This study examined parental factors associated with outcomes of youth in the Treatment Options for type 2 Diabetes in Adolescents and Youth (TODAY) clinical trial.
Of 699 youth with type 2 diabetes in the TODAY cohort, 623 (89.1%) had a parent participate and provide data at baseline, including weight, HbA1c, blood pressure, symptoms of depression, binge eating (BE), and medical history. Youth were followed 2–6.5 years. Data were analyzed using regression models and survival curve methods.
Parental diabetes (43.6% of parents) was associated with higher baseline HbA1c (P < 0.0001) and failure of youths to maintain glycemic control on study treatment (53.6% vs. 38.2% failure rate among those without a diabetic parent, P = 0.0002). Parental hypertension (40.6% of parents) was associated with hypertension in youth during TODAY (40.4% vs. 27.4% of youth with and without parental hypertension had hypertension, P = 0.0008) and with higher youth baseline BMI z scores (P = 0.0038). Parents had a mean baseline BMI of 33.6 kg/m2. Parental obesity (BMI >30 kg/m2) was associated with higher baseline BMI z scores in the youth (P < 0.0001). Depressive symptoms in parents (20.6% of parents) were related to youth depressive symptoms at baseline only (P = 0.0430); subclinical BE in parents was related to the presence of subclinical BE (P = 0.0354) and depressive symptoms (P = 0.0326) in youth throughout the study period.
Parental diabetes and hypertension were associated with lack of glycemic control, hypertension, and higher BMI z scores in youth. Further research is needed to better understand and address parental biological and behavioral factors to improve youth health outcomes.
Introduction
Youth-onset type 2 diabetes has been on the rise for two decades in the U.S. Diabetes-related complications and comorbidities, including retinopathy, hypertension, and albuminuria, are now observed in adolescents (1–8). Treatment Options for type 2 Diabetes in Adolescents and Youth (TODAY) was one of the first large randomized clinical trials to evaluate the safety and efficacy of different treatment regimens (9). The multicenter study group enrolled and monitored the largest ethnically diverse and well-characterized cohort of youth with type 2 diabetes in the world.
A family history of type 2 diabetes increases the risk of children developing diabetes (10), but whether the presence of diabetes in the parent is associated with more difficulty in achieving and maintaining glycemic control in youth and/or the earlier development of diabetes-related complications is unknown. The physical and mental health status of a parent may affect the general health or functioning of the child (11,12). However, whether the presence of parental obesity, hypertension, and/or depression is associated with greater hypertension, obesity, and depressive symptoms in their children with type 2 diabetes has not been well studied. The TODAY study collected data from the parents of youth participants, allowing for analyses that could help increase our understanding of how parental health status may affect the health and behavior of youth with type 2 diabetes. This information could inform the future development of targeted parent/child interventions to improve outcomes for youth with type 2 diabetes. The current analyses examined the following questions in TODAY study participants:
Is the presence of diabetes in a parent associated with poorer glycemic outcomes (failure to maintain glycemic control), youth medication adherence, BMI z score, and diabetes-related complications?
Is the presence of parental hypertension or obesity associated with hypertension or obesity in the offspring?
Is the presence of depressive symptoms or binge eating (BE) in parents associated with youth outcomes?
Is the occurrence of medical serious adverse events (SAEs) in parents (reflecting major medical problems in parents according to standard clinical trial SAE criteria) associated with failure to maintain glycemic control in their youth, youth medication adherence, BMI z score, and diabetes-related complications?
Research Design and Methods
The design of the TODAY study has been previously described (13,14). Overweight youth (n = 699, ages 10–17) with recent-onset type 2 diabetes (median 5 months) were randomized to receive metformin alone, metformin with rosiglitazone, or metformin with an intensive lifestyle intervention. The primary outcome was time to treatment failure based on glycemic control (HbA1c >8% [>64 mmol/mol]) over 6 months or inability to wean from temporary insulin therapy within 3 months after acute metabolic decompensation. Data collected on youth at baseline included age, sex, race/ethnicity, and family structure (with whom the participant lived). Height, weight, HbA1c, and blood pressure were assessed in youth at baseline, every 2 months in the first year, and quarterly thereafter. Urine albumin and creatinine were assessed at baseline and annually. Medication adherence based on pill count was assessed at each follow-up visit with a goal of ≥80%. SAEs (based on standard definitions for clinical trials) that occurred in youth during the study were also collected. Hypertension was defined as high blood pressure (≥130/80 mmHg or ≥95th percentile for age, sex, and height based on Centers for Disease Control and Prevention normative data) or use of appropriate medication. Microalbuminuria was defined as an albumin-to-creatinine ratio ≥30 µg/mg in two of three urine samples collected over ≥3 months or use of appropriate medication. Retinopathy was assessed by fundus photography in the last year of the study (7).
Depressive symptoms during the past 2 weeks were assessed using the Children’s Depression Inventory (CDI) for participants up to age 16 or the Beck Depression Inventory (BDI) for those 16 or older. A score ≥13 on the CDI or ≥14 on the BDI indicated clinically significant levels of depressive symptoms (15,16). Responses on the self-report Youth Eating Disorder Examination Questionnaire (Y-EDEQ) were used to identify the presence of BE, defined as the number of times in the past 28 days the participant reported episodes of overeating with an associated loss of control. Subclinical BE was defined as reporting 1–3 BE episodes in the past month, and clinical BE was ≥4 (17). The BDI, CDI, and Y-EDEQ were administered to the youth at baseline, month 6, and month 24 of the study. Youth hypertension, microalbuminuria, depressive symptoms, subclinical BE, and clinical BE were assessed at baseline and cumulatively during the entire study period (including baseline through end of study visits).
Each youth was accompanied by an adult caregiver, called the family support person (FSP), most of whom were parents. The FSP had to provide signed informed consent stating willingness to perform functions, including accompanying the youth to all study visits and participating in data collection. Baseline data were collected from FSPs, including age, sex, race/ethnicity, height, and weight measured by clinic staff, depressive symptoms (BDI), and an adult version of the EDEQ to determine subclinical and clinical BE defined in a similar manner as for the youth (18). History of diabetes and hypertension were obtained from the biological FSP parent via self-report at baseline. Highest household level of education and total household annual income were recorded. SAEs occurring in FSPs, such as death, hospitalization, birth of a baby with a congenital anomaly, disability, or any other life-threatening medical event, were documented during follow-up.
The protocol was approved by the institutional review board for each of the participating institutions. All participants provided informed consent and minor child assent.
Statistical Methods
Associations between parental characteristics and youth outcomes were analyzed using logistic regression models (for binary outcomes) and general linear models (for continuous outcomes).
Comparison of glycemic failure on assigned treatment in the youth by parental diabetes status and diagnosis of youth hypertension by parental hypertension were analyzed using survival curve methodology with log-rank tests. At baseline, all TODAY youth were free of glycemic failure per study protocol but could be enrolled if previously diagnosed with hypertension that was controlled with appropriate medical therapy.
All analyses were considered exploratory, with statistical significance defined as P < 0.05. SAS 9.3 software (SAS Institute Inc., Cary, NC) was used for all analyses.
Results
Of the 699 TODAY youth participants, 623 (89.1%) had a parent join as the FSP at baseline and were included in the analysis sample; 88.6% of FSPs were mothers (97.6% biological) and 11.4% were fathers (95.8% biological). There were no differences in baseline youth characteristics (sex, race/ethnicity, age, BMI, HbA1c, blood pressure) between those included or not included in the analyses. Characteristics of the parents and youth are reported in Table 1. At baseline, 63.3% of parents were obese, 43.6% reported diabetes, 40.6% reported hypertension, and 74.2% reported a total annual household income of <$50,000. Youth characteristics have been previously described (5,7,9,19–22): 63.4% were female, at baseline 42.6% lived with both parents, their mean BMI z score was 2.23, and 45.4% were unable to maintain glycemic control regardless of assigned treatment during an average follow-up of 3.8 years.
. | Parent at baseline* . | Youth at baseline . | Youth during study** . |
---|---|---|---|
Age (years) | 41.5 ± 6.7 | 13.9 ± 2.0 | — |
Female | 88.6 | 63.4 | — |
Race/ethnicity | |||
Non-Hispanic Black | 33.8 | 32.1 | — |
Hispanic | 35.5 | 40.0 | — |
Non-Hispanic White | 23.4 | 21.2 | — |
Other | 7.3 | 6.7 | — |
Household highest level of education | |||
Less than high school | 25.7 | — | — |
High school or GED | 24.8 | — | — |
Some college | 32.6 | — | — |
College degree or higher | 16.9 | — | — |
Household total annual income | |||
<$25,000 | 40.4 | — | — |
$25,000–$49,999 | 33.8 | — | — |
≥$50,000 | 25.8 | — | — |
Family structure | |||
Living with both parents | — | 42.6 | — |
Living with either one | — | 54.4 | — |
Living with neither | — | 3.0 | — |
BMI (kg/m2) | 33.6 ± 7.8 | 35.0 ± 7.7 | 36.0 ± 7.9 |
BMI >30 kg/m2 | 63.3 | 71.4 | 76.9 |
BMI z score | — | 2.23 ± 0.47 | 2.18 ± 0.50 |
Diabetes | 43.6 | 100 | — |
HbA1c, % (mmol/mol) | — | 6.0 ± 0.8 (42 ± 9) | 7.1 ± 1.6 (54 ± 17) |
Medication adherence ≥80% | — | — | 57.1 |
SAE(s) reported ≥1 | 16.4 | — | 21.5 |
Hypertension | 40.6 | 17.8 | 32.6 |
Diabetic retinopathy | — | — | 14.1 |
Microalbuminuria | — | 5.9 | 18.1 |
CDI score | — | 7.1 ± 6.4 | — |
BDI score | 8.3 ± 8.8 | 6.0 ± 7.5 | — |
Depressive symptoms† | 20.6 | 15.0 | 27.9 |
Categories of BE‡ | |||
Subclinical BE | 10.8 | 17.1 | 40.9 |
Clinical BE | 5.3 | 8.3 | 17.1 |
. | Parent at baseline* . | Youth at baseline . | Youth during study** . |
---|---|---|---|
Age (years) | 41.5 ± 6.7 | 13.9 ± 2.0 | — |
Female | 88.6 | 63.4 | — |
Race/ethnicity | |||
Non-Hispanic Black | 33.8 | 32.1 | — |
Hispanic | 35.5 | 40.0 | — |
Non-Hispanic White | 23.4 | 21.2 | — |
Other | 7.3 | 6.7 | — |
Household highest level of education | |||
Less than high school | 25.7 | — | — |
High school or GED | 24.8 | — | — |
Some college | 32.6 | — | — |
College degree or higher | 16.9 | — | — |
Household total annual income | |||
<$25,000 | 40.4 | — | — |
$25,000–$49,999 | 33.8 | — | — |
≥$50,000 | 25.8 | — | — |
Family structure | |||
Living with both parents | — | 42.6 | — |
Living with either one | — | 54.4 | — |
Living with neither | — | 3.0 | — |
BMI (kg/m2) | 33.6 ± 7.8 | 35.0 ± 7.7 | 36.0 ± 7.9 |
BMI >30 kg/m2 | 63.3 | 71.4 | 76.9 |
BMI z score | — | 2.23 ± 0.47 | 2.18 ± 0.50 |
Diabetes | 43.6 | 100 | — |
HbA1c, % (mmol/mol) | — | 6.0 ± 0.8 (42 ± 9) | 7.1 ± 1.6 (54 ± 17) |
Medication adherence ≥80% | — | — | 57.1 |
SAE(s) reported ≥1 | 16.4 | — | 21.5 |
Hypertension | 40.6 | 17.8 | 32.6 |
Diabetic retinopathy | — | — | 14.1 |
Microalbuminuria | — | 5.9 | 18.1 |
CDI score | — | 7.1 ± 6.4 | — |
BDI score | 8.3 ± 8.8 | 6.0 ± 7.5 | — |
Depressive symptoms† | 20.6 | 15.0 | 27.9 |
Categories of BE‡ | |||
Subclinical BE | 10.8 | 17.1 | 40.9 |
Clinical BE | 5.3 | 8.3 | 17.1 |
Data are presented as mean ± SD or %.
*Highest level of education and annual income refer to information on the household; parental diabetes and hypertension refer to information on the biological FSP parent; all other parent factors refer to information on the primary caregiver parent at baseline except for SAE(s) reported, which were based on data collected during the study.
**Youth factors during the study include baseline data.
†The presence of clinically significant mood impairment or depressive symptoms within the last 2 weeks was indicated by a CDI score ≥13 (for participants <16 years old) or a BDI score ≥14 (for participants ≥16). At baseline, 24% of the youth were ≥16 years old and completed the BDI; all others completed the CDI.
‡BE episodes were the number of times in the past 28 days the participant reported episodes of overeating with an associated loss of control; subclinical BE was defined as 1–3 episodes, and clinical BE was defined as ≥4 episodes.
Each parent characteristic from Table 1 (apart from age, sex, and race/ethnicity) was univariately related to each youth outcome collected at baseline or during the study. Tables 2 and 3 present the results from these associations, and significant results are further discussed below. Parental diabetes was significantly related to youth glycemic control (Table 2 and Fig. 1A). Baseline HbA1c was significantly higher in youth who had a parent with diabetes (mean 6.2% [SD 0.8]; 44 mmol/mol [SD 9]) compared with those whose parent did not have diabetes (mean 5.9% [SD 0.7]; 41 mmol/mol [SD 8]; P < 0.0001). Of those with a parent who had diabetes, 53.6% failed to maintain glycemic control (reached primary outcome) compared with 38.2% of those without a parent with diabetes (P = 0.0002). Figure 1A shows treatment failure by parental diabetes status, reflected in the statistically significantly higher youth baseline HbA1c in those whose parents had diabetes and the widening difference between failure rates over time. When analyses evaluating parental diabetes were restricted to only FSP biological mothers (n = 532), all the associations between maternal diabetes and youth glycemic control remained (P = 0.0002). Given the small number of FSP biological fathers (n = 67), similar analyses were not performed for the fathers.
Parent factors . | Youth factors . | |||||||
---|---|---|---|---|---|---|---|---|
Baseline . | During study** . | |||||||
BMI z . | HbA1c (%) . | HbA1c (mmol/mol) . | Hypertension (%) . | Medication adherence ≥80% (%) . | SAE(s) reported (%) . | Hypertension (%) . | Loss of glycemic control (%) . | |
BMI >30 kg/m2 | ||||||||
No | 2.04 ± 0.47 | 6.0 ± 0.7 | 42 ± 8 | 13.7 | 60.7 | 12.9 | 25.0 | 36.3 |
Yes | 2.29 ± 0.40 | 6.0 ± 0.8 | 42 ± 9 | 14.0 | 59.3 | 22.4 | 33.2 | 46.7 |
P value | <0.0001 | 0.6502 | 0.6502 | 0.9369 | 0.8109 | 0.0275 | 0.1115 | 0.0606 |
Diabetes | ||||||||
No | 2.23 ± 0.46 | 5.9 ± 0.7 | 41 ± 8 | 14.8 | 53.4 | 18.9 | 32.5 | 38.2 |
Yes | 2.24 ± 0.47 | 6.2 ± 0.8 | 44 ± 9 | 21.8 | 60.9 | 23.4 | 32.6 | 53.6 |
P value | 0.7910 | <0.0001 | <0.0001 | 0.0262 | 0.0724 | 0.1863 | 0.9953 | 0.0002 |
SAE reported | ||||||||
None | 2.21 ± 0.47 | 6.0 ± 0.8 | 42 ± 9 | 16.3 | 57.4 | 20.4 | 31.7 | 43.0 |
≥1 | 2.32 ± 0.45 | 6.1 ± 0.8 | 43 ± 9 | 25.5 | 55.6 | 27.5 | 37.3 | 57.8 |
P value | 0.0317 | 0.1647 | 0.1647 | 0.0330 | 0.7412 | 0.1188 | 0.2754 | 0.0060 |
Hypertension | ||||||||
No | 2.19 ± 0.47 | 5.9 ± 0.7 | 41 ± 8 | 13.1 | 59.1 | 19.6 | 27.4 | 46.9 |
Yes | 2.30 ± 0.45 | 6.2 ± 0.8 | 44 ± 9 | 24.9 | 53.5 | 22.9 | 40.4 | 42.9 |
P value | 0.0038 | 0.0004 | 0.0004 | 0.0002 | 0.1899 | 0.3286 | 0.0008 | 0.3237 |
Depressive symptoms | ||||||||
No | 2.23 ± 0.46 | 6.0 ± 0.7 | 42 ± 8 | 18.1 | 57.5 | 19.2 | 33.4 | 44.5 |
Yes | 2.22 ± 0.48 | 6.1 ± 0.7 | 43 ± 8 | 16.7 | 55.9 | 27.0 | 29.4 | 49.2 |
P value | 0.8328 | 0.1252 | 0.1252 | 0.6976 | 0.7566 | 0.0601 | 0.3858 | 0.3489 |
Subclinical BE† | ||||||||
No | 2.22 ± 0.48 | 6.0 ± 0.8 | 42 ± 9 | 17.9 | 56.5 | 21.4 | 33.0 | 45.3 |
Yes | 2.29 ± 0.35 | 6.1 ± 0.8 | 43 ± 9 | 16.7 | 65.6 | 24.2 | 28.8 | 51.5 |
P value | 0.2700 | 0.6707 | 0.6707 | 0.8087 | 0.1574 | 0.5963 | 0.4899 | 0.3399 |
Clinical BE | ||||||||
No | 2.23 ± 0.46 | 6.0 ± 0.8 | 42 ± 9 | 17.9 | 58.2 | 21.3 | 32.8 | 46.5 |
Yes | 2.25 ± 0.54 | 6.2 ± 0.8 | 44 ± 9 | 15.6 | 43.3 | 28.1 | 28.1 | 37.5 |
P value | 0.7898 | 0.1893 | 0.1893 | 0.7445 | 0.1103 | 0.3773 | 0.5817 | 0.3196 |
Household annual income | ||||||||
<$25,000 | 2.24 ± 0.46 | 6.0 ± 0.8 | 42 ± 9 | 21.1 | 53.4 | 26.8 | 33.3 | 46.9 |
$25,000–$49,999 | 2.24 ± 0.47 | 6.0 ± 0.8 | 42 ± 9 | 13.1 | 58.8 | 21.5 | 31.9 | 47.1 |
≥$50,000 | 2.20 ± 0.47 | 6.0 ± 0.7 | 42 ± 8 | 21.2 | 64.2 | 17.1 | 32.9 | 37.0 |
P value | 0.6917 | 0.5084 | 0.5084 | 0.0582 | 0.1292 | 0.0834 | 0.9542 | 0.1075 |
Parent factors . | Youth factors . | |||||||
---|---|---|---|---|---|---|---|---|
Baseline . | During study** . | |||||||
BMI z . | HbA1c (%) . | HbA1c (mmol/mol) . | Hypertension (%) . | Medication adherence ≥80% (%) . | SAE(s) reported (%) . | Hypertension (%) . | Loss of glycemic control (%) . | |
BMI >30 kg/m2 | ||||||||
No | 2.04 ± 0.47 | 6.0 ± 0.7 | 42 ± 8 | 13.7 | 60.7 | 12.9 | 25.0 | 36.3 |
Yes | 2.29 ± 0.40 | 6.0 ± 0.8 | 42 ± 9 | 14.0 | 59.3 | 22.4 | 33.2 | 46.7 |
P value | <0.0001 | 0.6502 | 0.6502 | 0.9369 | 0.8109 | 0.0275 | 0.1115 | 0.0606 |
Diabetes | ||||||||
No | 2.23 ± 0.46 | 5.9 ± 0.7 | 41 ± 8 | 14.8 | 53.4 | 18.9 | 32.5 | 38.2 |
Yes | 2.24 ± 0.47 | 6.2 ± 0.8 | 44 ± 9 | 21.8 | 60.9 | 23.4 | 32.6 | 53.6 |
P value | 0.7910 | <0.0001 | <0.0001 | 0.0262 | 0.0724 | 0.1863 | 0.9953 | 0.0002 |
SAE reported | ||||||||
None | 2.21 ± 0.47 | 6.0 ± 0.8 | 42 ± 9 | 16.3 | 57.4 | 20.4 | 31.7 | 43.0 |
≥1 | 2.32 ± 0.45 | 6.1 ± 0.8 | 43 ± 9 | 25.5 | 55.6 | 27.5 | 37.3 | 57.8 |
P value | 0.0317 | 0.1647 | 0.1647 | 0.0330 | 0.7412 | 0.1188 | 0.2754 | 0.0060 |
Hypertension | ||||||||
No | 2.19 ± 0.47 | 5.9 ± 0.7 | 41 ± 8 | 13.1 | 59.1 | 19.6 | 27.4 | 46.9 |
Yes | 2.30 ± 0.45 | 6.2 ± 0.8 | 44 ± 9 | 24.9 | 53.5 | 22.9 | 40.4 | 42.9 |
P value | 0.0038 | 0.0004 | 0.0004 | 0.0002 | 0.1899 | 0.3286 | 0.0008 | 0.3237 |
Depressive symptoms | ||||||||
No | 2.23 ± 0.46 | 6.0 ± 0.7 | 42 ± 8 | 18.1 | 57.5 | 19.2 | 33.4 | 44.5 |
Yes | 2.22 ± 0.48 | 6.1 ± 0.7 | 43 ± 8 | 16.7 | 55.9 | 27.0 | 29.4 | 49.2 |
P value | 0.8328 | 0.1252 | 0.1252 | 0.6976 | 0.7566 | 0.0601 | 0.3858 | 0.3489 |
Subclinical BE† | ||||||||
No | 2.22 ± 0.48 | 6.0 ± 0.8 | 42 ± 9 | 17.9 | 56.5 | 21.4 | 33.0 | 45.3 |
Yes | 2.29 ± 0.35 | 6.1 ± 0.8 | 43 ± 9 | 16.7 | 65.6 | 24.2 | 28.8 | 51.5 |
P value | 0.2700 | 0.6707 | 0.6707 | 0.8087 | 0.1574 | 0.5963 | 0.4899 | 0.3399 |
Clinical BE | ||||||||
No | 2.23 ± 0.46 | 6.0 ± 0.8 | 42 ± 9 | 17.9 | 58.2 | 21.3 | 32.8 | 46.5 |
Yes | 2.25 ± 0.54 | 6.2 ± 0.8 | 44 ± 9 | 15.6 | 43.3 | 28.1 | 28.1 | 37.5 |
P value | 0.7898 | 0.1893 | 0.1893 | 0.7445 | 0.1103 | 0.3773 | 0.5817 | 0.3196 |
Household annual income | ||||||||
<$25,000 | 2.24 ± 0.46 | 6.0 ± 0.8 | 42 ± 9 | 21.1 | 53.4 | 26.8 | 33.3 | 46.9 |
$25,000–$49,999 | 2.24 ± 0.47 | 6.0 ± 0.8 | 42 ± 9 | 13.1 | 58.8 | 21.5 | 31.9 | 47.1 |
≥$50,000 | 2.20 ± 0.47 | 6.0 ± 0.7 | 42 ± 8 | 21.2 | 64.2 | 17.1 | 32.9 | 37.0 |
P value | 0.6917 | 0.5084 | 0.5084 | 0.0582 | 0.1292 | 0.0834 | 0.9542 | 0.1075 |
*Mean ± SD or % of the factor in the youth are shown in the table by the status of a factor in the parent. The P values are from unadjusted models testing for an association between the parental factor and the youth factor (P < 0.05 in bold).
**Youth factors during the study include baseline data.
†Clinical BE (≥4 episodes) was excluded from the comparison group of normal plus overeaters vs. subclinical BE (1–3 episodes).
Parent factors . | Youth factors . | |||||
---|---|---|---|---|---|---|
Depressive symptoms . | Subclinical BE† . | Clinical BE . | ||||
Baseline . | During study . | Baseline . | During study . | Baseline . | During study . | |
BMI >30 kg/m2 | ||||||
No | 11.5 | 25.0 | 15.3 | 29.6 | 3.3 | 12.2 |
Yes | 12.0 | 22.0 | 18.7 | 36.2 | 10.5 | 17.5 |
P value | 0.8945 | 0.5245 | 0.4339 | 0.2538 | 0.0119 | 0.1875 |
Diabetes | ||||||
No | 14.8 | 27.2 | 19.3 | 34.6 | 8.0 | 16.9 |
Yes | 14.5 | 27.2 | 17.2 | 37.1 | 7.9 | 16.1 |
P value | 0.9206 | 0.9965 | 0.5211 | 0.5609 | 0.9752 | 0.8021 |
SAE reported | ||||||
None | 14.9 | 28.2 | 19.1 | 33.9 | 8.2 | 17.4 |
1 or more | 15.2 | 26.5 | 16.3 | 45.4 | 8.9 | 15.7 |
P value | 0.9552 | 0.7183 | 0.5255 | 0.0462 | 0.8064 | 0.6798 |
Hypertension | ||||||
No | 14.3 | 27.9 | 17.2 | 33.5 | 8.0 | 15.4 |
Yes | 15.1 | 26.1 | 20.4 | 39.0 | 8.1 | 18.4 |
P value | 0.7667 | 0.6231 | 0.3535 | 0.2116 | 0.9864 | 0.3420 |
Depressive symptoms | ||||||
No | 13.2 | 26.2 | 17.6 | 34.7 | 7.9 | 15.6 |
Yes | 20.8 | 33.3 | 21.3 | 41.0 | 10.7 | 22.1 |
P value | 0.0430 | 0.1150 | 0.3750 | 0.2464 | 0.3246 | 0.0936 |
Subclinical BE† | ||||||
No | 14.5 | 26.5 | 17.1 | 34.4 | 8.0 | 16.4 |
Yes | 17.2 | 39.4 | 30.9 | 50.0 | 12.7 | 25.0 |
P value | 0.5786 | 0.0326 | 0.0183 | 0.0354 | 0.2268 | 0.0980 |
Clinical BE | ||||||
No | 15.1 | 27.7 | 18.7 | 35.6 | 8.2 | 17.0 |
Yes | 9.4 | 31.3 | 14.8 | 41.7 | 12.9 | 22.6 |
P value | 0.3462 | 0.6690 | 0.6048 | 0.5501 | 0.3915 | 0.4386 |
Household annual income | ||||||
<$25,000 | 16.2 | 28.1 | 18.0 | 34.4 | 6.4 | 16.6 |
$25,000–$49,999 | 20.1 | 33.5 | 20.2 | 40.3 | 9.2 | 14.5 |
≥$50,000 | 7.8 | 18.5 | 17.0 | 30.4 | 9.1 | 19.6 |
P value | 0.0053 | 0.0074 | 0.7591 | 0.2281 | 0.4863 | 0.4771 |
Parent factors . | Youth factors . | |||||
---|---|---|---|---|---|---|
Depressive symptoms . | Subclinical BE† . | Clinical BE . | ||||
Baseline . | During study . | Baseline . | During study . | Baseline . | During study . | |
BMI >30 kg/m2 | ||||||
No | 11.5 | 25.0 | 15.3 | 29.6 | 3.3 | 12.2 |
Yes | 12.0 | 22.0 | 18.7 | 36.2 | 10.5 | 17.5 |
P value | 0.8945 | 0.5245 | 0.4339 | 0.2538 | 0.0119 | 0.1875 |
Diabetes | ||||||
No | 14.8 | 27.2 | 19.3 | 34.6 | 8.0 | 16.9 |
Yes | 14.5 | 27.2 | 17.2 | 37.1 | 7.9 | 16.1 |
P value | 0.9206 | 0.9965 | 0.5211 | 0.5609 | 0.9752 | 0.8021 |
SAE reported | ||||||
None | 14.9 | 28.2 | 19.1 | 33.9 | 8.2 | 17.4 |
1 or more | 15.2 | 26.5 | 16.3 | 45.4 | 8.9 | 15.7 |
P value | 0.9552 | 0.7183 | 0.5255 | 0.0462 | 0.8064 | 0.6798 |
Hypertension | ||||||
No | 14.3 | 27.9 | 17.2 | 33.5 | 8.0 | 15.4 |
Yes | 15.1 | 26.1 | 20.4 | 39.0 | 8.1 | 18.4 |
P value | 0.7667 | 0.6231 | 0.3535 | 0.2116 | 0.9864 | 0.3420 |
Depressive symptoms | ||||||
No | 13.2 | 26.2 | 17.6 | 34.7 | 7.9 | 15.6 |
Yes | 20.8 | 33.3 | 21.3 | 41.0 | 10.7 | 22.1 |
P value | 0.0430 | 0.1150 | 0.3750 | 0.2464 | 0.3246 | 0.0936 |
Subclinical BE† | ||||||
No | 14.5 | 26.5 | 17.1 | 34.4 | 8.0 | 16.4 |
Yes | 17.2 | 39.4 | 30.9 | 50.0 | 12.7 | 25.0 |
P value | 0.5786 | 0.0326 | 0.0183 | 0.0354 | 0.2268 | 0.0980 |
Clinical BE | ||||||
No | 15.1 | 27.7 | 18.7 | 35.6 | 8.2 | 17.0 |
Yes | 9.4 | 31.3 | 14.8 | 41.7 | 12.9 | 22.6 |
P value | 0.3462 | 0.6690 | 0.6048 | 0.5501 | 0.3915 | 0.4386 |
Household annual income | ||||||
<$25,000 | 16.2 | 28.1 | 18.0 | 34.4 | 6.4 | 16.6 |
$25,000–$49,999 | 20.1 | 33.5 | 20.2 | 40.3 | 9.2 | 14.5 |
≥$50,000 | 7.8 | 18.5 | 17.0 | 30.4 | 9.1 | 19.6 |
P value | 0.0053 | 0.0074 | 0.7591 | 0.2281 | 0.4863 | 0.4771 |
*% of the factor in the youth are shown in the table by the status of a factor in the parent; P values from unadjusted models testing for an association between the parental factor and the youth factor (P < 0.05 in bold). Youth factors during the study include baseline data.
†Clinical BE (≥4 episodes) was excluded from the comparison group of normal plus overeaters vs. subclinical BE (1–3 episodes).
Parental diabetes also related to youth hypertension at baseline (P = 0.0262) but not to hypertension diagnosed in youth during the entire trial period. No association was found between diabetes in a parent and youth diabetic retinopathy or microalbuminuria (data not shown), in youth BMI z score at baseline, or in medication adherence during the study. Among parents with a BMI >30 kg/m2, the baseline mean (SD) BMI z score of the youth was 2.29 (0.40) compared with 2.04 (0.47) among youth with a nonobese parent (P < 0.0001). Changes in youth BMI and BMI z scores were minimal during the study and were not explored further.
Parental hypertension was related to hypertension in the youth. Whereas there were higher rates of hypertension at baseline in youth with a parent with hypertension (24.9% vs. 13.1%, P = 0.0002), the change in the percentage of youth free of hypertension during the entire study period was similar in those with and without parental hypertension (Fig. 1B). By the end of the study, 40.4% of the youth with parental hypertension had been diagnosed with hypertension compared with 27.4% of those with no hypertensive parent (P = 0.0008). In 143 parents with both diabetes and hypertension, 30.8% of youth had hypertension at baseline compared with 13.9% of youth with parents with either diabetes or hypertension or neither (P < 0.0001). Among the parents with and without hypertension, the mean (SD) BMI z score of the youth at baseline was 2.30 (0.45) and 2.19 (0.47), respectively (P = 0.0038). Parental hypertension also related to higher HbA1c at baseline (P = 0.0004).
The percentage of youth with significant depressive symptoms at baseline was 20.8% in participants with a parent with depressive symptoms at baseline versus 13.2% in those without a parent with depressive symptoms (P = 0.0430; Table 3); the relationship was no longer significant by the end of the study. Also, depressive symptoms in the parent did not relate to youth BMI z score or HbA1c at baseline and did not correlate with medication adherence, development of diabetes-related complications, or occurrence of SAEs in youth during the study.
Parental subclinical BE (1–3 episodes in the past month) at baseline, reported by 10.8% of the FSP parents, related to greater depressive symptoms in the youth by the end of the study (P = 0.0326) and to subclinical BE in the youth at baseline (P = 0.0183) and cumulatively by the end of the study (P = 0.0354). Parental clinical BE at baseline was not significantly related to any other youth outcomes, but was only present in 5.3% of parents.
Low household education level (less than college degree) was associated with a higher baseline BMI zs core in the offspring (data not shown; P = 0.0340) and development of microalbuminuria (data not shown; P = 0.0091). Low household annual income (<$50,000) was related to the presence of youth depressive symptoms at baseline (P = 0.0053) and cumulatively by the end of the study (P = 0.0074). Youth who failed randomized diabetes treatment assignment (loss of glycemic control) were more likely to have a parent with at least one SAE (57.8%) versus 43.0% of the TODAY youth with no SAEs in a parent (P = 0.0060). There were 195 SAEs in 111 FSPs, including 11 deaths (7 related to cardiovascular disease and 1 cancer). SAEs were predominantly hospitalizations (94%), and the most common were related to cardiovascular disease (26%), infections (11%), and renal disease (9%). Offspring of parents with SAEs had higher BMI z scores (P = 0.0317) and more hypertension at baseline (P = 0.0330) and were more likely to be classified with subclinical BE episodes by the end of the study (P = 0.0462). However, parental SAEs were not associated with medication adherence, diagnosis of hypertension by the end of the study, retinopathy, microalbuminuria, or depressive symptoms in the youth.
Conclusions
A history of type 2 diabetes in a first-degree relative is a well-known risk factor for diabetes developing at a young age (23,24), and TODAY has already reported that 59.6% of the study cohort had a nuclear family member with diabetes, which rose to 89.5% when grandparents were included (9). Our analyses have shown a significant relationship between glycemic control in youth with type 2 diabetes and diabetes diagnosed in a parent. Most striking is the very early relationship between parental diabetes status and youth glycemic control, followed by a continuing significant effect over time shown in Fig. 1A. An area of future research could include investigations to further our understanding on how parents’ glycemic control affects youth’s glycemic control where one or both parents have a diagnosis of type 2 diabetes.
This inequity early in the natural history of youth onset type 2 diabetes may be due to environmental, lifestyle, genetic, and epigenetic factors. The possible role of an adverse intrauterine environment on the development of diabetes was introduced more than two decades ago (25) and was further studied in the Pima Indians and other populations (26). The extent to which youth-onset type 2 diabetes is associated with specific genetic abnormalities is currently under investigation. TODAY previously reported a significant association between higher HbA1c at baseline and failure to maintain glycemic control, possibly related to greater deterioration in pancreatic β-cell function (19). These findings highlight the importance of identifying and closely monitoring youth with type 2 diabetes who may require early intensification of therapy, which occurs more frequently in those with a parental history of diabetes.
Many youth (17.8%) already had hypertension at baseline, which was more common in those with a parental history of hypertension. This is not a surprising result, considering the occurrence of youth hypertension may be largely genetically determined. Youth hypertension at baseline was also more common in those with parents with diabetes. It is possible that a pediatrician or family practitioner, knowing about the positive family history of hypertension and/or diabetes, would be more diligent in checking the status of the youth. The higher prevalence of youth with hypertension at baseline in the 143 parents with diabetes and hypertension (30.8% vs. 13.9% of youth with parents with diabetes or hypertension or neither) may be due to increased clinician attention as well as to the compounding effect of a parent with multiple diagnoses. Environmental influences, such as parental obesity, presence of depressive symptoms or BE, and household annual income, were not associated with hypertension in youth.
Elevated depressive symptoms in parents were associated with elevated depressive symptoms in youth at baseline. This is consistent with the large body of literature reporting that indicators of parental depression relate to a range of youth problems in cognitive, behavioral, and emotional domains (27). This has been shown for maternal and paternal depression (28,29). The effect of socioeconomic status (as expressed by household annual income) on the relationship between depressive symptoms in parents and youth as a confounder or mediator bears further study.
We did not find a relationship between symptoms of depression in parents and youth baseline BMI, baseline HbA1c, medication adherence, or development of complications. In several studies of youth in the community, parental depression has been associated with poor parenting skills and especially low levels of warmth and high levels of parental criticism (30,31). These factors have been found to be associated with poorer adherence and glycemic control in children with type 1 diabetes (32), although Butler et al. (33), who found a relationship between parenting style and youth depression in a group of adolescents with type 1 diabetes, found no relationship between parenting style and adherence. In a sample of adolescents with type 1 and type 2 diabetes, Eckshtain et al. (34) found that parental depression had an indirect effect on youth depression, via less parental involvement, and that youth depression was related to glycemic control. They reported that parents with elevated depressive symptoms performed less monitoring of youth and that this behavior was associated with poorer glycemic control, as was parent-reported youth depression. However, youth reports of depressive symptoms in TODAY did not relate to glycemic control. In the current study, we relied on youth self-report of depressive symptoms, and this method may explain why we found no relationship. Also, youth were involved in an active and highly supportive intervention with a number of caring adults, and these relationships might have ameliorated any negative effects of parental depression on their outcomes.
Disordered eating is a significant comorbidity that can complicate the medical management of diabetes. Children who report engaging in disordered eating behaviors are at greater risk for developing increased shape and weight concerns than those who do not report these behaviors and are at increased risk for developing type 2 diabetes (35,36). TODAY previously reported that at baseline, type 2 diabetes youth with BE had higher levels of obesity, greater global eating, and weight and shape concerns (21). An association between disordered eating in parents without diabetes, particularly mothers, and children has been previously reported (37). Twin studies suggest that this relationship is due to the moderate heritability of disinhibited eating moderated by a shared environment (38,39). We did not find an association between parental BE and youth outcomes in our study but did find an association between subclinical BE in parents and subclinical BE in their youth with type 2 diabetes, suggesting that assessments for disordered eating in parents as well as in youth may be important.
Although the links that we found between parental mental health difficulties (depression, disordered eating) and youth outcomes are not dramatic, they are of concern. The data cited earlier suggest that depression and disordered eating in parents may contribute to these and other problems in their children, which in turn may predict problems in adulthood. It is important for researchers to explore these outcomes more fully. Also, practitioners should take these mental health issues into account when planning their interventions. This might include assessments of depression and disordered eating in parents as well as youth and behavioral/psychosocial interventions that target the family rather than only the child, using a family-centered approach to care. In the TODAY study, a parent was involved in the intervention itself, but his or her role was to support the child, and any potential benefit for the parent would have been indirect. A true family-centered approach would help the parent focus on his or her own needs and problems as well. For example, it would be interesting to learn whether an intervention for a parent with diabetes that addresses parental glycemic control and weight would result in improved youth glycemic control and weight.
Limitations of this study include the small number of FSP participants who were fathers and the use of self-reported recent depressive symptoms (2 weeks) and eating behaviors (28 days) rather than clinical diagnoses derived from structured psychiatric interviews. Furthermore, we lack information concerning depression and BE in non-FSP parents. Because we did not measure parenting behaviors, we cannot assess this potential connection. Diagnoses of diabetes and hypertension in parents were obtained by self-report, which may underestimate their prevalence. There was also lack of HbA1c data in parents. Strengths of the current study include the large, well-characterized multiethnic cohort of subjects with youth-onset type 2 diabetes and detailed data collected from their parents.
In conclusion, this treatment trial of youth-onset type 2 diabetes found parental diabetes, hypertension, and obesity were associated with failure to maintain glycemic control and a greater prevalence of hypertension and obesity in the youth. Further research is needed to better understand the factors responsible for these findings.
Appendix
The following individuals and institutions constitute the TODAY Study Group (*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, and 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, and 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, and 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, and 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, and E. Venditti; Columbia University Medical Center: R. Goland*, D. Gallagher, P. Kringas, N. Leibel, D. Ng, M. Ovalles, and D. Seidman; Joslin Diabetes Center: L. Laffel*, A. Goebel-Fabbri, M. Hall, L. Higgins, J. Keady, M. Malloy, K. Milaszewski, and 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, and B. Steiner; Saint Louis University: S. Tollefsen*, S. Carnes, D. Dempsher, D. Flomo, T. Whelan, and 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, and 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, and 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, and 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, and A. Wauters; Washington University in St. Louis: N. White*, A. Arbeláez, D. Flomo, J. Jones, T. Jones, M. Sadler, M. Tanner, A. Timpson, and R. Welch; and Yale University: S. Caprio*, M. Grey, C. Guandalini, S. Lavietes, P. Rose, A. Syme, and W. Tamborlane.
Coordinating Center.
George Washington University Biostatistics Center: K. Hirst*, S. Edelstein, P. Feit, N. Grover, C. Long, and 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* and J. Harting; DEXA Reading Center (University of California at San Francisco): J. Shepherd*, B. Fan, L. Marquez, M. Sherman, and J. Wang; Diet Assessment Center (University of South Carolina): M. Nichols*, E. Mayer-Davis, and Y. Liu; Echocardiogram Reading Center (Johns Hopkins University): J. Lima*, S Gidding, J. Puccella, and E. Ricketts; Fundus Photography Reading Center (University of Wisconsin): R. Danis*, A. Domalpally, A. Goulding, S. Neill, and P. Vargo; and Lifestyle Program Core (Washington University): D. Wilfley*, D. Aldrich-Rasche, K. Franklin, C. Massmann, D. O’Brien, J. Patterson, T. Tibbs, and 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; and University of Florida: J. Silverstein.
Clinical trial reg. no. NCT00081328, clinicaltrials.gov.
A complete list of the TODAY Study Group can be found in the appendix.
Article Information
Acknowledgments. The TODAY Study Group gratefully acknowledges 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. The opinions expressed in this article are those of the authors and do not necessarily reflect the views of the respective Tribal and Indian Health Service Institution Review Boards or their members. Materials developed and used for the TODAY standard diabetes education program and the intensive lifestyle intervention program are available to the public at https://today.bsc.gwu.edu/.
Funding. This work was completed with funding from National Institute of Diabetes and Digestive and Kidney Diseases/National Institutes of Health grant numbers U01-DK-061212, U01-DK-061230, U01-DK-061239, U01-DK-061242, and U01-DK-061254, from the National Center for Research Resources General Clinical Research Centers Program grant numbers M01-RR-000036 (Washington University School of Medicine), M01-RR-000043-45 (Children’s Hospital Los Angeles), M01-RR-000069 (University of Colorado Denver), M01-RR-000084 (Children’s Hospital of Pittsburgh), M01-RR-001066 (Massachusetts General Hospital), M01-RR-000125 (Yale University), and M01-RR-014467 (University of Oklahoma Health Sciences Center), and from National Center for Research Resources Clinical and Translational Science Awards grant numbers UL1-RR-024134 (Children’s Hospital of Philadelphia), UL1-RR-024139 (Yale University), UL1-RR-024153 (Children’s Hospital of Pittsburgh), UL1-RR-024989 (Case Western Reserve University), UL1-RR-024992 (Washington University in St. Louis), UL1-RR-025758 (Massachusetts General Hospital), and UL1-RR-025780 (University of Colorado Denver).
Duality of Interest. The TODAY Study Group thanks the following companies for donations in support of the study’s efforts: Becton, Dickinson and Company; Bristol-Myers Squibb; Eli Lilly and Company; GlaxoSmithKline; LifeScan, Inc.; Pfizer; and Sanofi. R.S.W.: Participation in multicenter clinical trials sponsored by Medtronic, Sanofi, Novo Nordisk, Intarcia Therapeutics, and Eli Lilly and Company. S.W.: Grant funding from VeroScience, Takeda Pharmaceutical Company, Bristol-Myers Squibb, and Novo Nordisk; consultant for MannKind Corporation. K.M.: Consultant for Medtronic insulin pumps. No other potential conflicts of interest relevant to this article were reported.
Author Contributions. The writing group was composed of Ruth S. Weinstock (Chair), Paula M. Trief, Laure El ghormli, Robin Goland, Siripoom McKay, Kerry Milaszewski, Jeff Preske, Steven Willi, and Patrice M. Yasuda.
R.S.W., P.M.T., S.M., S.W., and P.M.Y. researched data and wrote or edited parts of the manuscript. L.E. performed the analysis, researched data, and reviewed and edited the manuscript. R.G. researched data and reviewed the manuscript. K.M. reviewed or edited parts of the manuscript. J.P. wrote or edited parts of the manuscript. L.E. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.