Prospective study of the association between depressive symptoms at type 2 diabetes diagnosis and time to insulin initiation in the South London diabetes (SOUL-D) cohort

Aims: Initiation of insulin is usually delayed even when required. We aim to estimate the association between depressive symptoms on time to become insulin requiring and time to insulin initiation.

Methods: 8-year follow-up of a cohort of newly diagnosed people with T2D recruited in south-east London, UK (2008-2012). Baseline depressive symptoms were assessed using the Patient Health Questionnaire-9. Time to insulin-requiring was defined when optimal glycaemic levels were not achieved (HbA1c >58 mmol/mol) at least three months after the 2nd oral antidiabetic was prescribed, and time to insulin initiation was defined as first insulin prescription.

Results: Seventy percent (n = 1166) of the baseline cohort was followed up. Median time to insulin requiring was 84 months (IQR 63-100) and to insulin initiation 93 months (IQR 79-105). Participants with depressive symptoms at baseline required insulin earlier (mean [SD] 73.64 [32.16] vs. 79.05 [29.07] months, p = 0.007) and were prescribed insulin sooner (82.53 [30.19] vs. 89.72 [22.02] months, p < 0.001). In Cox regression, depressive symptoms at baseline were not associated with time to insulin requiring (HR [95 % CI]; 1.16 [0.86-1.57], p = 0.34) nor to insulin initiation (HR = 1.00 [0.99-1.00], p = 0.49).

Conclusions: Depressive symptoms were not associated with time to insulin requiring and initiation after adjusting for potential confounding.