | 24.9% | 29.7% | 23.9% | 29.7% | 22.7% | ||||||
| BMI Z-score | 2.2 ± 0.5 | 2.2 ± 0.5 | 0.5002 | * | 2.2 ± 0.5 | 2.2 ± 0.5 | 0.5032 | * | 2.2 ± 0.6 | 2.2 ± 0.5 | 0.9173 | * |
| HbA1c# | 6.0 ± 0.7 | 6.0 ± 0.8 | 0.9458 | * | 6.4 ± 1.3 | 6.4 ± 1.6 | 0.6041 | * | 7.3 ± 2.3 | 7.3 ± 2.4 | 0.9894 | * |
| Depression Z-Score | −0.5 ± 0.9 | −0.5 ± 0.8 | 0.1224 | * | −0.7 ± 0.7 | −0.7 ± 0.7 | 0.1995 | * | −0.7 ± 0.7 | −0.8 ± 0.7 | 0.0675 | * |
| Grades | <.0001 | 0.0351 | 0.0127 | 0.6056 | 0.0330 | 0.3352 | ||||||
| Mostly As | 10.2% | 25.2% | 19.0% | 27.4% | 17.5% | 26.8% | ||||||
| Mostly Bs | 44.9% | 42.7% | 42.1% | 41.0% | 44.8% | 40.2% | ||||||
| Mostly Cs | 36.5% | 25.6% | 33.3% | 23.6% | 30.2% | 25.0% | ||||||
| Mostly Ds & Fs | 8.4% | 6.5% | 5.6% | 7.9% | 7.5% | 8.0% | ||||||
| # of Days Missed School for Safety | 0.5693 | * | 0.5259 | * | 0.4809 | * | ||||||
| 0 Days | 92.8% | 88.6% | 92.8% | 92.9% | 92.4% | 92.6% | ||||||
| 1 Day | 3.0% | 4.7% | 2.1% | 2.7% | 1.4% | 2.2% | ||||||
| 2–3 Days | 1.8% | 3.0% | 4.1% | 3.0% | 4.3% | 2.6% | ||||||
| 4–5 Days | 2.4% | 1.9% | 1.0% | 0.5% | 0.5% | 1.8% | ||||||
| 6+ Days | 0% | 1.9% | 0% | 0.8% | 1.4% | 0.7% | ||||||
| Stress Score | 1.6 ± 1.3 | 1.4 ± 1.3 | 0.0598 | 0.1193 | 1.1 ± 1.0 | 0.9 ± 1.0 | 0.0218 | 0.0406 | 1.0 ± 1.0 | 0.9 ± 1.0 | 0.0112 | 0.0811 |
HbA1c at baseline was prior to any participants beginning insulin; 8.6% (N=53) were on insulin at 6 months and 30.2% (N=164) were on insulin at 24 months.
Variable not included in the multivariate model.
Multivariate model adjusted for treatment assignment.
Multivariate modeling indicated that age was significantly related to risk behaviors at every time period in the positive direction in all of the models (composite, smoking, and drinking) (p<0.001); i.e., older participants were more likely to report trying risk behaviors. Males were significantly different from females by the 24-month assessment in the composite outcomes model (p=0.0256); 70.2% of females and 29.8% of males had never tried either behavior. At baseline, almost twice as many non-Hispanic Whites had tried both smoking and drinking (15.5%) compared with Hispanics (7.9%), with non-Hispanic Blacks in between the other groups (10.2%) (p = 0.0328). A different pattern occurred at baseline in the smoking model with non-Hispanic blacks reporting having tried smoking at the highest percentage (non-Hispanic Whites 32.1%, Hispanics 28.6%, non-Hispanic Blacks 39.3%; p=0.021). Depressive symptomatology z-scores were associated with health risk behavior reporting in the composite model at the last two assessment points (p=0.0443 at 6 months, p=0.0329 at 24 months) and in the smoking model at baseline (p=0.017) only. Having worse grades was related to health risk behaviors in the composite model at baseline (p = 0.0492), drinking at baseline (p=0.035), and smoking at 6 months (p=0.027). Experiencing more stressful life events was associated with having tried cigarettes/alcohol at 24 months only (p=.0090), smoking at baseline (p=0.002), and drinking at 6 months (p=0.041). Finally, BMI z-scores were associated with having tried smoking (p=0.04) at the 6-month assessment only.
Table V shows the output from the longitudinal analyses of health risk behaviors conducted separately and then combined for depression z-scores, stressful life events, and BMI Z-scores, adjusting for age, sex, race/ethnicity, and treatment group. Higher depressive symptomatology z-scores were significantly related to greater health risk behavior engagement over time for the composite outcome and for drinking. More stressful life events were significantly related to engagement in health risk behaviors over time in all of the models. Higher BMI Z-scores were related to greater reporting of having tried smoking cigarettes over time but were not related to health risk behaviors in the composite health risk behavior or drinking-only models. As hypothesized, a youth psychosocial variable (depressive symptomology), parent/family variable (stress), and biologic variable (BMI Z-score) (the latter with smoking only) were significantly related to trying health risk behaviors over time.
Table 5.
Longitudinal Multivariate Modeling of Health Risk Behaviors
| Full Characterization Model* (Smoke & Drink, Smoke, Drink, Neither) | Smoking Model** (Smoke, No Smoke) | Drinking Model*** (Drink, No Drink) | |||||||
|---|---|---|---|---|---|---|---|---|---|
| β | Standard Error | Overall p-value | β | Standard Error | Overall p-value | β | Standard Error | Overall p-value | |
|
|
|
|
|||||||
| Depression Z-Score Only Model | |||||||||
| Depression Z-Score | 0.2895 | 0.0770 | 0.0005 | 0.3605 | 0.0918 | 0.0009 | 0.1977 | 0.0845 | 0.0228 |
| Stress Score Only Model | |||||||||
| Stress Score | 0.2265 | 0.0495 | <.0001 | 0.2560 | 0.0628 | 0.0002 | 0.1745 | 0.0537 | 0.0013 |
| BMI Z-Score Only Model | |||||||||
| BMI Z-Score | 0.2467 | 0.1525 | 0.1025 | 0.4997 | 0.2011 | 0.0101 | 0.0933 | 0.1668 | 0.5755 |
| Joint Model | |||||||||
| Depression Z-Score | 0.2485 | 0.0780 | 0.0030 | 0.3026 | 0.0937 | 0.0047 | 0.1659 | 0.0862 | 0.0604 |
| Stress Score | 0.1928 | 0.0505 | 0.0003 | 0.2119 | 0.0648 | 0.0022 | 0.1541 | 0.0547 | 0.0053 |
| BMI Z-Score | 0.2221 | 0.1559 | 0.1498 | 0.4845 | 0.2083 | 0.0165 | 0.0779 | 0.1693 | 0.6450 |
Model evaluating relationship with outcome of the full 4 health risk behavior categories (Smoke & Drink, Smoke Only, Drink Only, Neither)
Model evaluating relationship with outcome of smoking only (Smoke, No Smoke)
Model evaluating relationship with outcome of drinking only (Drink, No Drink)
All models adjusted for age, sex, race/ethnicity and treatment assignment.
Discussion
Despite already elevated health risks, a subset of youth with T2D in our sample reported engaging in activities that further increase the likelihood of life-threatening morbidities. Prevalence of trying smoking is lower than in a sample of youth with T2D participating in the SEARCH for Diabetes in Youth study.9 Reynolds et al found that 35.9% (N=579) of 10 to 22 year olds with T2D reported ever trying smoking compared with approximately 14–22% in our slightly younger sample.9 Our prevalence estimates are generally in line with the Centers for Disease Control and Prevention’s (CDC) 2009 Youth Risk Behavior Survey (YRBS) of high school students for which 19.5% had smoked cigarettes in the past 30 days, even though the CDC survey was conducted with an older sample and characterized current use rather than ever just trying smoking.23 For ever trying alcohol, our prevalence estimates (28–44%) are close to the CDC current use data with 41.8% reporting that they had drank alcohol in the past 30 days.23 In general, prevalence for at least trying smoking or drinking in this sample of youth with T2D is at the lower end of estimates that vary from 19.5–90% depending upon age, race, and operalization of health behavior.9,23,24 To our knowledge, no prevalence data exist for alcohol consumption in youth with T2D. Our study is unique in characterizing health risk behaviors for youth with T2D enrolled in a longitudinal research trial.
Many predictions based on findings from the prior literature were confirmed, including from school- and community-based studies as well as studies of youth with T1D.9–14 In multivariate modeling at each assessment point (ie, cross-sectional adjusted analyses), demographic factors related to increased risk for at least trying health risk behaviors are older age, male sex, and non-Hispanic White race-ethnicity for the composite health risk behaviors and non-Hispanic Black race-ethnicity for smoking. Child/family factors include lower grades, more depressive symptoms, and stressful life events. Of those, the most potentially modifiable factors warranting early identification and treatment are lower academic achievement and depressive symptoms. BMI Z-scores related to smoking at one time period. Previous research on the relationship between BMI z-scores and smoking among youth with T2D are difficult to interpret. BMI Z-scores were not significantly related to cigarette smoking status in the Reynolds study with youth who have T2D; 9 however, Reynolds et al found a significant difference between cigarette smoking variable levels for waist circumference (p=0.0198), with current smokers having the largest circumference.9 Finally, as expected, treatment group assignment was not related to engagement in health risk behaviors. The intensive lifestyle program arm focused on weight control and not on health risk behaviors and all three treatment groups received the same health risk prevention education.
Contrary to our a priori hypotheses, household income, perceived neighborhood safety, and HbA1c were not significantly related to ever having tried cigarettes and alcohol in cross-sectional multivariate analyses. The absence of a relationship between SES and health risk behaviors in our study is not unique. The national Monitoring the Future Study reported weak or negligible relationships between SES and alcohol/cigarette use.25 Reynolds et al found SES to be predictive of smoking among youth with T1D but not among those with T2D. 9 The fact that our youth were generally from families of low education level and low annual household income and were therefore rather homogeneous with regard to SES may also have influenced our findings.26 The lack of a relationship of perceived personal safety with health risk behaviors may be partially related to the low variability of this measure. Approximately 7–11% of the sample reported across the three assessment periods that there were one or more days that they had not gone to school because of feeling unsafe; these data are in line with national averages. In the 2009 CDC survey, only 5% of high school-aged children had not gone to school on at least one day due to safety concerns, and this percentage was slightly greater for younger children.23 Finally, we found that HbA1c was not associated with health risk behaviors in the cross-sectional models. In the Reynolds study, current smokers were more likely to have poor glycemic control than were non-smokers. 9 Our sample was distinctive in that participants had been recruited for a treatment trial, and achieving glycemic control on oral medications via a rigorous run-in period was a requirement for enrollment.
The TODAY study offered the unique opportunity to explore longitudinal associations over a two-year period between selected predictors and our outcome of ever having tried smoking/drinking or either one in separate models. Our findings fit with prior research with physically healthy youth in that depressive symptomatology and life stressors significantly predicted trying health risk behaviors.12 Finally, the finding that BMI Z-scores significantly predicted smoking over time is novel. As smoking has been reported to be a weight control technique in samples of physically healthy youth,15 it is possible that youth with T2D might be more likely to try this strategy as their BMI Z-scores increase. We would expect that weight management would be only one of a myriad number of potential reasons for these youth to try smoking cigarettes.
Our findings must be interpreted with a number of limitations in mind. The primary data analytic strategy adopted was to put together smoking/drinking into one composite, categorical variable. However, our hypotheses were also tested in separate models for smoking only and drinking only with relevant findings highlighted in the text and all of these analyses presented in Tables II and III. Another limitation is that this is an exploration of reports of trying health risk behaviors rather than the frequency and quantity of use. This decision was made due to the wording of the questions: “Have you ever tried cigarette smoking, even one or two puffs?” and “Have you ever had even one drink of alcohol?” Based on prior research, early initiation of smoking and drinking in youth, even in some cases of “just trying,” is an important metric to examine due to its relationship with a number of later adverse outcomes including nicotine/alcohol use disorders and other health conditions.3–8, 24 This is particularly concerning for our vulnerable sample of youth with T2D. Another study limitation relating to the outcome variable is that the exact age at initiation is not known due to the wording of the questions and the timing of the assessments. In addition, we do not have data on other health risk behaviors, including the use of marijuana or other drugs. For the longitudinal analyses, causation cannot be assumed between the predictors (depressive symptomatology, life stressors, and BMI Z-scores) and outcome (health risk behaviors) even though these variables were significantly related over time. Finally, when interpreting findings in cross-sectional and longitudinal models, it must be noted that at baseline all participants were on oral medications and not insulin. At the subsequent assessments, those who had failed to maintain glycemic control had transitioned to insulin. This means that the baseline study group was the most homogenous in terms of the treatment regimen.
Future research in samples of youth with T2D should include expanded measurement of these health risk behaviors to increase knowledge of the age of first trying, prevalence, frequency/severity, and incidents of binge drinking in addition to the use of other forms of tobacco, marijuana and other drugs. Our finding that BMI Z-scores significantly predict smoking over time merits further study to explore the interplay between T2D, obesity and engagement in specific types of health risk behaviors. Similarly, additional work is needed to better understand how engagement in health risk behaviors may have a greater impact on this group of youth and young adults compared with their physically healthy peers and also how these behaviors map onto other health behaviors (e.g., dietary intake, physical activity) that are part of the treatment regimen for T2D.
The likelihood is high that smoking cigarettes and/or drinking alcohol may have synergistic effects coupled with the already serious health consequences of T2D and overweight/obesity to hasten the further physical deterioration of this cohort.2 Therefore, the baseline and follow-up data available from the TODAY study provide an opportunity to explore the prevalence of these health risk behaviors in youth with T2D and the relationships between demographic, youth psychosocial, parent/family, environmental, and biologic predictors and these health risk behaviors. These models identify risk/protective factors for substance use among adolescents and youth with T2DM that are potentially modifiable, depressive symptomatology and school performance, and these may need to become targets for treatment interventions. Monitoring depressive symptomatology and life stressors in these youth takes on greater importance when considering of their longitudinal associations with health risk behaviors. Finally, increasing weight status (ie, BMI Z-scores) may be related to greater risk for trying smoking in this already vulnerable population.
Acknowledgments
We gratefully acknowledge the participation and guidance of the American Indian partners associated with the clinical center located at the University of Oklahoma Health Sciences Center, including members of the Absentee Shawnee Tribe, Cherokee Nation, Chickasaw Nation, Choctaw Nation of Oklahoma, and Oklahoma City Area Indian Health Service;
Funded by the National Institute of Diabetes and Digestive and Kidney Diseases and the National Institutes of Health Office of the Director (U01-DK61212, U01-DK61230, U01-DK61239, U01-DK61242, and U01-DK61254), the National Center for Research Resources General Clinical Research Centers Program (M01-RR00036 [Washington University School of Medicine], M01-RR00043-45 [Children’s Hospital Los Angeles], M01-RR00069 [University of Colorado Denver], M01-RR00084 [Children’s Hospital of Pittsburgh], M01-RR01066 [Massachusetts General Hospital], M01-RR00125 [Yale University], M01-RR14467 [University of Oklahoma Health Sciences Center]), and the NCRR Clinical and Translational Science Awards (UL1-RR024134 [Children’s Hospital of Philadelphia], UL1-RR024139 [Yale University], UL1-RR024153 [Children’s Hospital of Pittsburgh], UL1-RR024989 [Case Western Reserve University], UL1-RR024992 [Washington University, St Louis], UL1-RR025758 [Massachusetts General Hospital], UL1-RR025780 [University of Colorado Denver]).
Abbreviations
- BDI-II
Beck Depression Inventory II
- BMI
Body Mass Index
- CDC
Centers for Disease Control and Prevention
- CDI
Children’s Depression Inventory
- HBA1c
Glycated Hemoglobin
- NIDDK
National Institute of Diabetes and Digestive Kidney Diseases
- T1D
Type 1 diabetes
- T2D
Type 2 Diabetes
- TODAY
Treatment Options for Type 2 Diabetes in Adolescents and Youth
- YRBS
Youth Risk Behavior Survey
Appendix 1
Individuals and institutions associated with the TODAY Study Group include (* indicates principal investigator or director):
Clinical Centers--Baylor College of Medicine: S. McKay*, M. Haymond*, B. Anderson, C. Bush, S. Gunn, H. Holden, S.M. Jones, G. Jeha, S. McGirk, S. Thamotharan; Case Western Reserve University: L. Cuttler*, E. Abrams, T. Casey, W. Dahms (deceased), C. Ievers-Landis, B. Kaminski, M. Koontz, S. MacLeish, P. McGuigan, S. Narasimhan; Children’s Hospital Los Angeles: M. Geffner*, V. Barraza, N. Chang, B. Conrad, D. Dreimane, S. Estrada, L. Fisher, E. Fleury-Milfort, S. Hernandez, B. Hollen, F. Kaufman, E. Law, V. Mansilla, D. Miller, C. Muñoz, R. Ortiz, A. Ward, K. Wexler, Y.K. Xu, P. Yasuda; Children’s Hospital of Philadelphia: L. Levitt Katz*, R. Berkowitz, S. Boyd, B. Johnson, J. Kaplan, C. Keating, C. Lassiter, T. Lipman, G. McGinley, H. McKnight, B. Schwartzman, S. Willi; Children’s Hospital of Pittsburgh: S. Arslanian*, F. Bacha, S. Foster, B. Galvin, T. Hannon, A. Kriska, I. Libman, M. Marcus, K. Porter, T. Songer, E. Venditti; Columbia University Medical Center: R. Goland*, D. Gallagher, P. Kringas, N. Leibel, D. Ng, M. Ovalles, D. Seidman Joslin Diabetes Center: L. Laffel*, A. Goebel-Fabbri, M. Hall, L. Higgins, J. Keady, M. Malloy, K. Milaszewski, L. Rasbach; Massachusetts General Hospital: D.M. Nathan*, A. Angelescu, L. Bissett, C. Ciccarelli, L. Delahanty, V. Goldman, O. Hardy, M. Larkin, L. Levitsky, R. McEachern, D. Norman, D. Nwosu, S. Park-Bennett, D. Richards, N. Sherry, B. Steiner; Saint Louis University: S. Tollefsen*, S. Carnes, D. Dempsher, D. Flomo, T. Whelan, B. Wolff; State University of New York Upstate Medical University: R. Weinstock*, D. Bowerman, S. Bristol, J. Bulger, J. Hartsig, R. Izquierdo, J. Kearns, R. Saletsky, P. Trief; University of Colorado Denver: P. Zeitler* (Steering Committee Chair), N. Abramson, A. Bradhurst, N. Celona-Jacobs, J. Higgins, M.M. Kelsey, G. Klingensmith, K. Nadeau, T. Witten; University of Oklahoma Health Sciences Center: K. Copeland* (Steering Committee Vice-Chair), E. Boss, R. Brown, J. Chadwick, L. Chalmers, S. Chernausek, A. Hebensperger, C. Macha, R. Newgent, A. Nordyke, D. Olson, T. Poulsen, L. Pratt, J. Preske, J. Schanuel, S. Sternlof; University of Texas Health Science Center at San Antonio: J. Lynch*, N. Amodei, R. Barajas, C. Cody, D. Hale, J. Hernandez, C. Ibarra, E. Morales, S. Rivera, G. Rupert, A. Wauters; Washington University, St Louis: N. White*, A. Arbeláez, D. Flomo, J. Jones, T. Jones, M. Sadler, M. Tanner, A. Timpson, R. Welch; and Yale University: S. Caprio*, M. Grey, C. Guandalini, S. Lavietes, P. Rose, A. Syme, W. Tamborlane.
Coordinating Center--George Washington University Biostatistics Center: K. Hirst*, S. Edelstein, P. Feit, N. Grover, C. Long, L. Pyle.
Project Office--National Institute of Diabetes and Digestive and Kidney Diseases: B. Linder*.
Central Units--Central Blood Laboratory (Northwest Lipid Research Laboratories, University of Washington): S.M. Marcovina*, J. Harting; DEXA Reading Center (University of California at San Francisco): J. Shepherd*, B. Fan, L. Marquez, M. Sherman, J. Wang; Diet Assessment Center (University of South Carolina): M. Nichols*, E. Mayer-Davis, Y. Liu; Echocardiogram Reading Center (Johns Hopkins University): J. Lima*, S Gidding, J. Puccella, E. Ricketts; Fundus Photography Reading Center (University of Wisconsin): R. Danis*, A. Domalpally, A. Goulding, S. Neill, P. Vargo; Lifestyle Program Core (Washington University): D. Wilfley*, D. Aldrich-Rasche, K. Franklin, C. Massmann, D. O’Brien, J. Patterson, T. Tibbs, D. Van Buren.
Other--Hospital for Sick Children, Toronto: M. Palmert Medstar Research Institute, Washington DC: R. Ratner; Texas Tech University Health Sciences Center: D. Dremaine; University of Florida: J. Silverstein.
Footnotes
Registered with ClinicalTrials.gov NCT00081328
The other authors declare no conflicts of interest.
Funding information is available at www.jpeds.com (Appendix 2). Donations received from the following, but none participated in study design, conduct, data analysis, or report: Becton, Dickinson and Company, Bristol-Myers Squibb, Eli Lilly and Company, GlaxoSmithKline, LifeScan, Inc, Pfizer, Sanofi Aventis. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, the respective Tribal and Indian Health Service Institution Review Boards, or their members. L.K. is a consultant for Takeda and Janssen pharmaceutical.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Contributor Information
Carolyn E. Ievers-Landis, Department of Pediatrics, Division of Developmental/Behavioral Pediatrics and Psychology, Case Western Reserve University School of Medicine.
Natalie Walders-Abramson, Department of Pediatrics, Section of Endocrinology, University of Colorado School of Medicine.
Nancy Amodei, Department of Pediatrics, School of Medicine, University of Texas Health Science Center at San Antonio.
Kimberly L. Drews, Biostatistics Center, George Washington University.
Joan Kaplan, Child and Adolescent Psychiatry and Behavioral Science, Children’s Hospital of Philadelphia.
Lorraine E. Levitt Katz, Division of Endocrinology, Children’s Hospital of Philadelphia.
Sylvia Lavietes, Department of Pediatric Diabetes, Yale University.
Ron Saletsky, Department of Pediatrics: Child and Adolescent Psychiatry, SUNY Upstate Medical University.
Daniel Seidman, Medical Psychology, Columbia University Medical Center.
Patrice Yasuda, Department of Pediatics, Keck School of Medicine of University of Southern California.
References
- 1.Pinhas-Hamiel O, Zeitler P. Acute and chronic complications of type 2 diabetes mellitus in children and adolescents. Lancet. 2007;369:1823–31. doi: 10.1016/S0140-6736(07)60821-6. [DOI] [PubMed] [Google Scholar]
- 2.TODAY Study Group. A clinical trial to maintain glycemic control in youth with type 2 diabetes. N Engl J Med. 2012;366:2247–56. doi: 10.1056/NEJMoa1109333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Buchmann AF, Schmid B, Blomeyer D, Becker K, Treutlein J, Zimmerman US, et al. Impact of age at first drink on vulnerability to alcohol-related problems: Testing the marker hypothesis in a prospective study of young adults. Journal of Psychiatric Research. 2009;43:1205–12. doi: 10.1016/j.jpsychires.2009.02.006. [DOI] [PubMed] [Google Scholar]
- 4.Fidler JA, Wardle J, Brodersen NH, Jarvis MJ, West R. Vulnerability to smoking after trying a single cigarette can lie dormant for three years or more. Tob Control. 2006;15:205–09. doi: 10.1136/tc.2005.014894. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hegmann KT, Fraser AM, Keaney RP, Moser SE, Nilasena DS, Sedlars M, et al. The effect of age at smoking initiation on lung cancer risk. Epidemiology. 1993;4:444–48. doi: 10.1097/00001648-199309000-00010. [DOI] [PubMed] [Google Scholar]
- 6.Hingson RW, Heeren T, Winter MR. Age at drinking onset and alcohol dependence: Age at onset, duration, and severity. Arch Pediatr Adolesc Med. 2006;160:739–46. doi: 10.1001/archpedi.160.7.739. [DOI] [PubMed] [Google Scholar]
- 7.Hingson RW, Zha W. Age of drinking onset, alcohol use disorders, frequent heavy drinking, and unintentionally injuring oneself and others after drinking. Pediatrics. 2009;123:1477–84. doi: 10.1542/peds.2008-2176. [DOI] [PubMed] [Google Scholar]
- 8.Zakrajsek JS, Shope JT. Longitudinal examination of underage drinking and subsequent drinking and risky driving. Journal of Safety Research. 2006;37:443–51. doi: 10.1016/j.jsr.2006.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Reynolds K, Liese AD, Anderson AM, Dabelea D, Standiford D, Daniels SR, et al. Prevalence of tobacco use and association between cardiometabolic risk factors and cigarette smoking in youth with type 1 or type 2 diabetes mellitus. J Pediatr. 2011;158:594–601. doi: 10.1016/j.jpeds.2010.10.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Gau SSF, Chong M, Yang P, Yen C, Liang K, Cheng ATA. Psychiatric and psychosocial predictors of substance use disorders among adolescents: Longitudinal study. The British Journal of Psychiatry. 2007;190:42–48. doi: 10.1192/bjp.bp.106.022871. [DOI] [PubMed] [Google Scholar]
- 11.Gerrard M, Gibbons F, Michelle S, Vande Lune LS, Cleveland MJ. Images of smokers and willingness to smoke among African American pre-adolescents: An application of the prototype/willingness model of adolescent health risk behavior to smoking initiation. Journal of Pediatric Psychology. 2005;30:305–18. doi: 10.1093/jpepsy/jsi026. [DOI] [PubMed] [Google Scholar]
- 12.Simantov E, Schoen C, Klein JD. Health-compromising behaviors: Why do adolescents smoke or drink? Identifying underlying risk and protective factors. Arch Pediatr Adolesc Med. 2000;154:1025–33. doi: 10.1001/archpedi.154.10.1025. [DOI] [PubMed] [Google Scholar]
- 13.Ievers-Landis CE, Greenley NR, Burant C, Borawski E. Cognitive social maturity, life change events, and health risk behaviors among adolescents: Development of a structural equation model. Journal of Clinical Psychology in Medical Settings. 2006;13:111–20. [Google Scholar]
- 14.Hofer SE, Rosenbauer J, Gruuch-Henn J, Naeke A, Frohuch-Reiterer E, Holl RW. Smoking and metabolic control in adolescents with type 1 diabetes. J Pediatr. 2009;154:20–3. doi: 10.1016/j.jpeds.2008.07.052. [DOI] [PubMed] [Google Scholar]
- 15.Dowdell BE, Santucci ME. Health risk behavior assessment: Nutrition, weight, and tobacco use in one urban seventh-grade class. Public Health Nursing. 2004;21:128–36. doi: 10.1111/j.0737-1209.2004.021206.x. [DOI] [PubMed] [Google Scholar]
- 16.Shan X, Xi B, Cheng H, Hou D, Wang Y, Mi J. Prevalence and behavioral risk factors of overweight and obesity among children aged 2 to 18 in Beijing, China. International Journal of Pediatric Obesity. 2010;5:383–89. doi: 10.3109/17477160903572001. [DOI] [PubMed] [Google Scholar]
- 17.Schwab KO, Doerfer J, Hallermann K, Krebs A, Schorb E, Kristin K, et al. Marked smoking-associated increase of cardiovascular risk in childhood type 1 diabetes. International Journal of Adolescent Medicine and Health. 2008;20:285–92. doi: 10.1515/ijamh.2008.20.3.285. [DOI] [PubMed] [Google Scholar]
- 18.Gold M, Gladstein J. Substance use among adolescents with diabetes mellitus: Preliminary findings. Journal of Adolescent Health. 1993;14:80–84. doi: 10.1016/1054-139x(93)90089-8. [DOI] [PubMed] [Google Scholar]
- 19.TODAY Study Group. Treatment options for type 2 diabetes in adolescents and youth: a study of the comparative efficacy of metformin alone or in combination with rosiglitazone or lifestyle intervention in adolescents with type 2 diabetes. Pediatr Diabetes. 2007;8:74–87. doi: 10.1111/j.1399-5448.2007.00237.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Centers for Disease Control (CDC) National Center for Health Statistics 2000. CDC growth charts for the United States. Atlanta Centers for Disease Control; 2009. Available from: http://www.cdc.gov/growthcharts. [Google Scholar]
- 21.Kovacs M. Children’s Depression Inventory Manual. North Tonawanda, NY: Multi-Health Systems; 1992. [Google Scholar]
- 22.Beck AT, Steer RA. Beck Depression Inventory II. San Antonio, TX: Psychological Corp; 1996. [Google Scholar]
- 23.Eaton DK, Kann L, Kinchen S, Shanklin S, Ross J, Hawkins J, et al. Youth risk behavior surveillance -- United States, 2009. MMWR Surveill Summ. 2010;59:1–142. [PubMed] [Google Scholar]
- 24.Kandel DB, Hua M, Griesler PC, Schaffran C. On the development of nicotine dependence in adolescence. Drug and Alcohol Dependence. 2007;91:26–39. doi: 10.1016/j.drugalcdep.2007.04.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE. Monitoring the Future national results on drug use: 2012 Overview, Key Findings on Adolescent Drug Use. Ann Arbor: Institute for Social Research, The University of Michigan; 2013. [Google Scholar]
- 26.Copeland KC, Zeitler P, Geffner M, Guandalini C, Higgins J, Hirst K, et al. Characteristics of adolescents and youth with recent-onset type 2 diabetes: the TODAY cohort at baseline. J Clin Endocrinol Metab. 2011;96:159–67. doi: 10.1210/jc.2010-1642. [DOI] [PMC free article] [PubMed] [Google Scholar]