Persistently high and fluctuating trajectories of total and somatic depressive symptoms increase diabetes risk: Two prospective cohort studies.

Background: Depression is a significant risk factor for diabetes, particularly type 2 diabetes. However, depressive symptoms differ from clinical depression. Previous research has not fully considered the relationship between the trajectory of depressive symptoms and the risk of developing diabetes over time.

Aim: To investigate the association between depressive symptoms, their trajectories, and the risk of developing diabetes in two prospective cohort studies.

Methods: In the first phase we analyzed the association between depressive symptoms and the risk of developing diabetes separately using the Health and Retirement Study (HRS). Depressive symptom trajectories were assessed by examining changes in depressive symptoms at baseline and again 8 years later. We then identified specific depressive symptom trajectories that increased the risk of diabetes in the second phase. Finally, we confirmed the association between depressive symptoms and their trajectories with diabetes risk using the English Longitudinal Study of Ageing (ELSA) as a validation study. Depressive symptom trajectories were categorized into five states based on changes in the modified 8-item Center for Epidemiological Studies-Depression scores: Persistently high; increasing; fluctuating; decreasing; and persistently low. Diabetes mellitus was defined as self-reported, physician-diagnosed diabetes. Cox proportional hazards models were used to assess hazard ratios (HR) and 95% confidence intervals (CI), adjusting for potential confounders.

Results: In the first phase a total of 27658 participants were included (HRS: 18633, ELSA: 9025), among whom 6582 had depressive symptoms (HRS: 4547, ELSA: 2035), 6407 had somatic depressive symptoms (HRS: 4414, ELSA: 1993), and 26415 had cognitive-affective depressive symptoms (HRS: 17755, ELSA: 8660). We found that overall depressive symptoms (HRS: HR = 1.14, 95%CI: 1.07-1.22; ELSA: HR = 1.18, 95%CI: 1.03-1.34) and somatic depressive symptoms (HRS: HR = 1.14, 95%CI: 1.07-1.22; ELSA: HR = 1.25, 95%CI: 1.10-1.42) increased the risk of diabetes, while cognitive depressive symptoms were not associated with diabetes risk. Over an 8-year follow-up we identified 19729 trajectories of overall, somatic, and cognitive-affective depressive symptoms (HRS: 13918, ELSA: 5811). In the second phase we found that persistently high (HRS: HR = 1.22, 95%CI: 1.06-1.40, ELSA: HR = 1.54, 95%CI: 1.16-2.05 in total and HRS: HR = 1.24, 95%CI: 1.07-1.43, ELSA: HR = 1.79, 95%CI: 1.36-2.35 in somatic) and fluctuating (HRS: HR = 1.09, 95%CI: 1.01-1.17, ELSA: HR = 1.33, 95%CI: 1.14-1.55 in total and HRS: HR = 1.10, 95%CI: 1.02-1.18, ELSA: HR = 1.31, 95%CI: 1.13-1.53 in somatic) trajectories of overall and somatic depressive symptoms increased the risk of diabetes, while increasing trajectories may also raise diabetes risk. However, decreasing trajectories were not associated with diabetes risk. Cognitive-affective depressive symptoms showed no association with diabetes risk regardless of trajectory changes. Sensitivity analyses confirmed the reliability of the findings.

Conclusion: Persistently high and fluctuating trajectories of overall and somatic depressive symptoms increased the risk of diabetes, while decreasing trajectories were not associated with diabetes risk. In contrast trajectories of cognitive-affective depressive symptoms show no relationship with diabetes risk. Focusing on depressive symptom trajectories, particularly those of somatic depressive symptoms, represented a viable strategy for future diabetes prevention.