Asthma Inhalers: Reliever Risk, Montelukast Warnings and Funding

Table of contents

• Evidence summary
• What the asthma medicines are
• How they work
• What works and what does not
• Benefits by claim
• Risks and all side effects
• Interactions and who should take care
• Doses studied, devices and carbon footprint
• Follow the money: who makes them and who funded the evidence
• Related research
• Frequently asked questions
• Sources and funding notes

Evidence summary

Independent evidence and credibility scorecard

Independence tiers: 1 = regulator or government body; 2 = academic or public-funded team with no declared commercial funding for the work (or with a funding gap we flag); 3 = independent authors analysing mostly manufacturer-run trials, or minor declared industry ties; 4 = manufacturer-funded or manufacturer-authored. Credibility runs from A (highest) to D.

Note on clustering: nearly all the mild-asthma trial and follow-up evidence comes from the same New Zealand and Australian research group, and most of its industry ties run to one manufacturer. The regulator sources are UK and US. The evidence is therefore concentrated in one research network and one company's product.

What the asthma medicines are

Asthma medicines fall into a few groups. Short-acting beta-agonists (SABA), such as salbutamol (albuterol in the US), relieve symptoms quickly. Inhaled corticosteroids (ICS) reduce airway inflammation and are the core preventer. Long-acting beta-agonists (LABA) are usually combined with an ICS. Formoterol is unusual because it is both fast and long-acting, which is why it can be used in as-needed combination products. Leukotriene receptor antagonists, such as montelukast, are tablets taken daily. Biologics such as omalizumab, mepolizumab and benralizumab are injections for severe disease.

The Novel START and PRACTICAL trials tested as-needed budesonide-formoterol, which is an inhaled steroid plus a fast-acting long-acting beta-agonist in one device. The drug used in Novel START and the follow-up analyses is AstraZeneca's Symbicort. We did not open the full trial protocol to confirm the device supply arrangement in every trial.

Forms that matter for safety

• Reliever only: A SABA used without a preventer is the pattern the 1994 cohort and later work link to asthma death (Suissa et al. 1994).
• Combined reliever: An ICS plus formoterol taken as needed gives both anti-inflammatory and fast relief in one inhaler (Beasley et al. 2019).
• Metered-dose vs dry-powder devices: The two types differ in propellant and carbon footprint, not in the medicine itself (Wilkinson et al. 2019).

Current UK and international guidelines recommend an as-needed ICS-formoterol reliever for many adults. We could not open the NICE asthma guideline text (NG245) for this review, and the statements we rely on come from secondary summaries that we have flagged in the gaps section.

How they work

Salbutamol relaxes airway muscle within minutes, but it does not reduce inflammation. The 1994 cohort suggested that heavy reliever use marked patients with the most severe, poorly controlled disease, which is one reason the association is hard to read as cause (Suissa et al. 1994).

ICS reduce airway inflammation over days to weeks. In the 2000 cohort, each additional canister in the prior year was linked with fewer asthma deaths. That is an association in a population, not a trial of dose (Suissa et al. 2000).

Formoterol starts working quickly and lasts long. When combined with an ICS, it can be taken as needed. The mild-asthma trials test whether this reduces exacerbations when the steroid is delivered at the point of symptoms, rather than every day (Hardy et al. 2019).

What works and what does not

Benefits by claim

Does reliever-only treatment carry a death risk?

The clearest signal is the 1994 cohort of asthma drug users. Asthma death risk rose sharply once monthly inhaled beta-agonist use passed about 1.4 canisters, which the authors describe as the recommended limit (Suissa et al. 1994). In plain terms, heavy reliever use is a marker of poor control. The study cannot separate the drug's effect from the disease that drives heavy use. Alternative explanations, such as under-use of inhaled steroids, were not examined in this review.

Do inhaled steroids reduce asthma death?

In the 2000 population cohort, each additional canister of ICS used in the prior year was linked with a 21% lower rate of asthma death (adjusted rate ratio 0.79, 95% CI 0.65 to 0.97) (Suissa et al. 2000). The interval sits close to 1, so the benefit is real but modest in the data. It is an observational association, and the paper does not record a funder.

Does as-needed ICS-formoterol beat as-needed salbutamol?

In Novel START, adults with mild asthma who used budesonide-formoterol as needed had an annual exacerbation rate of 0.195, against 0.400 for as-needed albuterol. The relative rate was 0.49 (95% CI 0.33 to 0.72). Severe exacerbations were 9 with budesonide-formoterol and 23 with albuterol (RR 0.40, 95% CI 0.18 to 0.86). Against daily budesonide, the result was not different (rate 0.175; RR 1.12, 95% CI 0.70 to 1.79) (Beasley et al. 2019). The trial was open-label, so patients and clinicians knew which treatment was given.

The PRACTICAL trial, run in New Zealand and funded by the public Health Research Council, compared as-needed budesonide-formoterol with maintenance budesonide plus as-needed terbutaline in mild to moderate asthma. Severe exacerbations were 0.119 vs 0.172 per patient-year (RR 0.69, 95% CI 0.48 to 1.00; p=0.049) (Hardy et al. 2019). The confidence interval reaches 1.00, so the result is borderline. Two trials with different funders point the same way, which is the strongest convergence in this review. Both are open-label, and the maker co-funded only one of them.

Is salmeterol without a steroid safe?

The SMART trial was a 28-week randomised, double-blind, placebo-controlled study in people with asthma (Nelson et al. 2006). It was terminated after an interim analysis of 26,355 subjects, due to findings in African Americans and difficulties with enrollment. Its primary outcome (respiratory-related deaths or life-threatening experiences) occurred 50 times on salmeterol and 36 times on placebo, RR 1.40 (95% CI 0.91 to 2.14). That is not statistically significant, so it is a weak safety signal, not proof of harm. We could not open the sponsor statement.

Does montelukast help?

Montelukast is a tablet, not an inhaler, and it is used as an add-on for some patients. Its benefits are real but smaller than those of inhaled steroids in most guidelines. The safety issue is the bigger story. The UK regulator's 2019 notice reported 219 suspected neuropsychiatric reactions from 2014 to 2018 against about 14 million prescriptions, and said to weigh risks and benefits if reactions occur (MHRA 2019).

Do biologic injections help severe asthma?

Biologics such as omalizumab, mepolizumab and benralizumab are for severe disease. NICE has recommended them for adults who meet eosinophil-count and exacerbation criteria despite high-dose inhaled steroid and LABA treatment. We have not verified the appraisal text here: the NICE pages were blocked to us, and the summaries we saw name TA278 (omalizumab), TA671 (mepolizumab) and TA565 (benralizumab). NICE appraisals of these drugs are built on company submissions. We could not open those documents, so we do not state who made them.

Risks and all side effects

How common is the risk?

Nobody knows the true rate. Yellow Card reports capture only a fraction of events, and the same pattern of report counts does not show how many people were affected. The regulator's own figures give a reporting rate, not an incidence (MHRA 2024).

Reliever overuse and asthma death

A rising need for reliever inhalers is the warning sign. The 1994 cohort linked heavy beta-agonist use with asthma death, and the risk rose beyond about 1.4 canisters a month (Suissa et al. 1994). Anyone whose reliever use is climbing should talk to their prescriber about review, rather than adjusting the dose alone.

Inhaled steroid effects

We did not verify the side-effect profile of inhaled steroids (for example oral thrush or hoarseness with some devices) against a primary source in this review, so we do not give figures. A review of the labels and guideline text would be needed before making claims on that point.

Interactions and who should take care

We did not verify drug-interaction data for these products against a primary source in this review, so this section lists only what we confirmed. The MHRA asks prescribers to be alert for neuropsychiatric reactions in adults, adolescents and children (MHRA 2019). Pregnancy, older adults and people with other lung or heart disease need individual assessment, and we could not verify population-specific guidance against a primary source.

Doses studied, devices and carbon footprint

The trials used as-needed budesonide-formoterol, with the dose and device set by the trial protocol. Novel START and PRACTICAL were open-label and used the reliever on symptoms rather than a fixed daily schedule (Beasley et al. 2019; Hardy et al. 2019). We do not give dosing instructions. Changes to a treatment plan should be made with a prescriber.

Carbon footprint is a separate question. Metered-dose inhalers use a propellant with a high global warming potential, while dry-powder devices do not. A 2019 costing study modelled switching NHS prescriptions and found the cost and carbon effects of different switching scenarios (Wilkinson et al. 2019). Press coverage of that study quoted large carbon ratios between device types. We could not open the full text to check those ratios, so we give no figures. A 2024 post hoc analysis estimated the carbon footprint of the as-needed budesonide-formoterol strategy in mild asthma (Hatter et al. 2024), and a 2025 carbon-utility analysis compared three mild-asthma strategies (Vieira et al. 2025). Both are modelling or post hoc work, and the first is funded by the company whose product it studies. We did not open either full text to check its figures. Patients should not switch devices on their own; a change needs prescriber review.

Product quality and contamination issues were not reviewed in this article.

Follow the money: who makes it and who funded the evidence

Who makes and sells the products

AstraZeneca makes Symbicort (budesonide-formoterol), the product tested in Novel START and the follow-up analyses. The 2019 Novel START paper names AstraZeneca as a funder (Beasley et al. 2019). GSK is named in the 2011 US settlement over Advair (Nature blog, 2011) and, per a secondary summary, in the SMART trial. We could not verify the current owner of montelukast (Singulair) or its generic status in this review.

We did not open the company annual reports, so market share, prices and current owners of each device are not stated here.

Who funded the trials

The mild-asthma trials split into two groups. Novel START was funded by AstraZeneca and New Zealand's Health Research Council (Beasley et al. 2019). PRACTICAL was funded by the Health Research Council alone, according to its PubMed record (Hardy et al. 2019). The 2022, 2024 and 2026 analyses of the same programme declare AstraZeneca support or employment (Hatter 2022; Hatter 2024; Leung 2026). Most of the positive modern evidence therefore comes from one research network with close ties to one company.

Documented integrity events

In November 2011 GSK agreed to pay $3 billion to settle several US federal investigations. The Advair allegation was that GSK promoted the asthma drug for uses the government had not approved, known as off-label promotion (Nature blog, 2011). The settlement covered several drugs, so the Advair share is not separately stated here. We could not open the US Department of Justice release to confirm the detail.

The UK regulator's 2024 montelukast review led to new warnings. It did not withdraw the drug (MHRA 2024).

Who backs the research institutions

The Health Research Council of New Zealand is a public funder. Its funding of PRACTICAL and part of Novel START is documented in the PubMed records above. Several authors of the later analyses declare AstraZeneca or GSK payments, which the conflict statements in the 2026 cost-effectiveness paper set out (Leung 2026).

Related research

• Asthma Prevention and Management: Evidence and Safety Guide
• Asthma Supplements: Nutritional Roles, Evidence Limits and Safety
• Chronic Rhinosinusitis: Adult Care and Adjunct Evidence
• Eosinophilic granulomatosis with polyangiitis: EGPA symptoms, asthma, ANCA and treatment
• Eczema Prevention and Management: Evidence and Safety Guide
• Eczema Supplements: Evidence, Uncertainty and Safety

Frequently asked questions

Sources and funding notes

What we could not verify:

• NICE pages (NG245 asthma guideline; TA278, TA565 and TA671 appraisals) returned HTTP 403 and could not be opened. Statements about them come from search summaries and are not linked.
• The GINA 2025 report and the FDA labelling documents for ICS-LABA and montelukast were not opened (404 or 403). The FDA and GINA positions are not stated as verified facts here.
• The SMART trial's sponsor is not confirmed. The Chest abstract gives the numbers (26,355 subjects at interim analysis; RR 1.40) but names no funder, and GSK is named only in a secondary summary.
• The full text of the 2019 carbon costing study and the 1994 cohort results were not opened in full; we used the PubMed records and abstract summaries only.
• The identity of the trial device (AstraZeneca's Symbicort) comes from secondary trade-press summaries, not from the trial paper.
• Current prices, device market shares and the owner of Singulair were not verified.
• Interaction, pregnancy, paediatric and inhaler-technique guidance were not reviewed against a primary source.
• The MHRA 2019 wording that prescribers should weigh risks and benefits if reactions occur (montelukast section and FAQ) was not re-checked in this pass.
• PRACTICAL: the PubMed abstract, checked in this re-check, confirms 0.119 vs 0.172 per patient-year (RR 0.69, 95% CI 0.48 to 1.00) and the Health Research Council of New Zealand funding statement. The full Lancet text and the drug-supply arrangement were not opened.
• Novel START: the PubMed abstract, checked in this re-check, confirms the rates, severe-exacerbation counts (9 vs 23) and the daily-budesonide comparison (RR 1.12). The full NEJM text was not opened.
• Suissa 1994: the abstract puts the change point at about 1.4 canisters a month; a later review cites 1.7. The full paper was not opened.
• The 2024 MHRA review: one gov.uk page in search results says the expert advisory group recommended withdrawal. This conflicts with the published outcome and was not resolved.

1. MHRA Drug Safety Update, montelukast reminder, 19 September 2019 — UK regulator; government.
2. MHRA press release, outcome of review of neuropsychiatric reactions with montelukast, 2024 (exact day not confirmed in this check) — UK regulator; government.
3. Beasley R et al., Controlled Trial of Budesonide-Formoterol as Needed for Mild Asthma, NEJM 2019 — Novel START; AstraZeneca and Health Research Council of New Zealand.
4. Hardy J et al., Budesonide-formoterol reliever therapy versus maintenance budesonide plus terbutaline reliever therapy in adults, Lancet 2019 (PRACTICAL) — Health Research Council of New Zealand.
5. Suissa S et al., A cohort analysis of excess mortality in asthma and the use of inhaled beta-agonists, AJRCCM 1994 — academic; funder not in record.
6. Suissa S et al., Low-dose inhaled corticosteroids and the prevention of death from asthma, NEJM 2000 — academic; funder not in record.
7. Hatter L et al., Asthma control with ICS-formoterol reliever versus maintenance ICS and SABA reliever therapy, BMJ Open Respiratory Research 2022 — post hoc; AstraZeneca support declared.
8. Hatter L et al., The carbon footprint of as-needed budesonide/formoterol in mild asthma, ERJ 2024 — post hoc; AstraZeneca funds the manuscript.
9. Leung W et al., Cost-effectiveness of as-needed budesonide-formoterol in adults with mild asthma: the Novel START trial, ERJ Open Research 2026 — AstraZeneca employee among authors; AstraZeneca and GSK payments declared.
10. Vieira RJ et al., Comparison of Three Treatment Strategies in Mild Asthma: A Carbon-Utility Analysis, JACI In Practice 2025 — no funding statement in the record.
11. Wilkinson AJK et al., Costs of switching to low global warming potential inhalers, BMJ Open 2019 — academic and NHS authors; NHS England grant declared; no industry funding in the record.
12. Nelson HS et al., The Salmeterol Multicenter Asthma Research Trial, Chest 2006 — trial record names no funder; GSK named in a secondary summary.
13. Nature blog, GSK to settle federal investigations for $3 billion, November 2011 — secondary news report.

Educational research reviewed 9 October 2026. This article is not medical advice; talk to a qualified clinician or pharmacist before starting, stopping or combining treatments.