Weight is an established risk factor for asthma, and the relationship is graded. A meta-analysis of seven prospective studies covering 333,102 adults found overweight and obesity together raised the odds of incident asthma by 51% (OR 1.51; 95% CI, 1.27 to 1.80), with a dose-response running from OR 1.38 for overweight to 1.92 for obesity (95% CI, 1.43 to 2.59; p < 0.0001 for trend). The increase was similar in men (OR 1.46) and women (OR 1.68), p = 0.232 for the comparison.[1]
A larger and more recent synthesis of 16 cohort studies — 63,952 cases among 1,161,169 participants — put numbers on each axis: relative risk 1.32 per 5 kg/m² increase in body-mass index (95% CI, 1.21 to 1.44), 1.26 per 10 cm of waist circumference and 1.33 per 10 kg of weight gained. All three showed a clear dose-response, with heterogeneity around 90% for the first two.[2]
So the premise behind pointing a weight-loss drug at an airway is sound. Whether the drug does it is a separate and much thinner question, and the drug this market dispenses is covered in the compounded-versus-brand article rather than in any of the studies below.
What losing weight does to an asthmatic airway
A 2025 systematic review pooled 12 randomized trials and 1,052 participants testing weight loss by any means in people with asthma. Asthma control improved — weighted mean difference −0.67 on the Asthma Control Questionnaire (95% CI, −1.20 to −0.13; P = 0.02) — as did forced expiratory volume in one second as a percentage of predicted, at +13.08 points (95% CI, 2.42 to 23.73; P = 0.02). Quality of life did not improve significantly (0.53; −0.38 to 1.44; P = 0.25), nor did forced vital capacity. Heterogeneity ran above 91% on every outcome.[3]
A confidence interval on lung function running from 2 to 24 points is not a measurement; it is twelve trials disagreeing. What survives is the direction.
The observational record on this drug class
The foundational studies are electronic-health-record cohorts with an active-comparator, new-user design. In adults with type 2 diabetes and asthma, exacerbation counts at six months were lower on GLP-1 receptor agonists (n = 448) than on SGLT-2 inhibitors (IRR 2.98; 95% CI, 1.30 to 6.80), DPP-4 inhibitors (2.45; 1.54 to 3.89), sulfonylureas (1.83; 1.20 to 2.77) or basal insulin (2.58; 1.72 to 3.88), with the GLP-1 arm as the reference.[4] In 1,642 patients with COPD, adjusted exacerbation rates were significantly higher on DPP-4 inhibitors (IRR 1.48; 1.08 to 2.04) and sulfonylureas (2.09; 1.62 to 2.69) than on GLP-1 receptor agonists — and no significant difference appeared against SGLT-2 inhibitors.[5]
A 2025 systematic review pooled six retrospective studies covering 62,678 adults and laid the comparator ladder out in full. Against sulfonylureas the incidence rate ratio was 0.52 (95% CI, 0.42 to 0.64); against DPP-4 inhibitors 0.63 (0.47 to 0.86); against insulin 0.39 (0.26 to 0.58) in a single study. Against SGLT-2 inhibitors it was 0.66 with a confidence interval of 0.21 to 2.05, which crosses one. And against metformin, in a single Japanese cohort, exacerbations were higher on the GLP-1 receptor agonist: IRR 1.24 (1.08 to 1.43).[6]
That is the finding a favorable summary leaves out. The airway advantage is measured against the diabetes drugs that were already the worse choice. Against the two cheapest and oldest alternatives it either vanishes or reverses.
A comparative effectiveness study emulating three target trials across three US claims databases confirms the top of the ladder in COPD. Among 32,107 matched pairs, GLP-1 receptor agonists beat DPP-4 inhibitors on moderate or severe exacerbations (9.89 against 11.49 per 100 person-years; HR 0.86; 95% CI, 0.81 to 0.91). Among 36,218 pairs, SGLT-2 inhibitors against GLP-1 receptor agonists produced minimal difference — 9.47 against 10.00 per 100 person-years; HR 0.94; 95% CI, 0.89 to 1.00.[7]
The result that argues against the weight explanation
A 2025 study using UK primary-care records linked to hospital admissions triangulated two designs — a self-controlled case series in 4,278 patients and a new-user cohort with inverse probability weighting in 8,424 — drawn from more than two million adults with asthma. Metformin itself was associated with fewer asthma attacks in both (IRR 0.68; 95% CI, 0.62 to 0.75, and HR 0.76; 0.67 to 0.85). Among the add-on drugs, a GLP-1 receptor agonist was the only one with an additive association: IRR 0.60 (95% CI, 0.49 to 0.73). Negative control analyses found no evidence of significant bias.[8]
The part that complicates every summary of this literature is what did not modify the association: HbA1c, body-mass index, blood eosinophil count and asthma severity all failed to change it. The authors conclude the mechanism appears to be something other than glycemic control or weight loss, operating across asthma phenotypes.[8] If that is right, then the reason to expect an airway effect is not the reason this drug is being sold, and the weight-loss trials in the side-effect literature have nothing to say about it either way.
Who the effect appears in, and where it stops
A causal-forest analysis of commercial claims examined 10,989 patients starting a GLP-1 receptor agonist or a thiazolidinedione and 17,088 starting a GLP-1 receptor agonist or a sulfonylurea. The adjusted risk difference for asthma exacerbation within 180 days was −1.6% against sulfonylureas (95% CI, −2.2% to −1.1%) but −0.5% against thiazolidinediones, with a confidence interval of −1.1% to 0.1% that does not exclude zero. Within the sulfonylurea comparison, the subgroup with two or more emergency department visits in the prior year gained most, at −2.8% (−4.8% to −0.9%).[9]
A national-insurance cohort looked at asthma onset rather than exacerbation, following 1,345 GLP-1-treated patients drawn from 1,936,512 adults with type 2 diabetes and no pre-existing asthma over a mean 2.92 years. The adjusted hazard ratio for developing asthma was 0.67 (95% CI, 0.45 to 0.76), and a gradient ran across severity strata: 0.55 for asthma without acute exacerbations, 0.59 with exacerbations, 0.83 for status asthmaticus — and 0.96 (95% CI, 0.65 to 1.09) for cases requiring endotracheal intubation.[10]
The gradient does not merely weaken at the severe end. It disappears. The association is strongest for the mildest disease and absent for the episodes that put someone on a ventilator.
The one study in the population that buys this drug
Every figure above comes from people with type 2 diabetes, which most cash-pay buyers do not have. One retrospective cohort addresses the actual population. Using a US research network spanning 72 healthcare organizations, adults with a body-mass index of 30 or above and repeated asthma documentation who started GLP-1-based therapy were matched 1:1 against patients started on non-GLP weight-management pharmacotherapy, leaving 2,423 per arm. Over one year, recorded asthma exacerbation occurred in 1.7% against 4.0% (risk difference −2.2%; 95% CI, −3.2% to −1.3%; risk ratio 0.438; 0.306 to 0.626). Systemic glucocorticoid exposure ran 16.4% against 23.1%, emergency or critical-care use 7.1% against 13.0%, and acute respiratory failure 0.7% against 1.8%.[11]
Its authors state plainly that residual confounding, uncertainty about treatment persistence and exposure misclassification preclude causal inference, and that drug-specific adverse events were not systematically assessed.[11] A person who can afford a GLP-1 program and stays on it for a year differs from a person who cannot in ways no propensity score captures.
The signal pointing the other way
An analysis of the FDA Adverse Event Reporting System asked the opposite question and found something. Certain drugs in this class — exenatide, semaglutide and liraglutide — were associated with a higher proportion of respiratory adverse events, particularly asthma or asthma-like events, and the association reached statistical significance at least for semaglutide and liraglutide. Serious asthma-related events and deaths were also reported, with exenatide carrying the highest proportion of deaths.[12]
A disproportionality analysis measures what gets reported, not what happens, and the authors name pharmacological properties, patient characteristics and reporting bias as unresolved explanations. It is a hypothesis. It is also the only published work in this area asking whether the arrow points down rather than up, and it belongs in any honest summary of a literature otherwise made entirely of favorable retrospective cohorts.
How big the evidence base actually is
As of 15 September 2026, PubMed holds two records typed as randomized controlled trials carrying asthma, COPD or chronic obstructive pulmonary disease together with any of GLP-1, semaglutide, liraglutide, tirzepatide, dulaglutide or exenatide — and both come from the same 40-person Danish study. The identical query shape holds 263 records for the four asthma biologics and 193 for obesity trials of semaglutide or tirzepatide. The near-absence is real.
That trial randomized 40 people with obesity and COPD to liraglutide 3.0 mg or placebo for 40 weeks. Against placebo, liraglutide produced significant weight loss, increased forced vital capacity and carbon monoxide diffusing capacity, and improved the COPD Assessment Test score. It produced no significant change in FEV₁, in the FEV₁/FVC ratio, or in six-minute walking distance.[13]
FEV₁ is the measurement that defines airflow obstruction and grades COPD severity. The only randomized test this class has faced in an obstructive airway disease did not move it, in forty people, over forty weeks. That is the whole randomized record.
What is approved, and what to do with the rest
One airway indication does exist. Tirzepatide is approved to treat moderate to severe obstructive sleep apnea in adults with obesity, on the strength of two 52-week randomized trials in 469 patients; the label specifies a 10 mg or 15 mg maintenance dose for that indication.[14] Sleep apnea is an upper-airway disorder and not an obstructive lung disease, and those results are set out in the sleep apnea article. Nothing in this class is approved for asthma or COPD anywhere.
For someone with asthma or COPD considering a GLP-1 for weight, the practical reading is narrow. There is no reason from this literature to expect the injection to replace an inhaler, no randomized evidence that it improves airflow obstruction, and no reason to change a controller regimen. Weight loss itself has randomized evidence behind it for asthma control, which is the effect most likely to be doing the work. Smoking remains the dominant modifiable variable in COPD, and what this class does to that is in the smoking-cessation article. Anyone with airway disease facing sedation should also read the anesthesia article, where the drug’s effect on gastric emptying carries a real procedural consequence.
Every trial and cohort above studied FDA-approved product. Most sellers reviewed here dispense compounded tirzepatide or compounded semaglutide, which are not FDA-approved and are not reviewed by the FDA for safety, efficacy or quality before they are dispensed, and no study has measured a respiratory outcome on one.