The determination of what stress measure to recommend is partially determined by your study design and whether you are aiming to measure a profile of current stress response or chronic/prolonged stress. The former would best be measured by salivary cortisol or alpha-amylase whereas the later, identifying a pattern of chronic stress, would best be measured via hair cortisol. If you are looking to measure "change" based on your web-based intervention, it is likely that you will want to use salivary cortisol which is well-established and many labs can conduct the assays for you. You will want to make sure to establish a solid baseline, compare all subjects during the same time of day (due to dirunal regulation of coritsol) and consider the 20-minute lag in which cortisol is detectable in saliva in your protocool. In order to measure hair cortisol your intervention will need to take place over many weeks in order to detect change in hair. If you want to measure month-by-month stress, then using hair may be feasible. Hope this helps.

Hi there, I would not dare to say "the best" but we have had good experience with measuring heart rate variabilty. There's good literature available (see work from Julian Thayer) and the procedure is non-invasive. I am interested to hear others thoughts about that as well.

Best!

Hi,

For acute stress, I think is ACTH; and for accumulated stress, Cortisol.

Now are procedures non-invasives to take samples of these hormones, only with saliva.

Best

Burkhard, many thanks for your suggestion - I have just read Thayer's 2012 paper in Neuroscience and Biobehavioral Reviews. I'll continue to follow it up, but it may involve a measurement procedure which is to complicated for our purposes.

Hi Juan,

Thanks for your suggestion. Measurement of salivary cortisol appears to be our best bet at present as an acute stress marker, and there's now a possible way of using hair cortisol levels as a measure of more chronic stress levels. Any comments of about the reliabiity and validity of these two approaches would be most welcome (as well as any suggestions about other biological markers).

Dear Peter,

To measure stress reactivity of the HPA axis, I would measure cortisol (either in saliva, hair or finger nails) or plasma levels of copeptin (also in saliva). To measure stress reactivity of the ANS, I would measure systolic and diastolic blood pressure as well as heart rate.

Hope this paper can help.

Dyadic Coping, Insecure Attachment, and Cortisol Stress Recovery Following Experimentally Induced Stress.

Article Dyadic Coping, Insecure Attachment, and Cortisol Stress Reco...

Hi, depending on the time your web based training is planned (e.g. weekly sessions over a period of several weeks), hair cortisol would be recommendable. It is a measure for chronic stress, which you are presumably trying to diminish by your WBT (i.e. that the workers are better able to deal with stress in general). 1cm of hair from the sculp represents one month of accumulated cortisol secretion. This measure is also non invasive, the hair matrix is easy to store and the results are not potentially biased by circadianic rythm or food or beverage intake. Right now we are trying to establish that biomarker here at our Institute for further resarch and the handling seems easy. DHEAS might also be an option (also possible in hair).

For clinical purposes we measure dhea-s, cortisol (3 measurements/day) epinephrin, norepinephrin, serotonin, dopamin, gaba and glutamate - saliva seems to represent the acute state - see also 'neurolab.com' hrv- measurements give a good overview of the sympathicus/ parasympathicus-balance

To me, the addition of cortisol levels to heart rate variability is definitely the best combination of stress biomarkers

Hi, the most traditionally way is by saliva, blood or urine and this method is validated and established. A drawback is that it only reflects momentary stress levels last 20 minutes. A new method maturing cortisol in hair has been tested and published in a growing number of articles. This method reflects retrospectively a mean of cortisol levels, but is only available in special research laboratories, like in ours in Sweden.

Good luck!

Depending on the timing of your experiments, a suggestion to include alpha amylase in saliva samples as a marker for noradrenergic activation. In addition to salivary cortisol and heart rate measurements this will give you a nice picture of the entire (acute) stress response.

I think the ACTH is the best marker of acute stress together the heart rate (in the moment), more than cortisol that is the ACTH precursor. Cortisol is best to acummulated o sub-acute stress (few hours or days).

I would use saliva samples to measure cortisol and alpha-amylase response to awakening. Saliva cortisol as peripheral marker of HPA axis and alpha-amylase as peripheral marker of ANS activity together with BP and HR.

Article Abnormal diurnal patterns of salivary α-amylase and cortisol...

Cortisol in hair is the best marker to measure long-term stress..we are measuring retrospeklively 3months backwards

Salivary biomarkers including cortisol and alpha-amylase levels are easy to obtain and very reliable. It is important to control the measures for diurnal variation and have subjects keep a stress/activity journal. In addition, it may be important to use a depression scale as an important co-variate of measurement. Heart rate variability is another non-invasive easily obtained biomarker, and increase in the high-frequency component is particularly useful in measuring stress resiliency over time.

I agree that hair cortisol is a very good biomarker of long-term (average) stress levels, which also offers the unique possibility to create retrospective timeslines (of at least several months to even years with some methods). We have the hair analysis method available at our laboratory in Rotterdam, in the Netherlands as well.

See for examples/review: Groeneveld MG et al Stress. 2013 Nov;16(6):711-5 and Staufenbiel S Psychoneuroendocrinology. 2013 Aug;38(8):1220-35

The determination of what stress measure to recommend is partially determined by your study design and whether you are aiming to measure a profile of current stress response or chronic/prolonged stress. The former would best be measured by salivary cortisol or alpha-amylase whereas the later, identifying a pattern of chronic stress, would best be measured via hair cortisol. If you are looking to measure "change" based on your web-based intervention, it is likely that you will want to use salivary cortisol which is well-established and many labs can conduct the assays for you. You will want to make sure to establish a solid baseline, compare all subjects during the same time of day (due to dirunal regulation of coritsol) and consider the 20-minute lag in which cortisol is detectable in saliva in your protocool. In order to measure hair cortisol your intervention will need to take place over many weeks in order to detect change in hair. If you want to measure month-by-month stress, then using hair may be feasible. Hope this helps.

Saliva cortisol in combination of heart rate are the best variable for this purpose. the effect of circadian rythem is very important and should be considered in sampling. if it is possible, blood level of vassopressin can also be measured.

I recommend the following articles

Toker, S., Shirom, A., Shapira, A., Berliner, S. The Association Between Burnout, Depression, Anxiety, and Inflammation Biomarkers: C-Reactive Protein and Fibrinogen in Men and Women. Journal of Occupational Health Psychology 2005, Vol. 10, No. 4, 344–362.

I would suggest to combine measurements of cortisol and ACTH from peripheral blood as readout of the HPA axis. Additionally, I would recommend not to use baseline levels, but stimulated, as with dexamethasone, for example the dexamethasone suppression test, because this produces more robust findings. We could increase the sensitivity to uncover HPA dysregulation / GR impairment with dexamethasone stimulated gene expression profiling, one of the top regulated genes is FKBP5, also a crucial marker of GR functioning. For GR functioning assessment we take blood at 6 pm, then administer dexamethasone 1.5 mg orally, and 3 hours later second blood draw, each time cortisol, acth and mRNA (i.a. FKBP5).

http://www.ncbi.nlm.nih.gov/pubmed/22237309

http://www.ncbi.nlm.nih.gov/pubmed/23406438

skin temp, heart rate, sweat detection, pupil dilation

amylase, sIgA, chromogranin A, cortisol in saliva

ACTH, cortisol, testosteron, IL-6, substance P, CD4/CD8 ratio, Natural killer cell in blood

*Considerations

type of stress: acute or subacute or chronic; fatigue or tension

(mock interview=acute, tension, colorword test=sub acute and fatigue)

burden, cost VS accuracy, authorization

rhythm: circadian rhythm, menstrul cycle

disturbance: tooth decay, stomatitis, contraceptives, smoking

HRV is the best marker by far!

Full Stop.

''Everyone'' agrees :)

See:

Prof Martin Seligman's quote:

''Heart rate variability (HRV), surprisingly, is another candidate for protection against cardiovascular disease. HRV is the short-term variation in beat-to-beat intervals, which is partly controlled by the parasympathetic (vagal) system of the central nervous system. This is the system that produces relaxation and relief. Accumulating evidence suggests that people with high heart rate variability (HRV) are healthier, have less CVD, less depression, and better cognitive abilities.'' http://www.authentichappiness.sas.upenn.edu/newsletter.aspx?id=1569

Galvanic skin response (GSR) is a classic (electrode on the skin). Particularly useful to measure acute stress, but chronic stress can be measured by comparing average GSR to a standard population. In the latter case, you'd need additional measures (questionaires or biomarkers) to differentiate between chronic stress and e.g. hyperhydrosis or eczema.

Another ting to keep in mind: Cortisol/HPA axis measures do not always line up with measures of subjective stress. In fact, in some meta-analyses, the correlation between cortisol and subjective measures is near zero. This is not because people are "lying" or because cortisol is not a useful measure. Self-report and cortisol are kind of measuring two different things. The HPA axis is activated when the brain has calculated that energy is needed. One of the main functions of cortisol is to mobilize blood glucose and otherwise make energy available (esp. to the limbs). So, some highly negative mood states do not activate the HPA axis at all. In terms of chronic stress, in some cases (burn-out, PTSD) the HPA axis is dampened or blunted, so cortisol levels are lower than average even though the person is experiencing a great deal of stress. So you can't expect a one-to-one relationship between the amount of subjective, psychological stress a person is experiencing and their cortisol levels.

All of methods mentioned below are good. Depending what type of stress you would like to use. I would recommend hear or saliva cortisol, ACTH in blood and some hemodynamic features such as HRV.

Chronic stress can lead to a persistent increase in cortisol levels and decreasing levels of DHEA ( dehydroepiandrosterone ), which is detrimental to the functions of the brain and other organs of the body. Increased cortisol ( stress hormone ) affects immune responses such as the level of secretory IgA (sIgA) and the level of anti- gliadin antibodies ( AGA ) . Synthesis of secretory IgA, the first line of host defense against pathological microorganisms, and food antigens in the intestinal mucosa is reduced during stress. Level Antigliadin antibodies are elevated in people with celiac disease and intolerance to gliadin . Stress-related reduction in the gut immune defense results in increased permeability of the intestinal wall and enhance AGA in the bloodstream.

In the analysis of the profile of "Stress and adrenal hormones " used non-invasive procedure ( saliva test ) to monitor the activity of the adrenal cortex and its ability to respond to stress.

The procedure allows us to investigate the biological cycle of cortisol and DHEA , repeated every 24 hours .

If the analysis revealed the hormonal changes that accompany chronic stress , it is necessary to take measures to reduce the effects of stress on the body: to make lifestyle changes to adjust diet and enter into the diet of nutritional supplements to support the activity of the adrenal cortex and the recovery of a physical barrier in the intestine. As the prevention of mental disorders can be consumed yogurt rich in probiotics. Due to the content of the product, getting better bowel function, and it helps to communicate with probiotics brain through the nervous system. As it turned out, getting into the body, the nutrients being squeezed out hazardous to health bacteria and prevent them from multiplying.

It depends. Chronic stress, stress from trauma, environmental stress, workplace stress, PTSD: Depending on which is investigated, all or a few of the following could help

Saliva cortisol, ACTH in blood, and hemodynamic features, and stress-related survey questions (some surveys do a good job with just 7 questions). Here are some Cortisol links:

www.uni-duesseldorf.de/~ck/index.html

http://www.aeron.com/new_page_27.htm

Info

http://www.macses.ucsf.edu/Research/Allostatic/notebook/salivarycort.html

Some background values for cortisol

http://www.cushings-help.com/salivary_cortisol.htm

http://www.clinchem.org/cgi/content/full/49/1/203

http://www.clinchem.org/cgi/content/full/48/1/207

http://www.pituitarysociety.org/public/specific/salivarycortisol

http://patients.uptodate.com/topic.asp?file=adrenal/6117

http://www.sciencedaily.com/releases/2000/01/000120073039.htm

http://www.calgarylabservices.com/LabTests/AlphabeticalListing/C/Cortisol-Salivary.htm

http://www.docguide.com/news/content.nsf/news/8525697700573E1885256D0100680390

http://www.uchospitals.edu/news/2000/20000113-cortisol.html

I have also used stress-related serum lipid parameters such as Total Cholesterol, HDL Cholesterol (HDL-C) and their ratio, to assess the outcome of a stress management intervention program: at follow-up (7 months) the results were as predicted, with higher HDL-C in the treatment group as compared to controls; it can be a useful variable, if you need to assess a persistent effect. Other metabolic parameters can also be used, at least at short term, such as blood glucose.

Besides the good advice listed by others, you might also consider what disease outcome are you most interested in making links to (e.g., coronary artery disease, cancer, infectious disease etc.). Then determine what is the epidemiological / biomedical evidence that these measures are linked to either the development or exacerbation of the disease of interest (especially with the assessment context that you have in mind). However, if you simply want a reliable measure of the bodily response to stress then you'll need to consider the underlying biological mechanisms in terms of how these assessments may or may not be related so that you can make reasonable predictions about expected changes and potential discrepancies. Please also see Tom Kamarck's excellent work on how to increase the reliability of biological stress assessments via psychometric principles.

Hi everyone, I am following this one, because I am very interested in finding a survey that has been validated up against biomarker... @Gregory Zarus. do you have any references on this side issue?

Here is a recent publication that includes some (123) excellent references:

Critical Biological Pathways for Chronic Psychosocial Stress and Research Opportunities to Advance the Consideration of Stress in Chemical Risk Assessment;

Bruce S. McEwen from EPA and Pamela Tucker from CDC.

Am J Public Health. Published online ahead of print August 18, 2011: e1–e9. doi: 10.2105/AJPH.2011.300270)

(Dr. Tucker has assisted me with the development of stess evaluations during exposure events. She has facilitated some expert pannels on stress here at CDC --some of the minutes of those meetings are published internally at CDC's)

You could try to measure copeptin over time, although this could reflect more an acute rather than a chronic response.

Rachel Yehuda has found that the dexamethasone suppression test (DST) is a fairly reliable measure in people with PTSD to examine HPA axis dysregulation. The subject is given a dose the night before and blood is drawn the following morning for cortisol and ACTH . Some investigators use a dye in the DEX and a urine is collected the next morning to make sure the subject has actually taken the DEX. Salivary cortisol is not a reliable measure; even the morning cortisol awakening response (CAR) has not been found to always show differences. There are so many variable that affect circulating cortisol, such as sleep, diet, unaccounted for acute stress (traffic, arguments with a spouse etc), negative affect, co-morbid disease. Spot urine or plasma catecholamines have also been reliable for showing acute stress effects.

What a great discussion! I am wondering if anyone has any input about colic/excessive crying/fussing as a biological stress and what a good non-invasive biomarker would be? I am not sure that cortisol is the best measurement and was thinking of salivary IgA. Any other thoughts would be appreciated!

Methemoglobinemia is often unrecognized and underemphasized.It is a condition in which hemoglobin is oxidazed to the ferric form and is unable to transport oxygen to tissues, therefore causing hypoxia, and primarilly occuurs when erythocytes are affected by xenobiotics and farmaceutical compounds with toxicologic properties, such as volatile organic compounds, oxidants, nitrogen oxides, peroxinitrites, phenacetin, and sulfonamides

In attachment I present some my research results as contribution on the question:

What is the best biomarker of stress in human subjects?

In attachment is my full text which have been a good interest in the scientific community.

Apparently, you are looking for biological biomarkers of a condition of psychosocial distress, before and after a psychosocial intervention. Correct?

If so, you should also use measures of psychosocial stress, do you?

Lucio, we are planning also to use subjective measures of both stress and quality of life.

All, very many thanks for your helpful comments and suggestions so far.

Lucio, It is corect, but obviously not complete. I point-out the importance of biological biomarker , the condition for your and Peter sugested psychosocial interventions.My intention is to take aime at the confirmation of the expected significant relations. Are you agree with?

Peter, if you have not already chosen a self-report measure of stress, may I suggest the Stress Overload Scale?

I was very frustrated by the poor reliability and validity of the stress scales that were available in my own research, so I took time (5 years!) to develop a measure of my own. The SOS is consistent with stress theories, derived from a sequence of factor analytic and psychometric studies, and based on diverse community samples. Moreover, it has two sub scales: Event Load and Personal Vulnerability, which can be summed for a continuous score, OR they can be crossed to form a diagnostic grid. This grid classifies respondents according to their risk for subsequent stress-related pathology.

In a recent validity study (not yet published), high SOS scores proved to be predictive of a blunted cortisol response to a laboratory stressor (a sign of chronic stress).

The SOS is available for free download on my ResearchGate page...

James, many thanks for this. I will chase it up. Our selection of outcome measures is still a 'work in progress'. but Cohen's PSS has been high on our list.

Hi, Peter.

The PSS is a good scale, just not as psychometrically sound as the SOS. But it is shorter.

If you have a chance, read the lit review of my 2012 paper ("Stress Overload: A New approach to the assessment of stress"-- can be downloaded from ResearchGate). It compares all the popular measures of stress...

Yours, Jim

Jim, I have just read your 2012 paper (Table 1 contains a very useful summary), and the SOS looks very good. However, our subjects will be Thais, and one of the big advantages for us in using the PSS that there is a Thai version of the later 10-item scale which has been found to perform well in a Thai population.

Best, Peter

That is understandable. I have several translations of the SOS, but none for a Thai population! Good luck with your research,

Jim

Dear Mr. Bradshaw,

I just came across the following article:

http://www.reuters.com/article/2014/02/17/health-depression-biomarker-idUSL6N0LM27720140217

Good Luck with your experiments.

BR,

Zeynep

The time gap is essential: a basic distinction should be made between acute - subacute and chronic distress. In fact, with the passing of time, depressive symptoms arise and these can be tracked with the biological biomarkers already cited by the colleagues. Nothing new: remember Selye's General Adaptation Syndrome? The problem is: upto what time is the psychosocial stress to be considered "acute" and when does it become "chronic"? Sorry to ask even more questions, but I feel they are inescapable.

It is indeed an interesting topic. How does one best study it? Those stress events that are studied best are studied by several different researchers. Conclusions appear to tilt toward the researcher’s agenda or hypothesis, but when thoroughly studied by people looking at things from all sides, some kind of truth emerges.

I find the studies of stress associated with residents near Three Mile Island interesting. (There are many and I will just mention a few.) We know that exposure from the TMI event was relatively low (the gas was contained, measured, and then controllably released), yet the reporting of the incident and the preventative recommendations to the public suggested much worse. When accurate reports after the fact suggested much lower exposures, the residents did not believe them. Early stress of the event and follow-up stress of what appeared to be a cover up created a community stressed by publicity.

Flemming et al 1982 found that early stress was reduced by community support and connections. Yet stress returned, and other studies suggest from the reopening of TMI. Dew et al. 1987 suggested that acute stressors increase depression, hostility, and risk of accidental injury; Davidson et al 1990 found that the TMI residents performed worse on finding proof-reading errors and had higher measures of stress (blood pressure, urinary epinephrine and norepinephrine, answers to questions) than controls. The Dew and Davidson studies differed in their findings with Dew finding increased symptom reporting in the years that corresponding to the re-starting of TMI operations and Davidson found little in the way of symptom reporting (although they did find the other markers of stress.

http://link.springer.com/article/10.1007%2FBF00974585#page-2

Cortisol or ACTH presumably!

The question is interesting. Salivary cortisol , Dexamethasone Suppression Test were abandonned in the early 90' because failing to detect endogenous depressive disorders due to a lack of sensitivity and specificity.

It would probably make sense that some inflammatory parameters (CRP, etc, some Cytokines) show positive reactions . In practice , we do not observe a systematic response . Does the positive spot / flag depends on the type of Burn oout (seriousness, course, intercurrent problems?) Research deliver contradictory responses.

Finally, some studies about specific patterns / records profils observed with evoked potential and polysomnography seem to pop up .

Other publication

Best regards

I think increase in adrenalin level in the blood indicates the corresponding increase in heart rate. In crease heart rate is related to anxiety, fear and various types of stress. Also mental stress and anxiety is directly correlated with the level of neurotransmitter serotonin.

Hi - you also might want to check out the Galvanic skin response (GSR), which is a measure of a change in the voltage across a skin electrode due to increased sweating. It seems to be quite a good stress measure (under otherwise constant environmental conditions in terms of humidity, room temperature, amoun tof physical exercize, etc). I guess you might want to combine multiple complementary stress measures to provide the most accurate representation of an actual physical and psychological stress response.

Hello Peter, cortisol in hairs are the best stress marker, as hair samples can be stored for long time at room temperature only

Plasma and salivary cortisol is the best stress marker.. 

I think chronic the stress assessment cannot be done by a unique biomarker.

I agree with several of the biomarkers that have been proposed as cortisol and ACTH in saliva, for me, the first the best if you will do only one biomarker. You have to avoid the blood tests, because they can mask the results with an acute stress produced by the situation (the puncture). It would be not good to take only one measurement in saliva, is best do the adrenal stress index curve test, done since 8 am. till 12 pm. each 4 hours.

Also, it should be included a psychological test for the stress. It could be good to include a Holter monitor test for the measurement of the pressure and the heart frequency along the same day and collect urine during 24h for the analysis of catecholamines.

Other test would be a polysomnography, many times, the chronic stress is associated sleep disorders. Days before the adrenal test, could be done the blood test for other biomarkers, including gene biomarkers of cardiovascular risc.

An echocardiography for discard the heart hypertrophy also could be interesting, due the increase of adrenaline by the stress, that leads the increase of heart pressure, and hence, stress can lead to the increase of the left ventricle size.

In 2009 I encouraged a multidisciplinary group of colleagues (Internal Medicine, Cardiology, Psychiatry, Psychology, Genetics, Clinical Analysis and Pulmonology) of the hospital and we create a Stress Unit, focused on the cardiovascular effects of the stress and the prevention of them. The proposed protocol was similar to what I have commented. http://www.sanitas.es/sanitas/seguros/en/particulares/physicians-and-medical-centres/sanitas-hospital-cima/special-units/stress-test-unit/index.html