Psychological Medicine

Original Articles

Maternal separation in childhood and diurnal cortisol patterns in mid-life: findings from the Whitehall II study

M. Kumaria1 c1, J. Heada1, M. Bartleya1, S. Stansfelda2 and M. Kivimakia1

a1 Department of Epidemiology and Public Health, University College London, UK

a2 Centre for Psychiatry, Barts and the London School of Medicine and Dentistry, London, UK


Background Animal studies suggest that maternal separation is associated with alterations in the hypothalamic–pituitary–adrenal (HPA) axis through effects that occur in a critical period following birth. Evidence for an association of the diurnal cortisol rhythm with maternal separation in humans is equivocal.

Method We examined whether maternal separation in childhood is associated with diurnal cortisol pattern in 3712 middle-aged men and women. Two aspects of cortisol release were examined: the cortisol awakening response (CAR) and the diurnal slope in cortisol throughout the day.

Results Maternal separation in childhood was reported by 12% of participants. Those participants who reported maternal separation had a larger CAR and flatter slopes in cortisol levels compared to those who did not report maternal separation [adjusted mean CAR in those reporting no separation versus separation: 7.1, 95% confidence interval (CI) 6.7–7.5 v. 8.4, 95% CI 7.3–9.5, p = 0.02, corresponding to adjusted mean diurnal slope: −0.129 (95% CI −0.130 to −0.128) v. −0.126 (95% CI −0.128 to −0.124), p = 0.01]. In participants who reported maternal separation, the age of separation was not associated with either cortisol measure (p = 0.11). The association between maternal separation and slope in cortisol secretion was largely explained by smoking behaviour and marital status at the time of sample collection whereas that of the CAR was explained by childhood psychosocial, material factors and adult health behaviours.

Conclusions Our findings suggest that maternal separation in childhood is associated with alterations in the diurnal cortisol pattern in middle age. These associations are predominantly accounted for by adult circumstances and behaviours.

(Received April 14 2011)

(Revised May 21 2012)

(Accepted May 22 2012)

(Online publication July 11 2012)

Key words

  • Cortisol awakening response;
  • diurnal slope in cortisol;
  • lifecourse epidemiology;
  • maternal separation;
  • salivary cortisol;
  • stress


c1 Address for correspondence: Dr M. Kumari, Department of Epidemiology and Public Health, University College London, 1–19 Torrington Place, London WC1E 6BT, UK. (Email: