Elsevier

Psychoneuroendocrinology

Volume 35, Issue 9, October 2010, Pages 1404-1409
Psychoneuroendocrinology

Higher cortisol content in hair among long-term unemployed individuals compared to controls

https://doi.org/10.1016/j.psyneuen.2010.04.006Get rights and content

Summary

Unemployment and financial strain are chronic stressors that have been shown to be associated with an increase in mean salivary and serum cortisol levels. Hair analysis for cortisol content is a new promising tool by which hair segmental analysis may provide a retrospective calendar of cumulative cortisol exposure over time rather than momentary assessments.

Participants of this study were 31 unemployed and 28 employed individuals (46 women). Hair segmental analysis was conducted using 3-cm long segments starting with the scalp-near segment. Due to differing hair length, 52 individuals had values for the second segment and n = 33 individuals had values for the third segment. Univariate analysis of variance indicated that unemployed individuals had higher cortisol content in the first (p < 0.05, eta2 = 0.071) and second (p < 0.05, eta2 = 0.085) hair segment (a total of 6 cm long hair representing the preceding 6 months of collection). Consistent with other data from our laboratory, there was a wash-out effect for the third segment (p < 0.05 for segment 3 vs. segment 1 and 2). Unemployed individuals indicated increased levels of perceived stress and impairments in subjective well-being compared to employed individuals. These subjective measures of perceived stress and well-being were unrelated to cortisol content in hair.

We conclude that hair analysis for cortisol content may be a valid method to detect differences in cumulative cortisol exposure between chronically stressed individuals and healthy controls. Due to a wash-out effect, retrospective ascertainment of cortisol exposure may be limited to the preceding 6 months of specimen collection.

Introduction

Unemployment and financial strain are chronic stressors that have been shown to be associated with poor well-being and poor psychological and physical health (Hammarstrom and Janlert, 2002, McKee-Ryan et al., 2005), as well as with alterations of hypothalamus–pituitary–adrenal (HPA) axis functioning, one of the major endocrine stress axes. More specifically, studies have found an increase in mean serum cortisol levels (Arnetz et al., 1991, Grossi et al., 1999, Maier et al., 2006), as well as alterations of the diurnal cortisol profile with some reporting increased evening cortisol levels (Grossi et al., 2001) and others documenting increased morning cortisol levels in the presence of unchanged overall cortisol excretion rates (Ockenfels et al., 1995).

These inconsistencies in results are not uncommon in the literature on chronic stress and cortisol release, a circumstance that may stem from a methodological void consisting of the assessment of cortisol over longer periods of time such as weeks or months. That is, cortisol levels in blood or saliva represent “point” assessments, which are highly variable due to influencing factors such as circadian rhythm, food intake, and any stress that may have occurred shortly before sampling. While urine analysis for cortisol is suited for gaining a summary index of HPA activity covering a period of up to 24 h, the influence of chronic stress on cortisol excretion rates and possible associated health-outcomes needs to be assessed for time periods of up to several months. In sum, to capture the impact of a chronic stressor on cortisol excretion rates, a feasible and reliable approach to measuring cortisol concentrations over extended periods of time is needed.

Hair analysis appears to be a promising tool which has been used successfully in courts of Justice and doping control for well over a decade. Due to its wide window of detection of several months, noninvasive collection, easy storage, and inability to decompose like other body fluids or tissues, hair analysis has now also been introduced to the field of biopsychology. Hair analysis for cortisol was first undertaken by Raul and colleagues who demonstrated the detection of physiological concentrations of cortisol and cortisone in human hair (Raul et al., 2004). Then, Davenport et al. (2006) demonstrated for the first time in rhesus monkeys that the imposition of a major stressor (relocation) lead to increased hair cortisol levels 14 weeks post-move compared to pre-move levels and a recovery back to pre-move levels at 1 year. Hair cortisol concentrations correlated highly with eight (averaged) post-move salivary cortisol samples. More recently, Yamada et al. (2007) could replicate that hair cortisol concentrations in fact appear to be sensitive to stress-induced changes in HPA-axis activity. They demonstrated that in preterm and term infants in the neonatal intensive care unit, days on the ventilator, a circumstance the authors propose to be an index for chronic neonatal stress, influenced hair cortisol levels. Further support for hair cortisol levels as a reflection of stress-induced changes of the HPA-axis stems from Kalra and colleagues who reported associations of hair cortisol levels with measures of perceived stress in 25 healthy pregnant women (Kalra et al., 2007).

We recently conducted a proof-of-concept study in young mothers to use pregnancy as a model for increased cortisol excretion over a prolonged period of time in a healthy organism (Kirschbaum et al., 2008). Our aim was to determine if endogenous cortisol levels in hair remain stable over time and move along the hair shaft, thus allowing the usage of segmental hair analysis to create a retrospective calendar of endogenous cortisol secretion. We found elevated cortisol content in the hair segment of young mothers representing the third trimester of pregnancy, a period of time where cortisol levels increase by threefold (Erickson et al., 2001, Sandman et al., 2006), compared to controls. Furthermore, we found a decline in cortisol concentration from the scalp-near to the more distal hair segments. In sum, the literature to date on hair cortisol analysis indicates that human hair in fact is a medium by which increased cortisol production over an extended period of time can easily be captured. So far, a time period of 6 months seems to allow for valid data as preliminary data indicates a decline of cortisol levels in distal hair segments.

Based on these findings, the current study aimed at detecting whether hair cortisol levels differed between unemployed and employed individuals. That is, is the chronic stress of being unemployed reflected in increased cortisol incorporation into the hair over extended periods of time.

Section snippets

Study participants

A total of 59 individuals were recruited for the present study (n = 46 women) of whom 31 were unemployed (36.74 ± 11.04) and 28 were employed (32.55 ± 9.28). Inevitably, the unemployed group had a much lower household income than the employed group with 80.6% having a monthly sum of less than 1000 Euros while this was only the case for 5.5% in the employed group. More than half of the employed group (55.6%) received more than 2000 Euros monthly, a sum that none of the unemployed households achieved.

Results

Overall, the sample consisted of more women (78%) than men. Most participants held a graduate or college degree (26.5% and 51.0%, respectively). Almost half of the sample smoked (40.8%) and was on a regular medication regimen (42.9%). 36.7% were overweight or obese (38.9% and 36% for employed and unemployed individuals, respectively). As seen in Table 1, group differences were found for the following variables: gender, education, and hair treatment. To evaluate for possible confounders, these

Discussion

We found elevated cortisol levels in hair among long-term unemployed individuals compared to healthy controls. This group difference was observable for hair strands of a total of 6 cm length, which based on an average hair growth rate of 1 cm per month (Pragst and Balikova, 2006) represents a summary index of cortisol excretion for the preceding 6 months of sample collection. Consistent with prior data from our laboratory (Kirschbaum et al., 2008) we found a wash-out effect which was apparent in

Role of funding source

This study was in part supported by a research grant from the Deutsche Forschungsgemeinschaft to LD and CK (De 4-1).

Conflict of interest

We declare that none of the authors has any financial or other relationships that might lead to a conflict of interests in relation to this study or the content of this manuscript.

Acknowledgements

The authors would like to thank the study participants for their contribution of time and effort to the research.

References (30)

  • B.B. Arnetz et al.

    Neuroendocrine and immunologic effects of unemployment and job insecurity

    Psychother. Psychosom.

    (1991)
  • S. Cohen et al.

    The stability of and intercorrelations among cardiovascular, immune, endocrine, and psychological reactivity

    Ann. Behav. Med.

    (2000)
  • S. Cohen et al.

    A global measure of perceived stress

    J. Health Soc. Behav.

    (1983)
  • J. Denollet

    DS14: standard assessment of negative affectivity, social inhibition, and Type D personality

    Psychosom. Med.

    (2005)
  • K. Erickson et al.

    Preterm birth: associated neuroendocrine, medical, and behavioral risk factors

    J. Clin. Endocrinol. Metab.

    (2001)
  • Cited by (226)

    View all citing articles on Scopus
    View full text