Elsevier

Psychoneuroendocrinology

Volume 34, Issue 10, November 2009, Pages 1423-1436
Psychoneuroendocrinology

Assessing salivary cortisol in large-scale, epidemiological research

https://doi.org/10.1016/j.psyneuen.2009.06.011Get rights and content

Summary

Salivary cortisol measures are increasingly being incorporated into large-scale, population-based, or epidemiological research, in which participants are selected to be representative of particular communities or populations of interest, and sample sizes are in the order of hundreds to tens of thousands of participants. These approaches to studying salivary cortisol provide important advantages but pose a set of challenges. The representative nature of sampling, and large samples sizes associated with population-based research offer high generalizability and power, and the ability to examine cortisol functioning in relation to: (a) a wide range of social environments; (b) a diverse array individuals and groups; and (c) a broad set of pre-disease and disease outcomes. The greater importance of high response rates (to maintain generalizability) and higher costs associated with this type of large-scale research, however, requires special adaptations of existing ambulatory cortisol protocols. These include: using the most efficient sample collection protocol possible that still adequately address the specific cortisol-related questions at hand, and ensuring the highest possible response and compliance rates among those individuals invited to participate. Examples of choices made, response rates obtained, and examples of results obtained from existing epidemiological cortisol studies are offered, as are suggestions for the modeling and interpretation of salivary cortisol data obtained in large-scale epidemiological research.

Introduction

The addition of biological measures (“biomarkers”) to large-scale social science and epidemiological studies has recently been advocated by a number of funding bodies (Finch et al., 2001). In this interdisciplinary approach, extensive information on social and behavioural processes, health behaviours and self-reported health in existing large-scale, representative samples, is complemented by the addition of objective biological markers of physiological processes and pre-disease states.

Measures of the hypothalamic pituitary adrenal axis (HPA), and in particular levels of salivary cortisol are obvious candidates for inclusion in these studies, given the important role that the HPA axis plays in ‘transducing’ subjective social–environmental experience into physiological changes relevant to health. While salivary cortisol has been successfully gathered in a number of large-scale, population-based studies, the design, collection and interpretation of salivary cortisol data in naturalistic settings presents numerous challenges. Here we discuss these challenges, and related solutions, and review some existing results that have emerged from this new wave of research.

Section snippets

Background

The HPA axis serves as an important pathway by which social and psychological factors influence biology and health. Stressful stimuli serve to activate HPA function to cause an increase in peripheral cortisol. Cortisol can effect physiological changes that encompass most of the main organ systems, helping to provide the energetic resources needed to face the stressor at hand, and also helping to modulate and contain other components of the physiological stress response (Sapolsky et al., 2000).

Epidemiological research: strengths and challenges

Epidemiological research is characterized by: (1) representative sampling, in which the sample is carefully selected and retained to ensure that participants precisely reflect the characteristics of the larger population of interest; (2) large sample sizes (in the order of hundreds to tens of thousands), typically of a broader socioeconomic and racial/ethnic range than in samples of convenience; and (3) extensive measurement of and statistical control for potentially confounding variables.

As a

Methodological priorities in epidemiological research: design implications

These defining features of epidemiological research, and their relative advantages, also require a shift in several priorities regarding the design of cortisol data collection protocols for large-scale research.

Existing epidemiological cortisol research

While many salivary cortisol protocols have been implemented in convenience samples, we focus here on studies in which participants were purposefully sampled to accurately reflect the characteristics of a particular population of interest, such as samples chosen to be representative of a particular occupational status or age group (Ranjit et al., 2005, Rosmalen et al., 2005, Cohen et al., 2006, Badrick et al., 2007). The sample sizes, basic demographic characteristics, ages, data collection

Data analytic approaches and controversies

Approximately 1% of cortisol measures are found to be 3 standard deviations above the mean cortisol value (Whitehall II and Rotterdam Studies). It is unclear what these high cortisol values represent. In large-scale studies, the absolute numbers with these very high values can be substantial; it is not yet resolved whether it is best to remove these values altogether from analyses, or windsorize them in order to reduce their influence on the analysis. Even with such outliers removed, cortisol

Some findings from epidemiological cortisol research

The investment that researchers have made thus far in gathering salivary cortisol in population-based research studies is beginning to pay off. Studies are now appearing in the published literature, examining problems and processes that would have been difficult to capture with cortisol research utilizing smaller sample sizes. One area of research that these studies are uniquely suited to examine is association between race/ethnicity and/or socioeconomic position and cortisol levels. In

Research guiding design choices

The addition of salivary cortisol protocols to large-scale studies heralds a new and exciting period for HPA axis research, that should lead to a better understanding of the role of daytime cortisol secretion, its behavioural and biological correlates and how it relates to the development of disease. More research is however needed to quantify the extent of unreliability introduced by minimal protocols, and the extent to which that unreliability compromises investigators’ abilities to detect

Role of funding source

Whitehall II study has been supported by grants from the Medical Research Council; Economic and Social Research Council; British Heart Foundation; Health and Safety Executive; Department of Health; National Heart Lung and Blood Institute (HL36310), US, NIH: National Institute on Aging (AG13196), US, NIH; Agency for Health Care Policy Research (HS06516); and the John D and Catherine T MacArthur Foundation Research Networks on Successful Midlife Development and Socioeconomic Status and Health.

Conflict of interest

The authors have no conflicts of interests to declare.

Acknowledgements

We would like to thank Martin Shipley for providing the power calculations. We would like to thank all the principal investigators that provided information on the studies in which salivary samples have been incorporated.

References (53)

  • B.M. Kudielka et al.

    Awakening cortisol responses are influenced by health status and awakening time but not by menstrual cycle phase

    Psychoneuroendocrinology

    (2003)
  • J.C. Pruessner et al.

    Two formulas for computation of the area under the curve represent measures of total hormone concentration versus time-dependent change

    Psychoneuroendocrinology

    (2003)
  • J.C. Pruessner et al.

    Free cortisol levels after awakening: a reliable biological marker for the assessment of adrenocortical activity

    Life Sci.

    (1997)
  • N. Ranjit et al.

    Modeling cortisol rhythms in a population-based study

    Psychoneuroendocrinology

    (2005)
  • J.G.M. Rosmalen et al.

    Determinants of salivary cortisol levels in 10–12 year old children: a population-based study of individual differences

    Psychoneuroendocrinology

    (2005)
  • E.A. Shirtcliff et al.

    Use of salivary biomarkers in biobehavioral research: cotton-based sample collection methods can interfere with salivary immunoassay results

    Psychoneuroendocrinology

    (2001)
  • A.A. Stone et al.

    Individual differences in the diurnal cycle of salivary free cortisol: a replication of flattened cycles for some individuals

    Psychoneuroendocrinology

    (2001)
  • I. Wilhelm et al.

    Is the cortisol awakening rise a response to awakening?

    Psychoneuroendocrinology

    (2007)
  • C.E. Wright et al.

    Subjective socioeconomic position, gender and cortisol responses to waking in an elderly population

    Psychoneuroendocrinology

    (2005)
  • L.S. Adair et al.

    Maternal energy stores and diet composition during pregnancy program adolescent blood pressure

    Circulation

    (2001)
  • E.K. Adam

    Momentary emotion and cortisol levels in the everyday lives of working parents

  • E.K. Adam et al.

    Day-to-day dynamics of experience-cortisol associations in a population-based sample of older adults

    Proc. Natl. Acad. Sci. U.S.A.

    (2006)
  • E. Badrick et al.

    Alcohol consumption and cortisol secretion in an ageing cohort

    J. Clin. Endocrinol. Metab.

    (2008)
  • E. Badrick et al.

    The relationship between smoking status and cortisol secretion

    J. Clin. Endocrinol. Metab.

    (2007)
  • R.H. Belmaker et al.

    Major depressive disorder

    N. Engl. J. Med.

    (2008)
  • D.L. Bliwise

    Sleep in normal aging and dementia

    Sleep

    (1993)
  • Cited by (0)

    1

    The authors made equal contributions to this paper.

    View full text