A longitudinal study of fat mass accrual from adolescence through to emerging adulthood and its impact on cardiometabolic risk later in life
Overweight and obesity (OWO), specifically abdominal obesity, are linked with cardiovascular and metabolic disease risk (CMR) at every stage of life. The prevalence of OWO nearly doubles in Canadians between childhood/adolescence and adulthood, suggesting that normal weight (NW) status is not stable through the life course and that there is a period in adulthood when fat mass (FM) increases. Emerging adulthood (EA; 18-25 years) has been identified as a potential critical period when FM accrues, the degree to which is potentially influenced by childhood and adolescence FM accrual. EA is also a period which favors trunk fat depots and when CMR likely commences or increases from child and adolescent levels. However, there is a paucity of longitudinal data describing the patterns and predictors of FM accrual during EA. Furthermore, there is also a lack of information showing how EA fat mass trajectories relate to adult cardiometabolic health. Thus, the primary purpose of this thesis is to describe patterns and predictors of total body fat (TBF) and trunk fat (TrF) mass accrual and OWO status in EA. The second purpose is to identify if trajectories of FM accrual in EA influence later adult CMR. The thesis will also explore sex differences. In study 1, 126 participants (59 male) were drawn from the Pediatric Bone Mineral Accrual Study (PBMAS) (1991-2011). Participants of the PBMAS were aged 8 to 15 years at the initial measurement. Serial measures of participants included chronological age (CA), biological age (BA - years from peak height velocity (PHV)), body mass index (BMI), and percent total body fat (%TBF). Study 1 is divided into two papers - 1a and 1b. The results from paper 1a based on this study indicated that fat mass increased from PHV into EA. At PHV, 9% of males and 14% of females were OWO by BMI, rising to 65% and 32%, respectively, 15 years after PHV (approximately 27 years in females, and 29 years in males). The prevalence of OWO by %TBF, increased from 29% to 45% in males, and from 33% to 59% in females over the same period. Differences in values of %TBF and BMI at PHV between those identified as NW had disappeared by 19-22 years (p>0.05) (i.e. fat mass between NW and OWO youth became more similar with age). OWO status at PHV did not predict OWO status during EA (p<0.05). These results indicate that EA appears to be a major period of transition from NW to OWO status. In addition, sex and FM metrics showed differences in ages when NW individuals became OWO. Study 1 also attempted to address the potential discrepancy in OWO identification and age at onset of OWO by different metrics in paper 1b, using the same PBMAS cohort. Longitudinal measures, including anthropometrics and dual x-ray absorptiometry from 1991-2017 were used to create hierarchical random effects models. Coefficients from the models were then used to develop growth curves, and these were compared to known cut-points. The age at onset of OWO was considered that age at which the predicted line crossed metric specific cut-offs. Age at onset of OWO in males was identified as 23.5 years by BMI and 21.5 years for % fat mass (%FM). Waist circumference (WC) cut-offs for OWO classification were not reached by 39 years. In females, onset was at 22.5 years by BMI, 15 years by %FM and 33.5 years by WC. Cut-points for BMI failed to identify 21.4% of OWO males and 56% of OWO females identified by %FM. The discrepancy in age at OWO between measures suggests that the most conservative indicator of age at onset is sex specific. BMI identifies OWO in males sooner than %FM. The opposite is true in females; however, BMI likely misses “over fatness” - more in females. WC may not be as appropriate for indicating risk in young adults and youth as in older adults. Using the same participants identified in study 1, study 2 created longitudinal models of fat accrual during EA and beyond (18 to 30 years of age) and identified concurrent and childhood/adolescent predictors of FM accrual, including measures of physical activity (PA) and energy intake (EI). It was found that childhood and adolescent TBF and TrF (0.30 ± 0.05, p<0.05) predicted EA accrual in both sexes and that concurrent PA (-0.06 ± 0.02, p<0.05) was significant in males only. These results underscored the importance of maintaining lower amounts of TBF and TrF mass during childhood and adolescence, and maintaining high level of PA in EA in order to mitigate TBF and TrF mass accrual and reduce the risk of transitioning from NW to OWO during EA. In study 3 (1991-2017) participants of the PBMAS, now aged 32 to 40 years of age, were invited back for reassessment. Blood analysis was used to create a Continuous Cardiometabolic Risk (conCMR) score for each participant. Multi-level models of TrF and TBF accrual were created looking at the same predictors as study 2, with the addition of cardiometabolic risk (CMR) group. Childhood TBF and TrF z-scores were again found to be the most significant predictor of TBF and TrF accrual, this time from 18-39 years. PA was also significant. CMR group did not influence the trajectory of TBF or TrF accrual, potentially due to the homogeneity of the group and the small sample size. In conclusion it was found that FM continues to increase steadily from late adolescence through EA leading to a marked increase in the prevalence of OWO in young adulthood. Greater trajectories during EA are related to higher levels of FM accrual in childhood and adolescence, and higher scores on individual CMR factors in later adulthood. The results suggest that maintaining high levels of PA throughout the life span is beneficial to adult health directly and through its mitigating effect on FM accrual in EA, and indirectly by limiting FM accumulation in childhood and adolescence.
longitudinal, fat mass, cardiometabolic risk, emerging adulthood, physical activity
Doctor of Philosophy (Ph.D.)