A nonlinear relationship of generalized and central obesity with diurnal cortisol secretion in the Whitehall II study

Abstract

Context: Evidence for an association of measures of generalized and central obesity with salivary cortisol secretion is equivocal.

Objective: The objective of this study was to assess the relationship between body mass index (BMI), waist circumference, and salivary cortisol.

Design: The design was a cross-sectional study of BMI, waist circumference, and salivary cortisol from phase 7 (2002-2004) of the Whitehall II study.

Setting: The occupational cohort was originally recruited in 1985-1988.

Participants: Participants included 2915 men and 1041 women aged 50-74 yr with complete information on height, weight and waist circumference, and cortisol secretion.

Outcome measures: Saliva samples were taken on waking, waking plus 0.5, 2.5, 8, and 12 h, and bedtime for the assessment of cortisol. The cortisol awakening response and slope in diurnal secretion were calculated.

Results: After adjustment for age, sex, social position, waking time, and time since waking of sample collection, increasing central and generalized obesity was associated with lower waking cortisol (P = 0.001). U-shaped associations were apparent between diurnal slope in salivary cortisol and both BMI and waist circumference (P < 0.0001 for quadratic term). For example, the shallowest (most adverse) slopes in salivary cortisol were associated with highest (>31 kg/m(2)) and lowest (<21 kg/m(2)) levels of BMI, and the steepest slopes were apparent for those with BMI of 26 kg/m(2), independently of the 12 covariates examined. No associations were apparent for the cortisol awakening response (P > 0.05).

Conclusion: The associations of measures of generalized and central obesity with diurnal slope in salivary cortisol are not linear in older adults. These nonlinear associations may explain previously described mixed findings.

Figure 1

Diurnal slope in cortisol secretion (from a 2-class multilevel model with cortisol sampled at waking, at waking plus 2.5, 8, and 12 h, and at bedtime as level 1 variables and individual as a level 2 variable) by BMI. The figure represents mean slope in cortisol secretion adjusted for age, sex, last known employment grade, waking time, and time since waking (squares and continuous line), plus additional adjustments for alcohol consumption, smoking, poor self-rated health, short sleep, fatigue, depressive symptoms, financial insecurity, and stress on day of sampling (dots and dashed line). P < 0.0001 for quadratic term.

Figure 2

Diurnal slope in cortisol secretion (from a 2-class multilevel model with cortisol sampled at waking, at waking plus 2.5, 8, and 12 h, and at bedtime as level 1 variables and individual as a level 2 variable) by obesity status. Figure represents mean cortisol secretion adjusted for age, last known employment grade, waking time, and time since waking estimated from Table 3. P < 0.0001 for quadratic term.