Date of Award


Document Type

Open Access Dissertation


Epidemiology and Biostatistics



First Advisor

Anwar T. Merchant


BACKGROUND AND OBJECTIVES: Television watching and computer use are considered to be the main sedentary behaviors in children and youth. However, their longitudinal impact on HbA1c and serum lipids among youth with type 1 diabetes (T1D) and type 2 diabetes (T2D) are under-studied. Sedentary behavior is normally assessed by questionnaire in large epidemiological studies, however, its measurement by questionnaire has not been evaluated among youth with diabetes. Also, no recent studies have evaluated the optimal meal frequency for glycemic control in youth with T1D and T2D; for this reason nutritional guidelines on meal frequency for youth with diabetes are vague. In this study we investigated the longitudinal relationship between television watching, computer use and meal frequency and HbA1c and serum lipids among youth with T1D and T2D; and evaluated the performance of questions to measure sedentary behavior in youth with diabetes.

RESEARCH DESIGN AND METHODS: This study included 1049 US youth (≥ 10 years old at the initial visit) with T1D and T2D who participated in the SEARCH for Diabetes in Youth Study and provided baseline data from 2002 to 2005. These participants were followed-up prospectively at 1, 2 and 5 years after the initial visit. All participants included in these analyses had physician diagnosed diabetes, documented year of diagnosis, were less than 20 years old on December 31 of the year of diagnosis, and attended at least one follow-up visit. Data collection procedures were performed by trained and certified staff following standardized protocols for both initial and follow-up visits,. Data collection approaches included questionnaires, physical examination and laboratory tests. Between July 2003 and March 2006, SEARCH cases aged ≥10 years who participated in the on-site SEARCH visit were invited to participate in the SEARCH-CC study in South Carolina and Colorado centers. All youth without diabetes were selected from health care provider offices in the same geographic areas as cases. Overall, 49% (n=220) of invited youth without diabetes participated in SEARCH CC study.

RESULTS: Increased television watching on weekdays and during the week over time was associated with larger increases in HbA1c among youth with T1D and T2D (p-value<0.05). Among youth with T1D, significant positive longitudinal associations were observed between television watching and TG (p-value<0.05) (week days and whole week), and LDL (p-value<0.05) (whole week). The overall correlations between hours of television watching from the YRBS and number of half hour television watching blocks from the 3DPAR were 0.30 (p<0.05) and 0.45 (p<0.05) among youth with T1D and T2D respectively, and 0.41 (p<0.05) among youth without diabetes. The correlations tended to be higher in females, older participants, normal weight T1D participants and overweight T2D participants. Similarly, African Americans and Hispanics with T2D tended to have higher Pearson correlations. The YRBS questions tended to overestimate (by 0.23, 0.54 and 0.40 for youth with T1D, T2D and without diabetes respectively) television watching compared with 3DPAR after adjusting for age, gender, and race. Among youth with T1D, HbA1c increase over time was higher for those who consumed 1-3 meals/day compared with those who consumed 4-5 meals/day (reference group); changes in HbA1c for individuals who ate 6-10 meals/day were not different from the reference group after adjusting for potential confounders including BMI z-scores and total energy intake. This association was stronger among youth who were ≥ 15 years (p-value for interaction<0.05) and African Americans (p-value for interaction<0.05). Youth who consumed 6-10 meals/day consumed more calories on average than youth eating 6-10 meals/day. Meal frequency was not associated with changes in serum lipids among youth with T1D and T2D.

CONCLUSIONS: Youth with T2D who increased their television watching over time vs those that decreased it had larger increases in HbA1c over 5 years. Youth with T1D who increased their television watching over time had increases in LDL, TG and to a lesser extent HbA1c. Television watching measured by YRBS questions showed weak to moderate correlation with television watching measured by 3DPAR among youth with diabetes. The correlation was stronger among youth who were older, female, overweight, African American and Hispanic. Youth with T1D who ate 1-3 meals/day had higher HbA1c over 5 years compared with those who ate 4-5 meals/day, but not those who ate 6-10 meals/day. Frequent meals without increasing total energy intake may be beneficial for youth with T1D.

Included in

Epidemiology Commons