Elicit: GLP-1 and Anxiety: Clinical Insights (public)
Clinical studies evaluating the effects of GLP-1 on anxiety
Clinical studies of GLP-1 agents measured anxiety only as secondary outcomes using indirect methods rather than dedicated anxiety scales, with results showing modest changes that do not support a clear direct effect on anxiety.
Abstract
GLP-1 agent studies in patients with type 2 diabetes, overweight/obesity, or cardiovascular disease provide only indirect insights into anxiety outcomes. None of the five trials employed dedicated, validated anxiety scales. Instead, anxiety was measured as a subdomain of general quality‐of‐life instruments (including SF-36v2, DTR-QoL, EQ-5D/EQ-5D-5L, and TRIM-D) or recorded as infrequent neuropsychiatric adverse events.
- In one trial, DTR-QoL anxiety scores increased by 12.9 (SD 20.8) with an insulin degludec/liraglutide combination versus 0.4 (SD 17.2) with insulin degludec alone.
- Another trial reported an EQ-5D Visual Analog Scale improvement of +2.83 with albiglutide compared to +1.36 with placebo (p < 0.001).
- Where systematically documented, anxiety or anxiety disorder adverse events occurred in 0.1–0.8% of participants.
The studies consistently measured anxiety only as a secondary or adverse outcome, and the modest changes and low event rates do not support a clear, direct effect of GLP-1 agents on anxiety.
Methods
We analyzed 5 sources from an initial pool of 18, using 7 screening criteria. Each paper was reviewed for 6 key aspects that mattered most to the research question.
Papers identified with Elicit search
- n = 18
Papers screened using:
Human Participants
GLP-1 Intervention
Anxiety Outcomes
Study Design
GLP-1 Effect Attribution
Publication Type
Intervention Focus
n = 18
Papers screened out
- n = 13
Papers included for extraction
- n = 5
Characteristics of Included Studies
| Study Identifier | GLP-1 Agent(s) | Population | Primary Outcomes | Duration |
|---|---|---|---|---|
| NCT03015220 | Semaglutide (oral, 3/7/14 mg daily), Dulaglutide (0.75 mg weekly) | Japanese adults with type 2 diabetes (n=458), mean age 58, 74.5% male, all Asian | Glycemic control, safety | 52 weeks |
| NCT02911948 | Insulin degludec/liraglutide (fixed-ratio), Insulin degludec | Japanese adults with type 2 diabetes inadequately controlled on insulin (n=210), mean age 56, 63.3% male, all Asian | Glycemic control, safety | 26 weeks |
| NCT01644500 | Dulaglutide (0.75, 1.5 mg weekly), Glimepiride | Asian adults with type 2 diabetes (n=737), mean age 52.8, 55.4% male, all Asian | Glycemic control, safety | 26 weeks |
| NCT03574597 | Semaglutide (dose not specified, weekly) | Adults with overweight/obesity and prior cardiovascular disease (n=17,604), mean age 61.6, 72.3% male, 84% White | Major adverse cardiovascular events | 2.5–5 years |
| NCT02465515 | Albiglutide (30–50 mg weekly) | Adults with type 2 diabetes and established cardiovascular disease (n=9,463), mean age 64.1, 69.4% male, 85% White | Major adverse cardiovascular events | Median 1.65 years |
Summary
Absence of direct evidence: The available studies did not use dedicated, validated anxiety measures as primary or secondary outcomes. Anxiety-related outcomes were assessed only as subdomains of general quality of life instruments or as rare adverse events.
Magnitude and significance of effects: Where measured, changes in anxiety-related domains were small. Statistical significance was only reported in one study, and only for general psychological health domains, not for anxiety specifically.
Adverse events: Neuropsychiatric adverse events, including anxiety, were rare or very rare in studies that systematically reported them. Most studies did not systematically report on anxiety-specific adverse events.
Limitations of the evidence base: The evidence is limited by the lack of dedicated anxiety assessment and the indirect nature of available data. The findings do not provide clear evidence for or against an effect of GLP-1 agents on anxiety.