Longitudinal co-development of depressive symptoms and cardio-metabolic risk factors from childhood to young adulthood
Defina, S., Cecil, C.A.M., Felix, J.F., Walton, E., & Tiemeier, H. (in press) DOI
Check out this paper’s GitHub repository and interactive web application
Read the pre-print
Abstract \
Background: Depressive symptoms and cardio-metabolic risk factors often co-occur. However, our understanding of the potential mechanisms and temporal dynamics underlying their co-development remains elusive.
Methods: This population-based cohort study examined bidirectional longitudinal associations between depressive symptoms and cardio-metabolic risk factors from age 10 to 25 years, using prospective data from the ALSPAC Study. Participants with at least one (of six) follow-up measurement for each outcome were included in the analyses. We measured depressive symptoms through self- as well as parent-reports, and assessed several cardio-metabolic risk factors (including adiposity measures, lipid profiles and inflammation).
Results: Among our 7970 (47% male) participants, we found bidirectional, within-person associations between self-reported depressive symptoms and adiposity (i.e., fat/lean mass index, but not body mass index), across the study period. Adiposity was more stable over time (β [range] = 0.75 [0.54; 0.84]), compared to depressive symptoms (0.26 [0.12; 0.38]), and it had a stronger prospective (i.e., cross-lagged) association with future depressive symptoms (0.07 [0.03, 0.13]) compared to that between depressive symptom and future adiposity (0.04 [0.03, 0.06]). The magnitude of these associations reached its peak between 14 and 16 years. We did not find evidence of cross-lagged associations in either direction between depressive symptoms and waist circumference, insulin, triglycerides, LDL cholesterol or C-reactive protein.
Conclusions: These findings suggest a bidirectional relationship between depressive symptoms and cardio-metabolic risk factors, particularly adiposity (i.e., fat/lean mass). Adiposity showed a stronger prospective association with future depressive symptoms, than vice versa, however their relationship revealed more reciprocal than previously thought.