METHODS:
METHODS: Eligible participants were 326 children recruited from 6 socio-economically diverse areas in Melbourne, Australia, who took part in an earlier randomised trial for sleep problems at age 7 months. At 6 years, they participated in a follow-up assessment.
RESULTS: HbA1c collected by finger-prick. Exposures (collected simultaneously): proxy measures of child stress including: (i) child mental health; (ii) maternal mental health (depression, anxiety, stress), negative life events in the preceding year, life stresses and coping; and (iii) family socioeconomic status and financial stress.
METHODS: linear regressions, adjusted for original randomisation status and clustering.
RESULTS: Sixty percent (134/225) of children retained at 6 years provided HbA1c data, which ranged from 3.9%-5.8% (SD 0.3%). No child or family variable was associated with HbA1c. Of the maternal variables, only anxiety predicted HbA1c (adjusted difference per point increase: -0.01, 95% CI: -0.003 to 0.02, P = 0.01); this association was in the opposite direction to that hypothesised and clinically insignificant.
CONCLUSIONS: HbA1c was not associated with psychosocial stress in healthy 6-year-olds. This suggests that any link between HbA1c and psychosocial stress emerges after this age, and that HbA1c is unlikely to be a reliable biomarker for stress in early childhood or over the transition to school.