Article Text

G485(P) Higher salivary cortisol levels at 2 years of age are associated with impaired language performance and attentional problems, but not associated with prematurity
  1. EM Hurrion1,2,
  2. M Harris2,3,
  3. RM Greer2,
  4. PH Grey1,2
  1. Department of Newborn Services, Mater Mothers Hospital, Brisbane, Australia
  2. Program for Optimising Outcomes for Mothers and Babies at Risk, Mater Research Institute – The University of Queensland, Brisbane, Australia
  3. Department of Endocrinology, Lady Cilento Children’s Hospital, Brisbane, Australia


Aims To investigate cortisol levels before and after Bayley developmental assessment at 2 years of age for preterm compared with term born children.

To investigate any association between cortisol level and cognitive and language performance on Bayley III Scales of Infant and Toddler Development (Bayley-III), and attentional problems on Child Behaviour Checklist (CBCL).

Methods Preterm infants (24–30weeks), n=82, and term controls (37–41weeks), n=61, were recruited from a longitudinal study. At 2 years corrected age participants underwent assessment using Bayley- III, and parents completed CBCL. Immediately prior to, and on completion of the assessment, parents collected saliva samples using dental swabs. Saliva was extracted and cortisol concentration measured using a commercially-available enzyme immunoassay kit.

Results There was no statistically significant difference between preterms and terms for either pre- or post-assessment cortisol values. There was also no statistically significant change in cortisol values before and after assessment. There was no association between cortisol values and cognitive performance on the Bayley-III. There was a statistically significant negative association between pre and post-assessment cortisol values and language performance on the Bayley-III. Univariate analysis revealed many confounding factors, but on logistic regression analysis the independent factors were pre-assessment cortisol and maternal education level which together explained 27% of the variance in language scores. There was a statistically significant positive association between pre-assessment cortisol and attentional problems on CBCL. Univariate analysis revealed several confounding factors, but on logistic regression analysis, the independent factors were pre-assessment cortisol (p=0.029) and social support (p=0.001).

Conclusion Daytime cortisol levels do not differ between preterm and term born children. At 2 years, a standardised developmental assessment is not an effective stressor to evoke a cortisol response. Higher daytime cortisol levels are associated with attentional problems, modified only by maternal social support. Higher daytime cortisol levels immediately prior to Bayley-III assessment are also associated with poorer language performance, modified only by maternal education level. High cortisol is known to impair declarative memory, which may explain this finding as early language is predominantly declarative. It may also be explained by the impact of poorer attention.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.