Chronic Cough: Independent Evidence on Causes, Drugs & Funding

by
Key takeaways
  • Chronic cough lasts eight weeks or more. The European Respiratory Society (ERS) guideline names several recognised causes, including asthma, reflux, ACE inhibitors and post-infectious cough, and gives recommendations for each (ERS guidelines, Eur Respir J 2020). The same guideline puts chronic cough at about 15% of people taking an ACE inhibitor.
  • For refractory or unexplained cough, the strongest drug evidence comes from gefapixant. In two phase 3 trials of 2,044 people, the 45 mg twice-daily dose cut 24-hour cough frequency by 18.5% at 12 weeks (COUGH-1) and 14.6% at 24 weeks (COUGH-2), against placebo (McGarvey et al., Lancet 2022). Both trials were sponsored by Merck Sharp & Dohme (ClinicalTrials.gov NCT03449147).
  • Gabapentin has one small placebo-controlled trial: 62 adults randomised, 10 weeks, with a cough quality-of-life gain of 1.80 points (95% CI 0.56 to 3.04) over placebo (Ryan et al., Lancet 2012). The ERS guideline suggests a trial of gabapentin or pregabalin, but rates the evidence as low quality.
  • Speech and language therapy has two small randomised trials (162 adults) in the 2019 Cochrane review. The quality-of-life result from one trial was small and at low certainty, and it was not seen at three months (Cochrane review CD013067).
  • Regulators disagree on gefapixant. The European Medicines Agency authorised it as Lyfnua in September 2023 (EMA). The US Food and Drug Administration rejected it in December 2023 because the application did not show evidence of effectiveness, not because of safety (trade report of the FDA letter).
  • Money trail: Gefapixant is sponsored by Merck Sharp & Dohme, a subsidiary of Merck & Co. Eight authors of the Lancet paper are Merck employees who own stock, and several others report Merck grants or fees (Lancet 2022 declarations). The 2012 gabapentin trial was funded by Australia's National Health and Medical Research Council and the Hunter Medical Research Institute, according to the trial abstract. HMRI separately reports support for Dr Ryan's research from the Greaves Family (HMRI).
  • Evidence grade: Moderate that the ERS-recommended causes (asthma, reflux, ACE inhibitors) need checking first. Weak for gabapentin. Weak for speech therapy. Moderate for a modest drop in cough frequency with gefapixant, with Risk of taste change.

Independent evidence review · Prescription medicine and therapy

Most chronic cough has a cause that can be found and treated, and that is the first step. For cough that stays unexplained after tests, the independent evidence is thin. Gabapentin and speech therapy each rest on small, low-certainty trials. Gefapixant lowers cough frequency modestly in two large trials sponsored by its maker, and regulators have split on whether that benefit is enough. Our confidence: moderate that treating known causes comes first; low for gabapentin and speech therapy; moderate that gefapixant reduces cough frequency, but low that the benefit matters enough to justify its taste side effects for most people.

Best evidence for treating identified causes, and gefapixant 45 mg twice daily in two large maker-sponsored trials
Main risks taste change on gefapixant; side effects on gabapentin (nausea and tiredness in the trial); cough caused by ACE inhibitors
Key rule do not start, stop or change a prescribed cough treatment on the basis of this article; tell your prescriber about a new or persistent cough, especially if you take an ACE inhibitor
Get medical help promptly
Seek urgent care if a cough comes with coughing up blood, chest pain, breathlessness at rest, a high fever, weight loss you cannot explain, or a cough that wakes you from sleep with wheezing. Persistent cough in a smoker, or a cough that is changing, also needs a clinician's review. Do not stop an inhaler, blood-pressure tablet or other prescribed medicine without advice. If you are in crisis or thinking about harming yourself, contact emergency services now. In the US, call or text the 988 Suicide & Crisis Lifeline. In the UK, call Samaritans on 116 123. Elsewhere, use your local emergency number.

Table of contents

Evidence summary

Claim Evidence Source Funding / conflict Strength
Known causes should be treated first Guideline recommendations for asthma, reflux, ACE inhibitors and post-infectious cough, mostly conditional and low-quality evidence ERS guideline Funding statement not found in the text we read; the guideline states it was written with editorial independence from the ERS Moderate
Gabapentin reduces cough-specific quality of life 62 randomised; LCQ difference 1.80 (95% CI 0.56 to 3.04) at 10 weeks; NNT 3.58 Ryan et al., Lancet 2012 Funded by Australia's National Health and Medical Research Council and Hunter Medical Research Institute (per trial abstract); HMRI separately reports Greaves Family support for Dr Ryan's research Weak
Speech and language therapy improves cough quality of life Two trials, 162 adults; one trial showed an LCQ difference of 1.53 (95% CI 0.21 to 2.85) at four weeks, not seen at three months Cochrane CD013067 Review authors declare a device donation for a separate study and unrelated pharma payments Weak
Gefapixant 45 mg twice daily reduces cough frequency 24-hour cough frequency down 18.5% (COUGH-1, week 12) and 14.6% (COUGH-2, week 24) versus placebo; 15 mg not significant McGarvey et al., Lancet 2022 Sponsored by Merck Sharp & Dohme; authors include Merck employees who hold stock Moderate
Gefapixant causes taste change Dysgeusia in 16.2% (COUGH-1) and 21.1% (COUGH-2) of participants across all arms McGarvey et al., Lancet 2022 Maker-reported trial safety data Risk
Gefapixant is approved for refractory or unexplained chronic cough in the EU Marketing authorisation issued 15 September 2023 EMA EPAR Marketing holder is Merck Sharp & Dohme B.V. Moderate
Gefapixant was not approved in the US December 2023 complete response letter: no evidence of effectiveness shown Trade report Trade press; the FDA letter itself was not opened Insufficient
Camlipixant, a newer P2X3 drug, had fewer taste effects in a pooled analysis; GSK later ended its development Pooled analysis: camlipixant 50 mg or more twice daily cut cough frequency by about 34%, with dysgeusia in 5 to 7%. CALM-2 missed its primary endpoint and GSK stopped development in July 2026 (trade reports) 2026 meta-analysis No funding statement; authors declare no competing interests Insufficient

Independent evidence and credibility scorecard

Independence tiers: 1 = regulator or government body; 2 = academic, professional society or journal source with no declared commercial funding for the work that we could see; 3 = independent authors with minor declared industry ties or device donations; 4 = manufacturer-sponsored or manufacturer-authored. Credibility runs from A (highest) to D.

Source Who funds it Country Independence tier (1–4) Credibility (A–D) Why it would tell the truth / residual bias
EMA EPAR for Lyfnua EU regulator; applicant fees not checked in this review EU 1 A Legal mandate to judge benefit and risk. Residual bias: a positive decision only reflects the sponsor's submitted data.
ClinicalTrials.gov (NCT03449147, NCT03449134) US National Library of Medicine (NIH) US 1 A Shows the sponsor and design, not the results. Records are self-reported by sponsors.
ERS guideline (Morice et al.) European Respiratory Society; funding statement not located Europe 2 B Professional body with a clinical duty to get guidance right. Residual bias: most recommendations are conditional and low-quality, and the guideline cannot show what is missing from trials.
ACCP cough guideline (Irwin et al., Chest 2006) Not verified; PubMed record shows no grant or conflict statement US 2 C Academic guideline, but dated. It predates gefapixant and camlipixant, so we use it for background only.
Ryan et al., gabapentin trial (Lancet 2012) NHMRC (Australian government) and Hunter Medical Research Institute, per trial abstract Australia 2 B Academic trial with a clear primary outcome. Residual bias: small sample (62 randomised), one site, and no published pooled replication.
Cochrane review CD013067 (speech therapy, 2019) Cochrane; review authors declare a device donation (Breas Medical) for a separate study and unrelated pharma payments International Cochrane review (headquarters not checked) 3 B Systematic method and public protocol. Residual bias: only two trials, low-certainty outcomes, and a search that ended in February 2019.
McGarvey et al., COUGH-1 and COUGH-2 (Lancet 2022) Merck Sharp & Dohme (sponsor); eight authors are Merck employees who hold stock US (sponsor); trials in 17 and 20 countries 4 D The sponsor has a commercial interest in a positive result. The trials are large and placebo-controlled, which makes them the strongest data on the drug, but the sponsor ran and reported them.
Trade report of the FDA letter (Dec 2023) Ad-funded trade publication; ad funding not checked US 2 C Reports a regulatory decision, but is not the letter itself. Residual bias: trade media depends on industry advertising and access.
2026 meta-analysis of P2X3 antagonists No funding statement; authors declare no competing interests Not verified 2 B Pools nine trials; we did not check how many were sponsor-run, so it may inherit sponsor bias. Its benefit of a systematic search is offset by the lack of any funding disclosure.

Convergence and clustering: The sources do not converge on gefapixant. The EU regulator and the sponsor's trials point to a small but real effect on cough frequency. The US regulator decided the evidence of effectiveness was insufficient. Four of nine sources are US-based, and the two regulators sit on opposite sides of the Atlantic, so the verdict partly reflects which regulator a reader trusts. The Cochrane review is the most methodical source here, but it covers speech therapy, not the drugs, and it declares its own industry payments.

What chronic cough is, and its common causes

Chronic cough is a cough that lasts eight weeks or longer in adults. The ERS guideline defines it this way and gives a separate cut-off of four weeks for children (ERS guideline). Many people with a persistent cough have one of a few recognised causes. Others have a cough that stays unexplained after standard tests. Clinicians often call this refractory chronic cough or unexplained chronic cough.

Common causes named in the ERS guideline

  • Asthma and cough-variant asthma. The guideline suggests a two- to four-week trial of an inhaled corticosteroid in adults, and gives low-quality evidence for it (ERS guideline).
  • ACE inhibitors. The guideline says about 15% of people taking an ACE inhibitor develop chronic cough. It advises that people with cough should not be given ACE inhibitors, and states that angiotensin II receptor blockers do not affect the cough reflex (ERS guideline). Our ramipril page covers this side effect in more detail.
  • Reflux. The guideline recommends clinicians do not routinely use anti-acid drugs for chronic cough in adults. Acid-suppressing treatment gave no significant benefit over placebo without acid reflux and only a modest benefit with it. The guideline also says that 24-hour pH monitoring does not help without peptic symptoms (ERS guideline). For the reflux evidence itself, see our GERD article and lansoprazole page.
  • Post-infectious cough. The guideline names post-infectious cough as a common cause, alongside natural resolution, asthma and protracted bacterial bronchitis. It gives no separate treatment recommendation for adults in the sections we read (ERS guideline).

How the cough reflex becomes persistent

The working idea behind the unexplained cough is that the cough reflex becomes too sensitive. Nerves in the airway respond to small stimuli, and the brain amplifies the signal. The ERS guideline does not settle the mechanism, but it discusses this sensitisation as the reason some people cough without a clear cause (ERS guideline).

Two drug approaches follow from this idea. Gabapentin is a nerve-acting medicine used for nerve pain, and the trial authors chose it because the cough reflex behaves like neuropathic pain. Gefapixant and camlipixant block P2X3 receptors on airway nerves, which carry the signal that ATP sends to the cough reflex. Gefapixant also blocks P2X2/3 receptors in the taste pathway, which explains the taste side effects. The proposed mechanisms are supported by trial results, but they have not been shown to be the direct cause of any one person's cough (Ryan et al., Lancet 2012; 2026 meta-analysis).

What works and what does not

Approach What the evidence shows Status
Treating asthma, or a short inhaled steroid trial Guideline suggestion from low-quality evidence; benefit depends on a diagnosis of asthma or airflow obstruction Mixed
Stopping an ACE inhibitor when it causes cough Guideline advice; switching to an angiotensin II receptor blocker does not cause the same cough Works
Routine acid suppression with no reflux symptoms No significant benefit over placebo; guideline advises against routine use Does not work
Gabapentin for refractory cough One small trial with a quality-of-life gain; the guideline suggests a trial, low-quality Mixed
Speech and language therapy Two small trials; a quality-of-life gain at four weeks in one trial, not seen at three months Mixed
Gefapixant 45 mg twice daily Modest reduction in 24-hour cough frequency in two maker-sponsored trials; US regulator not convinced of effectiveness Mixed
Gefapixant 15 mg twice daily No significant reduction versus placebo in either trial Insufficient

Treatments by claim

Can gabapentin stop a chronic cough?

The placebo-controlled trial we found for gabapentin in refractory chronic cough randomised 62 adults in Australia. Participants had cough for more than eight weeks and no active infection. They were given gabapentin up to 1,800 mg a day, or placebo, for 10 weeks. The primary outcome was change in the Leicester Cough Questionnaire, which measures cough-specific quality of life. The gabapentin group improved by 1.80 points more than placebo (95% CI 0.56 to 3.04; p=0.004). The authors calculated a number needed to treat of 3.58 for this outcome (Ryan et al., Lancet 2012). Put in plain English, about one in three or four people in the gabapentin group had a meaningful gain that they would not have had on placebo. That is a single small trial, and it needs repeating in larger groups before it can be called established.

The ERS guideline suggests a trial of gabapentin or pregabalin in adults with chronic refractory cough, and grades the evidence as low quality (ERS guideline). This matches the picture from one trial: plausible, but not yet confirmed.

Does gefapixant reduce cough?

Gefapixant reduced 24-hour cough frequency in both large trials. In COUGH-1, which included 730 people, the 45 mg twice-daily dose reduced cough frequency by 18.5% at week 12 (95% CI 0.9% to 32.9%; p=0.041). In COUGH-2, which included 1,314 people, the same dose reduced it by 14.6% at week 24 (95% CI 1.4% to 26.1%; p=0.031). The 15 mg twice-daily dose did not reach significance in either study (McGarvey et al., Lancet 2022). Those are modest relative reductions. Whether they change a person's life is a separate question that the trials measured only partly.

A 2026 meta-analysis of nine randomised trials (1,934 participants in total) found that dual P2X2/3 antagonists such as gefapixant reduced cough frequency more than selective P2X3 antagonists: a mean of 38.1% against 19.9% (p=0.016). The same analysis found dysgeusia in 51% of the dual-antagonist arms against 8% of the selective arms (p<0.0001). The analysis reports that camlipixant at 50 mg or more twice daily gave a reduction of about 34%, with dysgeusia in 5 to 7% (2026 meta-analysis). The authors declare no competing interests. We did not check how many of the nine pooled trials were sponsor-run.

Camlipixant's phase 3 record for refractory cough is now mixed. Trade reports agree that GSK's CALM-1 trial met its primary endpoint at 50 mg twice daily, that CALM-2 did not, and that GSK announced on 17 July 2026 that it would not progress development in refractory chronic cough. We could read the GSK press release only as a PDF without extractable text, so these details rest on trade reports. The pooled analysis above was published in June 2026, and camlipixant is not approved for cough anywhere, so its figures are not a current treatment option.

Does speech and language therapy help?

The 2019 Cochrane review pooled two randomised trials with 162 adults with unexplained chronic cough. In both trials the control arm received healthy-lifestyle advice. One trial contributed quality-of-life data using the Leicester Cough Questionnaire. It found a mean difference of 1.53 points at four weeks (95% CI 0.21 to 2.85; 71 participants), but the difference was not seen between week four and three months. The reviewers rated the quality of evidence as low using GRADE. That trial also reported no serious adverse events in either arm, again at low quality (Cochrane review CD013067). The ERS guideline recommends cough-directed physiotherapy for experienced practitioners and also refers to speech and language therapy (ERS guideline).

Does treating reflux stop a chronic cough?

Not in general. The ERS guideline advises against routine anti-acid treatment for chronic cough in adults. Acid-suppressing drugs gave no significant benefit over placebo in people without reflux, and only a modest benefit in people with reflux (ERS guideline). If you have heartburn or regurgitation alongside the cough, that is a reason to see a clinician, but it is a different question from whether acid-suppressing medicine will treat the cough itself. Our GERD article covers the reflux evidence in more detail.

Can an ACE inhibitor cause chronic cough?

Yes. The ERS guideline puts the rate at about 15% of people taking an ACE inhibitor, and says that people who develop cough should not be given these drugs. Angiotensin II receptor blockers do not affect the cough reflex in the same way (ERS guideline). The decision to change a blood-pressure medicine belongs to the prescriber. Our ramipril page has more on this.

Risks and all side effects

Gefapixant: taste change
The most common adverse events in both trials were related to taste. Across all arms, dysgeusia (a distorted taste) was reported by 16.2% of participants in COUGH-1 and 21.1% in COUGH-2. Ageusia (loss of taste) was reported by 4.9% and 6.5%, respectively. We did not open the full safety section, so we cannot report how many of these effects resolved. The trial abstract describes taste disturbance as the most common adverse event (McGarvey et al., Lancet 2022).

Gabapentin

We could not open the side-effect table of the gabapentin trial, so we give no rate here. The abstract we read does not list the adverse-event rates. We did not check current drug safety notices for gabapentin in this review, so readers should look at their national medicines regulator's current advice.

ACE-inhibitor cough

The ERS guideline advises that people who develop cough on an ACE inhibitor should not be given these drugs, and that angiotensin II receptor blockers do not affect the cough reflex (ERS guideline). Do not stop a blood-pressure medicine on your own. Ask the prescriber to review it.

Speech therapy

The Cochrane review found no serious adverse events in the single trial that reported them, but the evidence on safety is thin (Cochrane review CD013067).

All interactions

  • ACE inhibitors and ARBs. ACE inhibitors can cause cough. Angiotensin II receptor blockers do not affect the cough reflex in the same way, so a switch is one option the prescriber may consider (ERS guideline).
  • Gabapentin with other medicines. We did not verify interaction data for gabapentin from a primary source in this review. Check the product information or a pharmacist before combining it with other medicines.
  • Gefapixant and taste-altering medicines. The trial data we read do not report interaction studies. Taste disturbance is the main known issue.
  • Inhaled corticosteroids and bronchodilators. These are used in asthma and fixed airflow obstruction. The ERS guideline advises against using a bronchodilator alone as maintenance treatment for asthmatic cough (ERS guideline).

Who should take extra care

  • People taking an ACE inhibitor with a new cough. Ask a prescriber to review the medicine. The ERS guideline advises that people with cough should not be given ACE inhibitors.
  • Pregnancy and breastfeeding. We did not check the pregnancy data in the product information or the trial reports, so this needs a clinician's advice.
  • People with kidney disease. We did not check the kidney-related dosing advice for gabapentin or gefapixant in this review, so this needs a clinician's advice.
  • Children. The ERS guideline uses different definitions and recommendations for children. None of the drug trials we read enrolled children (ERS guideline).

Doses studied and product notes

We report only doses used in trials. These are not instructions.

  • Gabapentin: up to 1,800 mg a day, the maximum tolerable dose, for 10 weeks (Ryan et al., Lancet 2012).
  • Gefapixant: 45 mg twice daily for 12 weeks (COUGH-1) or 24 weeks (COUGH-2), with extension periods to 52 weeks. The 15 mg twice-daily dose was also tested and did not reach significance (McGarvey et al., Lancet 2022).
  • Camlipixant: the pooled analysis reports 50 mg or more twice daily (2026 meta-analysis).

Product quality matters less here than trial design, because the drugs are prescription medicines made to regulated standards. We did not check the labels or batch testing for any product in this review.

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

Who makes and sells the drugs

Gefapixant is developed and sponsored by Merck Sharp & Dohme. The ClinicalTrials.gov registry lists the sponsor as Merck Sharp & Dohme LLC for both COUGH-1 (NCT03449134) and COUGH-2 (NCT03449147) (ClinicalTrials.gov). The Lancet paper's declarations say the sponsor is a subsidiary of Merck & Co, and that eight authors are employees who own stock (McGarvey et al., Lancet 2022). In the EU, the marketing authorisation holder for Lyfnua is Merck Sharp & Dohme B.V., based in Haarlem, the Netherlands (EMA). The COUGH trials ran at 156 sites in 17 countries (COUGH-1) and 175 sites in 20 countries (COUGH-2).

Gabapentin is an established medicine, so the 2012 trial did not test a new product. We did not trace a manufacturer or supplier for that trial, so we cannot say whether a product sponsor shaped it.

Camlipixant was developed by Bellus Health. GSK completed its acquisition of Bellus Health on 28 June 2023 (about US$2.0 billion), which included camlipixant. GSK later ended development of camlipixant in refractory chronic cough (see Treatments by claim).

Who funded the trials

  • Gefapixant (COUGH-1 and COUGH-2). Sponsored by Merck Sharp & Dohme. Tier 4. Several other authors report grants, fees or advisory roles from Merck and from other companies (Lancet 2022 declarations).
  • Gabapentin (Ryan et al., 2012). Funded by the National Health and Medical Research Council of Australia and the Hunter Medical Research Institute, according to the trial abstract. HMRI separately says Dr Ryan's research was supported by the Greaves Family (HMRI). Its registration is ACTRN12608000248369 (Australian New Zealand Clinical Trials Registry); we did not open the registry entry.
  • Speech therapy (Cochrane review). The review authors declare a device donation from Breas Medical for a separate study on laryngoscopy during cough-assist use, plus unrelated honoraria and conference support from pharma companies (Cochrane review CD013067). These are unrelated to the review, but they are declared.
  • EU regulator. The EMA is a public agency. Whether applicants pay fees that fund the agency was not checked here.

Documented integrity events

The FDA complete response letter in December 2023 is the main documented regulatory event. The agency said the application did not provide evidence of effectiveness, and it did not raise a safety concern (trade report). We did not find a documented fraud, recall or safety action for gefapixant in this search. This is not proof that none exists, only that we did not find one.

Follow the money

Research funding at a glance

Funding & backersSource & studyClaim & limits

9 sources graded. The counts below are the independence tiers assigned in the scorecard above. They count sources, not funding amounts or evidence quality.

Tier 12Independent
Tier 25Indirect ties
Tier 31Interested party
Tier 41Self-interested

Public funding alone does not prove independence, and commercial funding alone does not prove a claim false. Named funders, countries and relationships are in the scorecard and source notes.

Frequently asked questions

How long does a cough have to last to be chronic?

Eight weeks or longer in adults, according to the ERS guideline. The guideline uses four weeks for children (ERS guideline).

Can ACE inhibitors cause a dry cough?

Yes. The ERS guideline puts the rate at about 15% of people taking the drug, and advises that people who develop cough should not continue it (ERS guideline). The prescriber should review the medicine. Do not stop it on your own.

Does gabapentin work for chronic cough?

One small placebo-controlled trial of 62 adults found a quality-of-life gain of 1.80 points more than placebo at 10 weeks (95% CI 0.56 to 3.04) (Ryan et al., Lancet 2012). The ERS guideline suggests a trial but rates the evidence low quality (ERS guideline).

Is gefapixant approved?

It is authorised in the EU as Lyfnua for adults with refractory or unexplained chronic cough, since 15 September 2023 (EMA). The US FDA issued a complete response letter in December 2023 because the application did not show evidence of effectiveness (trade report). We could not open the FDA letter itself.

Does gefapixant make food taste strange?

Taste change is the main side effect. Dysgeusia was reported by 16.2% of participants in COUGH-1 and 21.1% in COUGH-2, across all arms. We could not confirm from the trial abstract how many of these reversed (McGarvey et al., Lancet 2022).

Does speech therapy stop chronic cough?

The 2019 Cochrane review found two small trials. One showed a quality-of-life gain at four weeks, low certainty, and no difference by three months (Cochrane review CD013067). The evidence is too thin to say it stops the cough.

Are over-the-counter cough medicines useful for chronic cough?

This article does not recommend any product. Over-the-counter cough remedies were not reviewed here. A cough that lasts more than eight weeks needs a clinician's assessment for the causes listed above before a treatment is chosen.

Sources and funding notes

What we could not verify: The gabapentin side-effect rates and the gefapixant safety section were not opened in full. The funding statements for the gabapentin trial (the Lancet full text returned a 403 block), for the COUGH trials (we read the declarations from the PubMed record but not the printed funding line), and for the ERS guideline in the sections we read. The Australian New Zealand Clinical Trials Registry entry for the gabapentin trial could not be opened. The ACCP guideline was read as a PubMed record only, not in full. The Cochrane review was read from its PubMed abstract, not the full review. Camlipixant's phase 3 outcomes and the July 2026 discontinuation rest on trade reports; we could not extract text from GSK's press release. The FDA letter was read through a trade publication. The Bellus acquisition was checked against GSK's completion announcement (June 2023); camlipixant's status after GSK's decision was not checked further. The Japanese approval of gefapixant was not verified. We did not check current drug safety notices for gabapentin, or the EMA's fee structure.

  1. Morice AH, Millqvist E, Bieksiene K, et al. ERS guidelines on the diagnosis and treatment of chronic cough in adults and children. Eur Respir J 2020;55:1901136. Professional guideline; funding statement not located in the text we read.
  2. Irwin RS, Baumann MH, Bolser DC, et al. Diagnosis and management of cough executive summary: ACCP evidence-based clinical practice guidelines. Chest 2006. Guideline; funding not verified.
  3. Ryan NM, Birring SS, Gibson PG. Gabapentin for refractory chronic cough: a randomised, double-blind, placebo-controlled trial. Lancet 2012;380:1583-89. Funded by the NHMRC and Hunter Medical Research Institute, per the abstract. Support note from Hunter Medical Research Institute (Greaves Family support for Dr Ryan's research).
  4. Speech and language therapy for management of chronic cough. Cochrane Database Syst Rev 2019 (CD013067). Systematic review; declarations include a device donation and unrelated pharma payments.
  5. McGarvey LP, et al. Efficacy and safety of gefapixant in refractory and unexplained chronic cough (COUGH-1 and COUGH-2). Lancet 2022;399:909-23. Sponsored by Merck Sharp & Dohme; authors include employees with stock.
  6. ClinicalTrials.gov NCT03449147 (COUGH-2) and NCT03449134 (COUGH-1). Registry; US National Library of Medicine; sponsor Merck Sharp & Dohme LLC.
  7. European Medicines Agency: Lyfnua EPAR. Regulator; marketing holder Merck Sharp & Dohme B.V.
  8. Managed Healthcare Executive: Merck receives complete response letter for gefapixant (December 2023). Trade press; advertising-funded.
  9. Systematic review and dose-response meta-analysis of P2X3 versus dual P2X2/3 antagonists in refractory chronic cough (2026). No funding statement; authors declare no competing interests.

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.

Have a question — or want us to cover something?

Ask about anything on this page, or request the next deep dive: an ingredient, a supplement, or a health concern. We use published research, evidence syntheses, and regulatory guidance, with clear source links.

We store your topic, message, optional email, and this page so we can manage and reply to the request. Do not include diagnoses, medications, or other sensitive medical information. See our Privacy Policy.