Psychosocial correlates of cortisol levels in fingernails among middle-aged workers

Abstract

It was recently suggested that cortisol levels in fingernails reflect cumulative hormone exposure over a relatively long period. This exploratory study cross-sectionally investigated the relationships between fingernail cortisol level and psychosocial stress in a sample of middle-aged workers (94 men and 29 women). The participants were asked to grow their fingernails for ∼2 weeks and then provide fingernail samples from every digit by using nail clippers. Further, they completed questionnaires for assessment of exposure to psychosocial stress in the past (stressful life events in the workplace in the previous year; e.g. change to a different line of work) and in the present (job stress and perceived stress). Results of a regression analysis adjusting for the effects of demographic variables showed that experience of stressful life events, but not job stress and perceived stress, was associated with elevated fingernail cortisol level. These findings indicate the potential of fingernail samples to retrospectively reflect individual differences in cortisol levels related to past psychosocial stress.

Keywords:

• Cortisol
• fingernail
• job stress
• stressful life events
• hair cortisol
• perceived stress

Introduction

Psychosocial stress triggers a number of physiological changes such as activation of the sympathetic nervous system and hypothalamic–pituitary–adrenal (HPA) axis. One consequence of this activation is the release of cortisol, an adrenal cortex hormone secreted in response to acute stress (Dickerson & Kemeny, 2004). Traditionally, cortisol has been measured in blood and saliva samples. The analysis of saliva and blood hormone level provides an index for a short period. Several recent studies have shown that endogenous hormones such as cortisol can reliably be measured in scalp hair (Russell, Koren, Rieder, & Van Uum, 2012). Scalp hair grows at an average rate of 1.0 cm/month; therefore, 1.0 cm of scalp hair may reflect hormone levels secreted in 1 month. While analysis of saliva and blood hormone levels provides an index for a short period, hair sample analysis provides a retrospective index of cumulative hormone exposure over a longer period, which could have advantages in the investigation of cortisol levels and chronic stress.

It was recently suggested that cortisol levels in fingernails also reflect cumulative hormone exposure over a relatively long period (de Berker, André, & Baran, 2007; Warnock et al., 2010). Neutrally charged endogenous hormones passively diffuse from capillaries into the nail matrix, and are incorporated into keratin during nail formation (de Berker et al., 2007). In this process, free cortisol may be incorporated into keratin through the same mechanism by which cortisol is incorporated into hair (Russell et al., 2012). Fingernails grow at an average rate of 1.0 mm/10 days (Gupta et al., 2005); therefore, 1.0 mm of fingernail may retrospectively reflect hormone levels over 10 days. In addition, previous work that traced fluoride in nails found that 3–4 months were required for nails to fully extend from the nail matrix (Buzalaf, Pessan, & Alves, 2006). Therefore, fingernail samples may reflect cortisol levels in the 3–4 months prior to clipping. In a recent study (Izawa et al., 2015), it was reported that cortisol levels in fingernails were associated with cortisol levels in the past that were assessed by hair and saliva samples. However, no previous studies clearly demonstrated the relationship between fingernail cortisol and psychosocial stress as well demographic variables, and minimal information is available on the fundamental aspects of fingernail cortisol.

This exploratory study cross-sectionally investigated the relationships between fingernail cortisol level and psychosocial stress in the past (stressful life events in the workplace in the previous year) and in the present (job stress and perceived stress) in a sample of middle-aged workers. This population was chosen because psychosocial stress in the workplace is associated with stress-related disease and could cause numerous biological changes (e.g. Eller et al., 2009). We expected that fingernail cortisol level would be positively associated with psychosocial stress, especially for stressful life events in the previous year, because fingernail samples reflect cumulative hormonal exposure in the past. Furthermore, we additionally investigated the effects of demographic variables (age, gender, body mass index, smoking status, and manicure use) on fingernail cortisol level.

Methods

Participants

White-collar workers were recruited from hospitals and research institutes in Kanagawa Prefecture in Japan. The sample consisted of hospital personnel and research institute staff who were mainly engaged in desk work. The sample initially comprised 140 workers; however, 17 were excluded for various reasons: six for recent steroid injection or medication use, two for antidepressant use, one for history of an adrenal gland tumor, one for recently giving birth to a child, three for missing questionnaire data, and four for containing statistical outliers regarding fingernail cortisol (see results section for additional details). Therefore, the final sample consisted of 123 workers (94 men; 29 women). Of the 123 participants, none reported adrenal gland disease such as Cushing’s syndrome or Addison’s disease, and none were taking steroid medications or antidepressants. Written informed consent was obtained from participants, and the ethical committees of the National Institute of Occupational Safety and Health approved the study.

Questionnaires

Stressful life events in the workplace were assessed using the following six items derived from the Social Readjustment Rating Scale (Holmes & Rahe, 1967): “collapse of the company”, “disemployment”, “job change”, “change in job responsibility”, “change to different line of work”, and “merger and reorganization”. Participants were asked whether they had experienced any of these six life events during the previous year using a yes/no format.

Job stress was assessed on a six-item measure derived from the Brief Job Stress Questionnaire (Shimomitsu, Yokoyama, Ono, Maruta, & Tanigawa, 1998), which was based on the job stress model of Karasek and Theorell (1990). Three items pertained to job demands (e.g. “have an extremely large amount of work to do”) and three items concerned perceived job control (e.g. “can choose how and in what order to do work”). Each item was rated on a four-point Likert scale ranging from “disagree” to “agree”, and the Cronbach’s alpha coefficient of job demands and control items was .77 and .75, respectively. The job strain index was calculated by dividing job demands by job control. A job strain of six items indicated a balance between demands and control, with higher scores reflecting high demand coupled with low control.

Perceived stress was assessed using the Japanese version of the Perceived Stress Scale (Iwahashi, Tanaka, Fukudo, & Hongo, 2002; PSS). The PSS is a 14-item questionnaire with five response options that assess the perceived degree of stressfulness of situations over the past month (e.g. “how often have you felt that things were going your way?”). The internal reliability (alpha coefficients) range from .82 to .89.

Procedure

For collection of fingernail samples, participants were asked to grow their fingernails for ∼2 weeks and provide samples from every digit by using nail clippers with a catcher to minimize tissue loss. Participants with manicured hands were asked to remove any nail polish before clipping. Fingernail samples were transferred into a Ziploc bag and frozen at −30 °C prior to the assay.

Participants also completed a questionnaire that assessed smoking status, height, weight, and frequency of nail polish use at the time of sample collection. Collection of fingernail samples and the questionnaire survey were conducted once in January 2012 or in February 2012.

Nail hormone extraction and enzyme immunoassay

Our nail hormone extraction method was identical to that used in a previous study (Izawa et al., 2015). Each sample was transferred into a 15-mL Falcon tube. Then, 5 mL of isopropanol was added, and the tube was vortexed twice for 60 s. This washing procedure was repeated two times. Samples were air-dried overnight. Dried samples were transferred to a 2.0-mL polypropylene micro-tube with a zirconia ball (diameter, 5 mm) and ground for 40 min using a mixer mill (Retsch MM300, Haan, Germany) set at 30 Hz. Fifteen milligrams of fingernail powder was transferred to another 2.0-mL micro-tube, followed by the addition of 1.5 mL of pure methanol. The micro-tube was slowly rotated for 24 h at room temperature to allow for steroid extraction. Following this, the micro-tube was centrifuged at 10,000 rpm for 2 min. One milliliter of the clear supernatant was transferred into a 50-mL Falcon tube, and subjected to evaporation for ∼20 min at 60 °C until completely dry.

Cortisol level was determined by an enzyme immunoassay method using the EIA Kit (Salimetrics LLC, State College, PA). The evaporated samples were re-suspended in 100 μl of the assay diluent included in the EIA Kit, and the levels of cortisol in the diluent were analyzed according to the manufacturer’s instructions. The inter-assay and intra-assay variations were <6.41 and 3.65%, respectively. The findings are presented as pg cortisol/mg fingernail (pg/mg). The lowest detectable level of cortisol is 0.56 pg/mg. We previously confirmed that the dilution curve of cortisol levels measured in serially diluted fingernail extracts significantly paralleled the cortisol standard curve from the assay kit (Izawa et al., 2015).

Statistical analyses

Cortisol levels in fingernails were logarithmically transformed (base 10) because their distribution in this sample was skewed. Further, the Smirnov–Grubbs tests for transformed values were performed to find statistical outliers. Correlational analyzes, independent t-tests, and a multiple linear regression analysis were subsequently conducted to evaluate the effect of stressful life events in the workplace, job strain, and perceived stress on cortisol level in fingernails. Age, gender (male/female), body mass index (BMI), and smoking status (yes/no) were included in the regression analysis. The distribution of the number of experienced stressful life events was skewed; we coded this variable as “0” (did not experience any stressful events) and “1” (experienced one or more stressful events).

Furthermore, the effect of nail polish could be evaluated in 19 women, and nail hormone levels were compared between women who used nail polish (N = 11) and those who did not (N = 8) using an independent t-test. All statistical calculations were performed using PASW Statistics 18 for Windows (SPSS Inc., Tokyo, Japan).

Results

The Smirnov–Grubbs test identified four statistical outliers ranging from 46.0 to 142.4 pg/mg, which were excluded from analyzes. The means and standard deviation (SD) of fingernail cortisol, as well as demographic and psychosocial variables, are presented in Table 1. Participants who experienced one or more stressful life events (i.e. the 1+ group) experienced an average of 1.4 (±0.5) stressful life events. Fingernail cortisol levels between women who used nail polish (mean ± SD: 4.3 ± 4.1 pg/mg) and those who did not (3.4 ± 1.0 pg/mg) were not significantly different.

Table 1. Demographics of the participants.

Demographic variable	N or mean ± SD
Male/female	94/29
Age (years), mean ± SD	43.4 ± 10.2
BMI (kg/m^2), mean ± SD	23.2 ± 3.5
Smoking status (no/yes)	91/32
Stressful life events in the workplace (0/1+)	75/48
Job strain (demand/control), mean ± SD	1.1 ± 0.5
Perceived stress, mean ± SD	22.5 ± 6.4
Cortisol level in fingernails (pg/mg), mean ± SD	4.6 ± 2.4

The means and SD of fingernail cortisol in the participants without and with stressful life events in the workplace were 4.2 ± 2.1 and 5.2 ± 2.7 pg/mg, respectively. Cortisol levels significantly differed between the groups [t(121) = 2.62, p = .010]. Job strain and perceived stress did not correlate significantly with fingernail cortisol levels. The results of a multiple linear regression analysis are also shown in Table 2. Smoking status and experience of stressful life events in the workplace were significantly associated with higher fingernail cortisol levels.

Table 2. Results of the multiple regression analysis for cortisol level in fingernails.

Independent variables	b (SE)		β	p
Gender (male/female)	−0.050	(0.044)	−.104	.263
Age	0.001	(0.002)	.051	.581
BMI	−0.002	(0.005)	−.039	.664
Smoking status (no/yes)	0.114	(0.041)	.248	.006
Stressful life events in the workplace (0/1+)	0.087	(0.036)	.209	.019
Job strain	0.056	(0.036)	.152	.125
Perceived stress	−0.003	(0.003)	−.082	.406

We found a moderate correlation between job stress and perceived stress (r = .45, p < .01), implying a possibility of collinearity in the regression analysis. However, excluding job stress or perceived stress from the regression analysis did not alter the results.

Discussion

This exploratory study investigated the relationship between fingernail cortisol level and psychosocial stress (stressful life events, job stress, and perceived stress) in a sample of middle-aged workers. We found that stressful life events in the workplace, but not job strain and perceived stress, were associated with higher cortisol levels in fingernails after adjusting for the effects of demographic factors. Fingernail samples reflect cumulative hormonal exposure in the past. Therefore, it could be interpreted that fingernail cortisol is associated with psychosocial stress in the past (stressful life events in the previous year), rather than in the present (job stress and perceived stress). Life events in the workplace such as changing to a different line of work are generally recognized as stressful (Holmes & Rahe, 1967). A recent retrospective study (Grassi-Oliveira et al., 2012) found a positive association between hair cortisol levels and number of negative life events, consistent with the findings of this study. To the best of our knowledge, this is the first study to demonstrate a relationship between fingernail cortisol level and psychosocial stress.

In this study, we also investigated the effects of demographic variables on fingernail cortisol, and found that smokers had higher fingernail cortisol levels. This result was consistent with previous findings on salivary cortisol, in which smokers exhibited higher cortisol levels (e.g. Steptoe & Ussher, 2006). However, we did not find significant associations between fingernail cortisol and other demographic variables (age, gender, BMI, and manicure use). BMI has been frequently reported to be associated with hair cortisol (Stalder et al., 2012). However, the lower mean BMI and comparatively limited BMI range of this study could have affected its results.

Fingernail samples could have some advantages, compared with use of saliva and hair samples. Salivary cortisol is known to have large diurnal rhythms and reflect acute increases in hormones, which would confound an investigation of the relationship between cortisol and chronic stress. In contrast, fingernail cortisol level would not be affected by such transient increases and diurnal rhythms. Furthermore, many strands are needed for the measurement of hair cortisol, and hair self-sampling may be difficult for some participants. In contrast, fingernail samples can be self-collected, and only small amounts are required.

This study has certain limitations, which warrant careful interpretation of its findings. First, in this study, we only investigated fingernail cortisol, but not salivary or hair cortisol. More information on relationships between fingernail and salivary or hair cortisol is a key component to advancing the study of fingernail cortisol. Second, we did not account for individual differences in nail growth rate, which could affect observed steroid concentrations. Further, in this restricted study design, participants clipped their fingernails at home (i.e. not in accurately controlled conditions). Third, we did not assess the effects of stressful life events outside the workplace or those of positive life events. While we did ask participants whether they experienced stressful life events outside the workplace (e.g. death of spouse, divorce, marital separation), the experience rate of most events was quite low, which made it difficult to correlate these events with hormone levels. Fourth, for the assessment of job stress, we did not clearly define the time period of job stress that participants should consider. However, reported job stress may largely reflect participants’ experience of stress in the present, but not in the past, because job stress is relatively unstable and variable across time (Kawada & Otsuka, 2014). Fifth, the sample size in this study was small, particularly that of the nail polish use analysis. Therefore, associations of psychosocial stress with fingernail cortisol could not be investigated separately for male and female participants. Furthermore, this study was conducted in healthy middle-aged workers, so not all age groups were represented in our sample.

In conclusion, we demonstrated that experience of stressful life events in the workplace in the previous year was associated with elevated fingernail cortisol level in a sample of middle-aged workers. Thus, we provided evidence that fingernail samples may retrospectively reflect individual differences in cortisol levels related to past psychosocial stress.

Disclosure statement

The authors report no conflicts of interest.

Additional information

Funding

This study was supported by a fundamental research grant from the National Institute of Occupational Safety and Health, Japan [H23026].