A frequent chief complaint in my office is anxiety and depression. Many years ago when I was in training at UCSF, I often noticed when collecting a very detailed history in depressed individuals that they reported a period of anxiety about six to nine months before they became depressed. Lifetime prevalence of major depressive disorder (MDD) is 16.2% and 28.8% for anxiety (ANX). Fifty-nine percent of individuals with lifetime MDD were co-morbid for anxiety. ANX tends to present between the ages of 6 - 20 whereas MDD is spread across adolescence through adulthood.
Understanding the nature of the relationship between ANX and MDD is important because individuals with these co-morbid conditions tend to have greater symptom severity and disability, higher utilization of treatment, and poorer therapeutic response. The two models researchers have used to explore this relationship are shared etiology (cause) or direct causation. Direct causation states that one disorder causes or makes it more likely to develop the other disorder. My observation that ANX often precedes MDD and foreshadows the course of MDD is supported by many researchers and has led to the speculation that ANX is causally related to MDD. The shared etiology model postulates that a common group of risk factors lead to the development and expression of both ANX and MDD. The factors that have been researched include parental mental health problems, non-intact family structure, family dysfunction, socio-economic disadvantage, and peer dysfunction.
Now come A.R.Mathew and colleagues, reporting a study exploring the relationship between risk factors and the development of ANX and MDD as well as the relationship between the development of ANX on MDD and MDD on the development of ANX. They studied 1709 adolescents (aged 14-18) from nine high schools in western Oregon. They made observations at four points with the final point being when the subjects were 30 years of age. I don't plan to discuss the statistics or methods but simply set out the results.
As expected, they found significant associations between MDD and ANX. They were shown to have shared and distinct risk factors. Low self-esteem was predictive of ANX while attributional style and subthreshold depression were predictive of MDD. Common predictors of ANX and MDD included female gender, family social support, worry, loneliness, friend social support, and emotional reliance.
The temporal order of onset seems to suggest different etiological models. ANX remained a significant predictor of MDD even when common risk factors were controlled for. This extends prior research in that this study controlled for risk factors where prior studies did not. On the other hand, MDD was not predictive of ANX when common risk factors were included. Therefore the development of ANX is most likely related to common risk factors.
Some additional findings are:
In conclusion, this study supports different etiologic models of ANX-MDD co-morbidity depending on the order of onset. The path from ANX to MDD is most congruent with the causation model but the path from MDD to later ANX is most congruent with the shared etiology model.