Effect of Depression Treatment on Somatic Depressive Symptoms and Cardiometabolic Biomarkers among People without Diabetes

Abstract

While depression is a risk factor for type 2 diabetes, little is known about the effect of depression treatment on diabetes risk markers. Using data from the recently completed eIMPACT trial (NCT02458690, supported by R01 HL122245), I examined if depression intervention improves diabetes risk markers and if improvements in somatic depressive symptoms mediate potential intervention effects. 216 participants (primary care patients ≥50 years with depression and elevated cardiovascular disease risk from a safety net healthcare system) were randomized to 12 months of the eIMPACT intervention (modernized collaborative care intervention involving internet cognitive-behavioral therapy [CBT], telephonic CBT, and/or select antidepressants; n=107) or usual primary care for depression (primary care providers supported by embedded behavioral health clinicians and affiliated psychiatrists; n = 109). Given my focus on diabetes risk, I excluded participants who did not attend the post-treatment visit (n = 17) or who had a diabetes history at pre-treatment (n = 73), leaving a final sample of 126 (n=66 intervention, n=60 usual care; Mage = 58 years, 79% women, 50% Black, 47% with income <$10k/year). I computed depressive symptom severity variables from the Hopkins Symptom Checklist-20 (SCL-20) items: hyperphagia (“overeating” item), poor appetite (“poor appetite”), hypersomnia (“sleeping too much”), disturbed sleep (“sleep that is restless or disturbed”) and SCL-15 (mean of items not pertaining to appetite or sleep). I calculated insulin resistance from fasting plasma glucose and insulin using the Homeostasis Model Assessment for Insulin Resistance (HOMA-IR)-2 calculator, body mass index (BMI) from measured height and weight, and plasma concentrations of high-sensitivity C-reactive protein (hsCRP), leptin, and ghrelin using ELISA kits. Parallel mediation analyses revealed that 12 months of modernized collaborative care for depression improved both directions of sleep symptoms but did not improve poor appetite or hyperphagia – the somatic symptom most consistently linked with increases in HOMA-IR, BMI, hsCRP, and leptin. Of the five cardiometabolic biomarkers examined, the eIMPACT intervention decreased only hsCRP and ghrelin. There were no intervention effects on HOMA-IR, BMI, or leptin. In addition, no somatic depressive symptoms mediated intervention effects on the cardiometabolic biomarkers, nor did race moderate any mediation effects. Further research is warranted to determine best practices for targeting hyperphagia and reducing cardiometabolic disease risk among people with depression.