Tirzepatide Users Had Fewer Anemia Diagnoses Than GLP-1 Drug Users in a Large Cohort
In 135,216 matched adults with type 2 diabetes and obesity, tirzepatide was linked to a modestly lower hazard of nutritional deficiency anemia, mostly iron deficiency.
Summary
Weight-loss drugs can change how much people eat, which raises a fair question: do they raise the risk of nutrient-related anemia? Researchers used a large U.S.-based electronic health records network to compare adults with type 2 diabetes and overweight or obesity who started tirzepatide with those who started a GLP-1 receptor agonist such as semaglutide or liraglutide. After matching, each group had 67,608 people. Over one year, 2.13% of tirzepatide users and 2.84% of GLP-1 users were diagnosed with nutritional deficiency anemia, a modestly lower hazard with tirzepatide (HR 0.88). The difference was driven by iron deficiency anemia. Vitamin B12 and folate results were inconclusive. This is an observational association, not proof that tirzepatide protects against anemia, and the authors note residual confounding and differences in follow-up and diagnosis. Notably, neither drug class showed a clear signal of harm here.
Detailed Summary
Incretin-based drugs are now widely used for diabetes and weight loss, and they reduce appetite and food intake. That raises a practical safety question for long-term users and their physicians: could eating less, or absorbing nutrients differently, lead to deficiencies such as iron, vitamin B12 or folate shortfalls and the anemia that follows? Anemia matters for aging because it is tied to fatigue, reduced physical capacity and poorer functional outcomes.
This study was a retrospective cohort analysis using the TriNetX database. It included adults with type 2 diabetes and overweight or obesity who started tirzepatide or a GLP-1 receptor agonist between January 2022 and August 2026. People with prior use of either drug, prior nutritional deficiency anemia, type 1 diabetes or bariatric surgery were excluded. After 1:1 propensity score matching, each arm had 67,608 patients, followed from day 1 through day 365. The primary outcome was a composite of iron, B12 or folate deficiency anemia.
Nutritional deficiency anemia was diagnosed in 1,440 tirzepatide users (2.13%) and 1,917 GLP-1 users (2.84%), giving a hazard ratio of 0.88 (95% CI 0.83-0.95). Iron deficiency anemia was lower with tirzepatide (HR 0.87; 0.81-0.93). Estimates for B12 deficiency anemia (HR 1.07; 0.83-1.38) and folate deficiency anemia (HR 0.98; 0.56-1.71) were inconclusive, with wide intervals that include both benefit and harm.
The finding is reassuring in that tirzepatide did not appear to carry more anemia risk than the older drug class, and may carry slightly less. The abstract offers no mechanism, so any explanation, such as differences in weight loss, glycemic control or inflammation, would be speculation. Absolute differences were small, under one percentage point.
Caveats are important. This is observational, so association is not causation. The authors acknowledge residual confounding and differences in follow-up and ascertainment, since only diagnosed anemia was counted. No hemoglobin or nutrient levels were reported in the abstract, and this summary draws on the abstract alone.
Key Findings
- Matched cohorts of 67,608 adults each; one-year nutritional deficiency anemia occurred in 2.13% on tirzepatide versus 2.84% on GLP-1 receptor agonists.
- Composite anemia hazard ratio was 0.88 (95% CI 0.83-0.95), a modest association favoring tirzepatide.
- Iron deficiency anemia drove the difference (HR 0.87; 0.81-0.93).
- Vitamin B12 (HR 1.07) and folate (HR 0.98) deficiency anemia estimates were inconclusive, with wide confidence intervals.
- Authors state residual confounding and differences in follow-up and ascertainment limit causal interpretation; no mechanism was tested.
Methodology
Retrospective cohort study in the TriNetX electronic health records network, comparing new users of tirzepatide with new users of GLP-1 receptor agonists among adults with type 2 diabetes and overweight or obesity. After 1:1 propensity score matching (67,608 per arm), diagnosed iron, B12 or folate deficiency anemia was tracked from day 1 to day 365. Endpoints were diagnosis-based, not laboratory-measured, in the information available.
Study Limitations
This summary is based on the abstract only. Observational design with residual confounding, differences in follow-up and diagnostic ascertainment means causality cannot be inferred. B12 and folate results were inconclusive, outcomes were diagnoses rather than measured hemoglobin or nutrient levels, and one-year follow-up says nothing about longer-term risk.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe:
