Nasal Insulin: Effectiveness and Clinical Evidence
Based on the highest quality evidence, intranasal insulin is NOT effective for treating cognitive impairment or Alzheimer's disease and should not be used for this purpose in clinical practice. The most recent and rigorous phase 2/3 multisite randomized controlled trial showed no cognitive, functional, or biomarker benefits 1.
Glucose Regulation
Intranasal insulin is not approved or indicated for glucose regulation in diabetes. The FDA-approved insulin formulations (Humulin R U-100) are designed exclusively for subcutaneous or intravenous administration 2. While early research from the 1980s demonstrated that nasal insulin with surfactants could raise serum insulin levels and lower blood glucose 3, this route has never been developed into an approved therapy for diabetes management. The bioavailability remains poor, and systemic spill-over is minimal 4.
Cognitive Function: The Evidence Contradicts Earlier Promise
The Definitive Negative Trial
The 2020 JAMA Neurology trial 1 represents the gold standard evidence:
- 289 participants with mild cognitive impairment (MCI) or Alzheimer's disease
- 12-month treatment with 40 IU daily intranasal insulin
- Primary outcome (ADAS-cog-12): No difference between insulin and placebo (0.0258 points; 95% CI: -1.771 to 1.822; P = .98)
- No benefits on any secondary clinical, functional, or cerebrospinal fluid biomarkers
- Device reliability issues affected the first 49 participants, but even the primary cohort (n=240) with the reliable device showed null results
The Most Recent Meta-Analysis Confirms Ineffectiveness
A 2026 systematic review and meta-analysis 5 pooled five RCTs (540 participants) and found:
- No significant cognitive improvement: ADAS-Cog13 mean difference -1.09 (95% CI: -4.89 to 2.71)
- No functional benefit: ADCS-ADL mean difference 0.06 (-0.33 to 0.45)
- No biomarker effects: No changes in CSF Aβ42, total tau, or phosphorylated tau
- Adverse events: More gastrointestinal side effects (RR: 1.57; 95% CI: 1.15-2.14)
- Conclusion: "Current evidence does not support its routine clinical use"
Why Earlier Studies Showed Promise
Earlier smaller trials from 2012 6 and 2015 7 suggested benefits, but these findings have not been replicated:
2012 pilot trial (n=104,4 months):
- Showed improved delayed memory and preserved function
- Critical limitation: Much smaller sample size and shorter duration than the definitive 2020 trial
2015 insulin detemir trial (n=60,21 days):
- Reported memory improvement in APOE-ε4 carriers with 40 IU dose
- Critical limitations: Very short treatment duration, small sample, and paradoxical worsening in non-carriers
Recent Exploratory Data (Not Practice-Changing)
A 2025 phase 2A/B trial 8 (4 weeks only) suggested potential benefits on cognition and biomarkers, but this was:
- Too short to establish clinical efficacy
- Explicitly stated that "a longer and larger trial is needed to confirm these results"
- Contradicted by the longer, larger 2020 trial
A 2024 meta-analysis 9 reported positive effects on delayed recall (effect size: 1.37) and overall cognition (effect size: 0.58), but this analysis:
- Included the same small, short-duration studies that were contradicted by the 2020 definitive trial
- Did not adequately weight the negative findings from the largest, longest trial
Clinical Bottom Line
Do not prescribe intranasal insulin for cognitive impairment or Alzheimer's disease. The phase 2/3 multisite trial—the highest quality evidence available—definitively showed no benefit over 12 months 1, and this was confirmed by the 2026 meta-analysis 5.
Do not use intranasal insulin for diabetes management. It is not FDA-approved for this indication, and approved insulin formulations are designed for subcutaneous or intravenous routes only 2.
What Guidelines Say About Cognitive Dysfunction in Diabetes
Current guidelines 10 emphasize:
- Active screening for cognitive impairment in older adults with diabetes
- Tailoring diabetes therapy to prevent hypoglycemic episodes in those over 60-65 years
- No evidence that intensified glycemic control preserves cognitive function
- Recognition that cognitive impairment increases risk of treatment-related complications
The focus should remain on established diabetes management strategies and prevention of acute metabolic complications, not experimental intranasal insulin therapy.