Moderate-to-severe depression symptoms are associated with poorly controlled glucose levels in a real-world sample of individuals with mental illness and type 2 diabetes mellitus

Objective

To investigate whether depression symptom severity is associated with poor long-term glycemic control among individuals with mental illness and co-morbid type 2 diabetes mellitus (T2DM).

Methods

2842 psychiatry outpatients (PCARES Registry, 2015–2020) were included. T2DM diagnosis and all available glucose labs were extracted from electronic health records. Scores on the 9-item patient health questionnaire (PHQ-9) provided baseline depression severity: 0–9 (none-mild), 10–27 (moderate-to-severe). With baseline lab within (±) 90 days of the baseline PHQ-9 date, all follow-up labs had to be ≤ 365 days of the preceding lab, and not exceed one-year after the second follow-up lab date. 1255 individuals met the timeline criteria for glucose. Linear mixed-effects models provided coefficients for the association between depression symptom severity and long-term glucose levels, after adjusting for socio-demographics, BMI, anti-psychotic medications, and follow-up time.

Results

Among 1255 patients with a mean ± SD age (45.9 ± 16.8 years), PHQ-9 score (11.9 ± 7.1), and glucose (111.7 ± 45.1 mg/dl), 65% identified as females, 85% as non-Hispanic white, 60% had moderate-to-severe depression symptoms (N = 753) and 31% (N = 390) had T2DM. Individuals with moderate-to-severe depression symptoms (N = 245) showed a significant long-term increase in glucose levels at 6.7 (2.7) mg/dl over follow-up, indicating poor glycemic control (P = 0.01), whereas those with none-to-mild depression symptoms (N = 145) showed no significant long-term changes in glucose levels (P = 0.40); (P _{depression symptom severity X follow-up time} = 0.03)

Conclusions

Our findings support the need for improved diabetes care for patients with mental illness and T2DM and regular depression screening among individuals with T2DM.