Depression, Depressive Symptoms, Dysthymic Disorder, Insulin Resistance, Major Depressive Disorder, PreDiabetes, Type 2 Diabetes
Conditions
Keywords
Primary Care, Cognitive-Behavioral Therapy, Computer-Based Psychotherapy, Antidepressant Medications
Brief summary
This pilot randomized controlled trial seeks: (1) to determine the preliminary efficacy of our modernized collaborative care intervention for depression in improving the diabetes risk markers of hemoglobin A1c and insulin resistance and (2) to explore whether somatic depressive symptoms - i.e., hyperphagia (increased appetite/weight) and/or hypersomnia (increased sleep) - moderate the effect of the eIMPACT-DM intervention on diabetes risk markers.
Detailed description
Diabetes affects 31 million (12%) U.S. adults, and another 82 million (34%) adults have prediabetes, a precursor to diabetes. The ramifications of diabetes are grave and include cardiovascular disease, disability, and death. While these statistics highlight the importance of diabetes prevention, current approaches have only partial effectiveness. This has created a clear need to identify new primary prevention targets and approaches for diabetes, and depression and depression treatment are strong candidates in this regard. Over 20 years of evidence indicates that depression is an independent, clinically important, robust, biobehaviorally plausible, and modifiable risk factor for diabetes. However, research has yet to determine whether depression treatment can prevent the development of diabetes in people with prediabetes. Given that depression is still receiving limited attention in settings where diabetes prevention occurs (e.g., primary care), there is a large cohort of patients with an underdetected or undertreated diabetes risk factor (depression). This status quo and the strong state of the depression-to-diabetes science create the need for a pilot randomized controlled trial to evaluate the utility of depression treatment as a new diabetes prevention strategy. Thus, we propose a pilot RCT of 64 primary care patients (50% minority) with a depressive disorder and prediabetes. Patients will be randomized to 6 months of eIMPACT-DM (intervention) or Active Control (comparator). eIMPACT-DM is our modernized collaborative stepped care intervention consisting of (1) computerized and telephonic cognitive-behavioral therapy for depression and (2) select antidepressant medications included in an algorithm optimized for diabetes risk reduction. Our preliminary data establish the feasibility and antidepressive efficacy of eIMPACT-DM. The Active Control consists of depression education, symptom monitoring, and primary care for depression. Our primary aim is to determine the preliminary efficacy of eIMPACT-DM in improving the diabetes risk markers of hemoglobin A1c (primary outcome) and insulin resistance (secondary outcome). Our exploratory aim is to explore whether somatic depressive symptoms - i.e., hyperphagia (increased appetite/weight) and/or hypersomnia (increased sleep) - moderate the effect of eIMPACT-DM on diabetes risk markers. A positive pilot trial would pave the way to an R01-level RCT by: (1) generating critical proof-of-concept data (eIMPACT-DM can improve A1c) to support the premise of the definitive trial; (2) providing preliminary effect sizes for eIMPACT-DM on diabetes risk markers to help justify future power analyses; (3) identifying a potentially important moderator of eIMPACT-DM efficacy that may need to be incorporated into the definitive trial. Ultimately, demonstrating that depression treatment reduces diabetes risk would identify a novel target (depression) for diabetes prevention efforts, and it would equip healthcare providers with a new practical, scalable, and disseminable intervention (eIMPACT-DM) to help lower diabetes risk for a large cohort of high-risk patients. These practice changes should translate into reduced diabetes morbidity, mortality, and costs.
Interventions
BEHAVIORALGood Days Ahead (GDA)
GDA (Empower Interactive) is an empirically supported, HIPAA compliant, computerized CBT for depression appropriate for primary care patients and people with little computer experience. GDA uses an interactive, multimedia format to deliver 9 45-minute sessions, the structure and content of which mirror face-to-face CBT. General topics include identifying and modifying automatic thoughts, using behavioral activation and other behavioral methods, identifying and modifying schemas, using effective coping strategies, and employing other core CBT methods. GDA is empirically supported - it is acceptable to patients, achieves superior depression outcomes to waitlist comparators, and yields equivalent (noninferior) depression outcomes to standard face-to-face CBT. To minimize time/transportation barriers, GDA sessions occur at the PI's lab or a location with internet access selected by the patient (patient's, family member's, or friend's home).
PST-PC is an established, manualized, empirically supported CBT developed for primary care. During the 6-10 30-minute sessions, patients are taught skills for solving problems contributing to depression. We will deliver PST-PC by phone, which has been found to be feasible and efficacious.
We first considered all FDA-approved antidepressants and excluded those with weight gain effects (tricyclics, paroxetine, mirtazapine) and those rarely used in primary care (MAOIs). Then, we used existing evidence to inform the structure. We made bupropion (an aminoketone) and fluoxetine (an SSRI) our first-line and second-line antidepressants, as meta-analyses indicate that their use is associated with weight loss. We made other SSRIs (escitalopram, sertraline) and SNRIs (desvenlafaxine, duloxetine, venlafaxine) our third-line antidepressants, given their negligible effects on weight. Our team will make recommendations to the patient's PCP, who will write prescriptions. Our team and the PCP will then collaboratively manage pharmacotherapy.
OTHERActive Control
(1) The graduate research assistant (RA) will have a 50-minute call with AC patients to review depression materials. The RA will provide a list of Eskenazi Health mental health services and will encourage patients to follow-up with their PCP. We will then send an electronic health record message to the PCP encouraging them to address their patient's depression, note that there are no care restrictions, and provide the same list of services. (2) The RA will call AC patients every 4 weeks to assess depressive symptoms and will notify clinical staff to encourage additional care when indicated. (3) AC patients will receive current primary care for depression. The Eskenazi Health primary care clinics utilize a team care approach, with PCPs supported by embedded behavioral health clinicians and affiliated psychiatrists.
Sponsors
National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)
Indiana University
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
SINGLE (Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Current primary care patient in Eskenazi Health * Age ≥18 years * Depressive disorder at screening * Prediabetes at screening
Exclusion criteria
* History of type 1 or type 2 diabetes * Major inflammatory conditions: HIV/AIDS, chronic kidney disease, systemic inflammatory disease (e.g., rheumatoid arthritis, lupus, Crohn's disease, and ulcerative colitis), or active cancer/current cancer treatment * Current pregnancy * Severe cognitive impairment * Acute risk of suicide * History of bipolar disorder or psychosis or current use of an atypical antipsychotic medication: * Participation in our prior eIMPACT Trial
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Hemoglobin A1c at 6 Months | 6 months | Fasting blood samples were collected, and whole blood and plasma aliquots were frozen. Hemoglobin A1c will be measured by a standard method. A1c is the primary outcome because: (1) it is the gold standard measure of glycemia and a common surrogate endpoint, (2) it strongly predicts future diabetes, (3) interventions decreasing A1c improve clinical diabetes endpoints, and (4) diabetes prevention interventions targeting glycemic control result in lower rates of progression from prediabetes to type 2 diabetes. Higher hemoglobin A1c values indicate greater diabetes risk. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Homeostatic Model of Assessment-Insulin Resistance (HOMA-IR) Score at 6 Months | 6 months | Higher HOMA-IR scores indicate greater insulin resistance. Homeostatic Model of Assessment-Insulin Resistance (HOMA-IR) scores were derived from fasting glucose and insulin values measured by standard assays. HOMA-IR score is an established index of insulin resistance that correlates highly with the more invasive euglycemic clamp and is appropriate for assessing change. Higher HOMA-IR scores indicate greater insulin resistance. |
| Depressive Symptoms | 6 months | Participants completed the reliable and valid Hopkins Symptom Checklist-20 (SCL-20) to assess depressive symptoms. Total scores (mean of items responses, range: 0-4) were computed, with higher scores indicating greater depressive symptoms. |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| eIMPACT-DM Intervention eIMPACT-DM is a 6-month, modernized, collaborative, stepped care intervention consisting of (1) internet and telephonic CBT for depression and (2) select antidepressant medications in an algorithm optimized for diabetes risk reduction. A multidisciplinary team delivers established depression treatments consistent with patient preference. Good Days Ahead (GDA; MindStreet, Inc.) is an empirically supported internet CBT for depression that uses an interactive, multimedia format to deliver 9 45-minute sessions, the structure and content of which mirror face-to-face CBT. GDA sessions occurred at a location with internet access selected by the patient or the PI's lab. Problem Solving Treatment in Primary Care (PST-PC) is an empirically supported CBT. During the 6-10 30-minute sessions, patients are taught skills for solving problems contributing to depression. We delivered PST-PC by phone. Regarding medications, we first considered all FDA-approved antidepressants and excluded those with weight gain effects and those rarely used in primary care. We then made bupropion and fluoxetine our first-line and second-line antidepressants, given their association with weight loss. We made other SSRIs (escitalopram, sertraline) and SNRIs (desvenlafaxine, duloxetine, venlafaxine) our third-line antidepressants, given their negligible effects on weight. Our team made recommendations to the patient's PCP, who wrote prescriptions. Our team and the PCP collaboratively managed pharmacotherapy. | 24 |
| Active Control Active Control (AC) consists of depression education (study staff), symptom monitoring (study staff), and primary care for depression (clinical staff).
Active Control: (1) The graduate research assistant (RA) will have a 50-minute call with AC patients to review depression materials. The RA will provide a list of Eskenazi Health mental health services and will encourage patients to follow-up with their PCP. We will then send an electronic health record message to the PCP encouraging them to address their patient's depression, note that there are no care restrictions, and provide the same list of services. (2) The RA will call AC patients every 4 weeks to assess depressive symptoms and will notify clinical staff to encourage additional care when indicated. (3) AC patients will receive current primary care for depression. The Eskenazi Health primary care clinics utilize a team care approach, with PCPs supported by embedded behavioral health clinicians and affiliated psychiatrists. | 22 |
| Total | 46 |
Baseline characteristics
| Characteristic | Active Control | Total | eIMPACT-DM Intervention |
|---|---|---|---|
| Age, Continuous | 50.7 years STANDARD_DEVIATION 10.4 | 49.6 years STANDARD_DEVIATION 11.1 | 48.7 years STANDARD_DEVIATION 11.8 |
| Depressive Symptoms | 1.73 Score on a scale STANDARD_DEVIATION 0.91 | 1.78 Score on a scale STANDARD_DEVIATION 0.91 | 1.82 Score on a scale STANDARD_DEVIATION 0.83 |
| Ethnicity (NIH/OMB) Hispanic or Latino | 1 Participants | 4 Participants | 3 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 21 Participants | 42 Participants | 21 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants |
| Hemoglobin A1c | 4.91 % (percentage of total hemoglobin) STANDARD_DEVIATION 0.32 | 4.94 % (percentage of total hemoglobin) STANDARD_DEVIATION 0.38 | 4.97 % (percentage of total hemoglobin) STANDARD_DEVIATION 0.44 |
| HOMA-IR score | 2.10 HOMA-IR index STANDARD_DEVIATION 0.97 | 2.58 HOMA-IR index STANDARD_DEVIATION 1.46 | 2.99 HOMA-IR index STANDARD_DEVIATION 1.69 |
| Race (NIH/OMB) American Indian or Alaska Native | 0 Participants | 1 Participants | 1 Participants |
| Race (NIH/OMB) Asian | 1 Participants | 2 Participants | 1 Participants |
| Race (NIH/OMB) Black or African American | 17 Participants | 33 Participants | 16 Participants |
| Race (NIH/OMB) More than one race | 0 Participants | 2 Participants | 2 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) White | 4 Participants | 8 Participants | 4 Participants |
| Region of Enrollment United States | 22 participants | 46 participants | 24 participants |
| Sex: Female, Male Female | 16 Participants | 36 Participants | 20 Participants |
| Sex: Female, Male Male | 6 Participants | 10 Participants | 4 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 0 / 24 | 0 / 22 |
| other Total, other adverse events | 1 / 24 | 3 / 22 |
| serious Total, serious adverse events | 0 / 24 | 0 / 22 |
Outcome results
Primary
Hemoglobin A1c at 6 Months
Fasting blood samples were collected, and whole blood and plasma aliquots were frozen. Hemoglobin A1c will be measured by a standard method. A1c is the primary outcome because: (1) it is the gold standard measure of glycemia and a common surrogate endpoint, (2) it strongly predicts future diabetes, (3) interventions decreasing A1c improve clinical diabetes endpoints, and (4) diabetes prevention interventions targeting glycemic control result in lower rates of progression from prediabetes to type 2 diabetes. Higher hemoglobin A1c values indicate greater diabetes risk.
Time frame: 6 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| eIMPACT-DM Intervention | Hemoglobin A1c at 6 Months | 4.94 % (percentage of total hemoglobin) | Standard Deviation 0.47 |
| Active Control | Hemoglobin A1c at 6 Months | 4.88 % (percentage of total hemoglobin) | Standard Deviation 0.38 |
p-value: 0.64ANCOVA
Secondary
Depressive Symptoms
Participants completed the reliable and valid Hopkins Symptom Checklist-20 (SCL-20) to assess depressive symptoms. Total scores (mean of items responses, range: 0-4) were computed, with higher scores indicating greater depressive symptoms.
Time frame: 6 months
Population: Missing data was handled using within subject mean imputation when \< 25% of the data was missing.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| eIMPACT-DM Intervention | Depressive Symptoms | 1.11 Score on a scale | Standard Deviation 0.73 |
| Active Control | Depressive Symptoms | 1.58 Score on a scale | Standard Deviation 0.89 |
p-value: 0.09ANCOVA
Secondary
Homeostatic Model of Assessment-Insulin Resistance (HOMA-IR) Score at 6 Months
Higher HOMA-IR scores indicate greater insulin resistance. Homeostatic Model of Assessment-Insulin Resistance (HOMA-IR) scores were derived from fasting glucose and insulin values measured by standard assays. HOMA-IR score is an established index of insulin resistance that correlates highly with the more invasive euglycemic clamp and is appropriate for assessing change. Higher HOMA-IR scores indicate greater insulin resistance.
Time frame: 6 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| eIMPACT-DM Intervention | Homeostatic Model of Assessment-Insulin Resistance (HOMA-IR) Score at 6 Months | 3.48 HOMA-IR index | Standard Deviation 2.16 |
| Active Control | Homeostatic Model of Assessment-Insulin Resistance (HOMA-IR) Score at 6 Months | 2.38 HOMA-IR index | Standard Deviation 0.93 |
p-value: 0.046ANCOVA