What Children are Eating and the Risk of Type 2 Diabetes Mellitus

aut.embargoNoen_NZ
aut.thirdpc.containsNoen_NZ
dc.contributor.advisorRush, Elaine
dc.contributor.advisorMearns, Gael
dc.contributor.authorJalili-Moghaddam, Shabnam
dc.date.accessioned2018-10-12T03:14:59Z
dc.date.available2018-10-12T03:14:59Z
dc.date.copyright2018
dc.date.issued2018
dc.date.updated2018-10-12T01:55:36Z
dc.description.abstractType 2 diabetes mellitus (T2DM) is an insidious, intergenerational disease. Although T2DM presents most often in adults, it is starting to present in childhood. In New Zealand (NZ), the overall prevalence of diagnosed T2DM in the adult population (≥15 years) is 5.8% with double the prevalence in the most deprived quintile of society (8.6% vs 4.4%). Prevalence differs by ethnicity: Māori (7.3%), Pacific (12.5%) and Asian (6.8%) compared with European (4.7%). Patterns of eating as well as rate of childhood growth and body size are three closely connected risk factors for T2DM that may be modifiable across the life course. The key to T2DM prevention is taking into account one’s whole diet and combinations of consumed foods, rather than one food item or nutrient. One in nine NZ children aged 2-14 years are obese (a major risk factor for T2DM). The rate of obesity among children living in the most deprived areas (20%) is higher than children living in the least deprived areas (4%). Diagnosis of diabetes is a common clinical challenge. The measurement of glycated haemoglobin A1c (HbA1c) in human blood is an indicator of average blood glucose over the previous 2-3 months. This measurement plus elevated serum uric acid (SUA) concentration - a risk factor for T2DM- can assist in the identification of the profile for risk for T2DM. The focus of this exploratory PhD was to identify groups of children with distinct eating patterns and investigate the relationships of eating patterns with risk of T2DM, using the blood biomarkers, HbA1c and SUA concentrations. In addition, information concerning conditions at birth (i.e., maternal age, maternal education, birth weight, and gender) and body size (i.e., current weight, height and waist), were examined as possible cofactors and covariates. Four connected investigations were undertaken to collect the information from high-risk children living in Auckland. To this aim, two longitudinal birth cohorts were studied: 1) the Pacific Islands Families (PIF) study at age 14 years, and 2) Metformin in Gestational Diabetes the Follow-up (MiGTOFU) study, at age 9 years. Firstly, a performance and utility investigation of the AfinionTM point of care test (POCT) for HbA1c was conducted with 94 girls and 96 boys aged 15 years in a nested subsample of the PIF cohort. In doing so, HbA1c was measured three times on two different occasions. The first occasion was collecting samples at schools using a capillary finger-prick sample. A year after, the second sample collection was completed, using the same model POCT and finger-prick. Simultaneously, HbA1c from a venous sample was analysed by boronate affinity chromatography at a certified laboratory. For the same day analysis, the mean difference in capillary and venous measures was 0.54 mmol.mol-1 (0.05%) (95% CI mean: 0.25, 0.83, p < 0.001) and the ± 1.96SD limits of agreement: 4.48, -3.40 mmol.mol-1. There was a moderate to strong correlation between the two POCT measures taken one year apart (r = 0.55, 95% CI [0.44, 0.65], p< 0.001) with a mean difference of 0.14 ± 2.18 (SD) mmol.mol-1. The within-day difference between the reference and the POCT was less than the precision of the POCT and was not biologically or clinically meaningful. The Afinion POCT TM AS100 test provided a valid and biologically reliable measure of HbA1C and had the potential to identify children at risk of elevated HbA1c. The second investigation, studied the association of eating patterns of 931 PIF children at 14 years. This included a cluster analysis of the data from a self-reported dietary habits questionnaire with the concentration of HbA1c measured with POCT. As a result of this, four eating patterns were derived for 740 children. In addition, the effect of body anthropometric measurements, body composition (i.e., measured by Bioimpedance Analysis) and some early life factors (i.e., maternal age at conception, maternal education, type of baby feeding, birth weight, and number of siblings) were evaluated. The mean of HbA1c concentration was not different between four derived eating patterns. Waist-to-height ratio and mothers’ age were positively (standardised β 0.108 p = 0.03) and negatively (standardised β -0.091 p = 0.012) related to HbA1c R2 = 0.020, but not the eating patterns. In the third investigation, the associations of reported consumption of added-sugar foods and risk of T2DM, was measured among a nested subsample from the PIF cohort with 204 Pacific children aged 15 years old. Findings suggest that boys drank more sugary drinks and ate fewer snacks and sweets than girls. After adjusting for gender ‘snacks and sweets’ group and ‘sugary drinks’ group were negatively and positively related to SUA respectively (R2 = 0.309), but no associations were found with HbA1c concentration. There was a positive and significant association between SUA and all the body size measurements. However, HbA1c concentration was positively and significantly related to weight, BMI, waist circumference, waist-to-height ratio, FM and FFM. For the last investigation data was sourced from the ‘Metformin in Gestational Diabetes the Follow-up’ study when the offspring (n = 99) of mothers treated for gestational diabetes mellitus were aged 7-9 years. Food consumption, HbA1body measurements including dual energy x-ray absorptiometry measurements and early life factors (e.g., maternal age at conception, maternal education, mother’s type of treatment, type of baby feeding, birth weight, and ethnicity) were included in the analysis. Five eating patterns were derived as a result of cluster analysis. After adjusting for waist-to-height ratio, there was almost a positive significant difference (p = 0.052) between ‘refined carbs’ and ‘not refined carbs’ eating patterns for HbA1c (1.4 mmol.L-1) (95% CI [-0.2, 2.8]). Eating patterns were associated with weight z scores, height z scores and visceral abdominal fat. Evidence presented in this thesis demonstrates that these high-risk groups of children are not meeting the NZ food guidelines and that their eating patterns are weakly associated with T2DM risk factors. Age and relative homogeneity of the children are the probable reasons for not finding associations of eating patterns among these apparently healthy children with a relatively high prevalence of overweight and obesity. Additionally, the methods and methodology used to collect dietary information, methodological difficulties in assessing the diets and the cross-sectional design of the analyses may have contributed. The findings of this thesis support initiatives to inform positive actions such as research, benchmarking, and quality improvement of the food environment. From this point of view, this research would benefit individuals, communities, and policy makers.en_NZ
dc.identifier.urihttps://hdl.handle.net/10292/11876
dc.language.isoenen_NZ
dc.publisherAuckland University of Technology
dc.rights.accessrightsOpenAccess
dc.subjectRisk for type 2 diabetes mellitusen_NZ
dc.subjectEating patternsen_NZ
dc.subjectChildrenen_NZ
dc.subjectHbA1cen_NZ
dc.subjectPoint of careen_NZ
dc.subjectFooden_NZ
dc.titleWhat Children are Eating and the Risk of Type 2 Diabetes Mellitusen_NZ
dc.typeThesisen_NZ
thesis.degree.grantorAuckland University of Technology
thesis.degree.levelDoctoral Theses
thesis.degree.nameDoctor of Philosophyen_NZ
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
JaliliMoghaddamS.pdf
Size:
10.16 MB
Format:
Adobe Portable Document Format
Description:
Whole thesis
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
889 B
Format:
Item-specific license agreed upon to submission
Description:
Collections