Getting under the skin: does biology help predict chronicity of depression?

Abstract Background Depressed patients are at risk of an unfavourable course including chronic episodes. Various psychiatric characteristics have shown to be predictive of depression's course trajectory, but whether indicators of somatic health further contribute to course prediction remains unclear. This study aimed to identify somatic health indicators (i.e. biomarkers, health status and lifestyle) that predict 2-year chronicity above and beyond an extensive list of sociodemographic and psychiatric characteristics. Methods Data are from patients with current depression at baseline (n = 903) and available 2-year follow-up participating in a longitudinal cohort study. Baseline demographic, psychiatric and somatic health indicators were associated with 2-year course trajectories, classified as non-chronic versus chronic Results At 2-year follow up, 40% of the patients showed a chronic course. Of the twenty tested somatic health indicators, short sleep and high interleukin-6 improved the regression model predicting chronicity with a significant, but modest, effect (ROC = 0.78; p = 0.03). Limitations Due to the observational design we did not have the ability to reliably consider the impact of psychiatric treatment. More elaborate information on somatic health such as dietary patterns would strengthen the study. Conclusions This study showed that short sleep duration and high interleukin-6 contributed significantly to the regression model as independent predictors, suggestive of clinical implications for patients with sleep disturbances and elevated inflammation levels. Other somatic health indicators did not add to the model. Overall, somatic health indicators showed modest additive value for predicting chronic course above and beyond sociodemographic and psychiatric indicators.