A Saliva Test to Detect Risk of Future Depression? Not Yet

New research, published earlier this week in the Proceedings of the National Academies of Science, purported to identify the first biomarker for clinical depression.

What most media outlets failed to note was that this was not the first study to look at cortisol levels and their relationship to depression. In fact, it’s an area of research that has quite a few studies.

And what has the vast majority of the research in this area found? That a saliva biomarker test for depression is still a long ways away from becoming a reality.

A lot of the mainstream media is mis-reporting on the new study (Owens et al., 2014), leaving out a pretty important component of the research. You can see it right there in the title of the study:

Elevated morning cortisol is a stratified population-level biomarker for major depression in boys only with high depressive symptoms

You see that there? It’s a “biomarker” only in boys with high depressive symptoms. If the boy doesn’t have sub-clinical depressive symptoms, this biomarker tells you nothing.

And if a teen already has depressive symptoms, well, it’s not too far a leap to understand that that is a population at greater risk for depression. Depression symptoms beget depression — you’re at greater risk for future depression if you have depressive symptoms now.

But here’s what we already know, summed up nicely here in this 2013 study’s findings (Dietrich et al):

The study’s most convincing support is for higher cortisol in the morning (mainly CAR) in relation to overall depressive problems (both cognitive-affective and somatic) in children aged 10-to-12 years. […]

Patterns of higher cortisol (basal morning cortisol levels, CAR) in relation to anxiety and depression, and lower cortisol in relation to aggression (in boys) were generally consistent with the literature.

Our study further supported an association between higher cortisol and aggressive problems specifically in girls, as already suggested in previous TRAILS analyses in the population cohort. […]

Finally, this study suggests an association of higher morning cortisol levels with cognitive anxiety (referring to worry and anticipatory anxiety) rather than with somatic anxiety problems (referring to bodily panic symptoms).

But here’s the kicker. The researchers then go on to note that the associations they found are “very weak and therefore only of limited relevance.”

In the dozen or so studies looking at depression and cortisol, the evidence has shown time and time again that this is a weak link. Not one you’d hang you hat on, much less base a scientific test on.

As the 2013 researchers said, “it has been increasingly recognized that associations between cortisol and internalizing and externalizing problems [such as depression or anxiety] are weaker and more inconsistent than previously assumed.”

This remains true today, despite the hype.


Dietrich et al. (2013). Cortisol in the morning and dimensions of anxiety, depression, and aggression in children from a general population and clinic-referred cohort: An integrated analysis. The TRAILS study. Psychoneuroendocrinology, 38, 1281-1298.

Owens et al. (2014). Elevated morning cortisol is a stratified population-level biomarker for major depression in boys only with high depressive symptoms. Proceedings of the National Academies of Science. DOI 10.1073/pnas.1318786111