Depression, the most frequent and harmful mental disorder, has been associated with specific somatic diseases as the leading cause of death. The purposes of this prospective study were to predict incident chronic diseases based on baseline depressive symptoms and to test sex-dependent effects.

In a representative German community sample of over 12 000 participants, baseline depressive symptoms (assessed using the Patient Health Questionnaire-9) were tested as a predictor of new onset of cardiovascular disease (CVD), chronic obstructive lung disease, diabetes, cancer, and migraine at 5-year follow-up. To study disease incidence, we created subsamples for each chronic disease by excluding participants who already had the respective disease at baseline. Potential confounders were included in logistic regression models and sex-specific analyses were performed.

Controlling for demographic characteristics and loneliness, in men and women, baseline depressive symptoms were predictive of CVD, chronic obstructive lung disease, diabetes, and migraine, but not of cancer. When we additionally adjusted for metabolic and lifestyle risk factors, there was an 8% increase of chronic obstructive lung disease and migraine per point of depressive symptoms. There was a trend for CVD (4%; p = 0.053). Sex-sensitive analyses revealed trends for the relevance of depressive symptoms for CVD in men (p = 0.065), and for diabetes in women (p = 0.077).

These findings underscore the need to implement screening for depression in the treatment of major somatic illnesses. At the same time, depressed patients should be screened for metabolic and lifestyle risk factors and for somatic diseases and offered lifestyle interventions.

Major depression and anxiety disorders are known to negatively influence cognitive performance. Moreover, there is evidence for greater cognitive decline in older adults with generalized anxiety disorder. Except for clinical studies, complex executive planning functions and subclinical levels of anxiety have not been examined in a population-based sample with a broad age range.

Planning performance was assessed using the Tower of London task in a population-based sample of 4240 participants aged 40–80 years from the Gutenberg Health Study (GHS) and related to self-reported anxiety and depression by means of multiple linear regression analysis.

Higher anxiety ratings were associated with lower planning performance (β = −0.20; p < 0.0001) independent of age (β = 0.03; p = 0.47). When directly comparing the predictive value of depression and anxiety on cognition, only anxiety attained significance (β = −0.19; p = 0.0047), whereas depression did not (β = −0.01; p = 0.71).

Subclinical levels of anxiety but not of depression showed negative associations with cognitive functioning independent of age. Our results demonstrate that associations observed in clinical groups might differ from those in population-based samples, also with regard to the trajectory across the life span. Further studies are needed to uncover causal interrelations of anxiety and cognition, which have been proposed in the literature, in order to develop interventions aimed at reducing this negative affective state and to improve executive functioning.

To test the vascular depression hypothesis in the general population, we analyzed the association between current depression, medical history of depression, cognitive and somatic depressive symptom dimensions and measures of atherosclerosis [intima–media thickness (IMT) and carotid plaques].

We included a representative sample of 5000 participants from the Gutenberg Health Study (GHS). Depression was assessed by the nine-item Patient Health Questionnaire (PHQ-9), and IMT and carotid plaques were measured at both common carotid arteries using an edge detection system. Regression analyses were performed separately for participants with and without cardiovascular disease, adjusting for medical history, cardiovascular risk factors and psychotropic medication.

Contrary to hypotheses, we found no increased IMT for somatic symptoms of depression; the same was true for depression and cognitive symptoms in the fully adjusted model. Only a moderate relationship between medical history of depression and the presence of atherosclerotic plaques was maintained after correction.

The relationship between depression and atherosclerosis may be more complex than previously assumed. Although the vascular depression hypothesis was not supported, our results support the hypothesis that lasting depression leads to arteriosclerosis.