Most coughs come from a cold and settle on their own over about two to three weeks, longer than most people expect, so the usual job is to wait it out. Honey eases a cold-related cough a little and is safe over the age of one, while cough syrups and antibiotics for a chest cough do close to nothing.
A cough that lingers past eight weeks almost always has a treatable cause, usually postnasal drip, asthma, or reflux, and finding that cause is what makes the cough stop. Coughing up blood is the one sign that always needs a doctor promptly.
Practice Ranking
Every practice we track for Cough, ranked by how well the evidence supports it for this condition. Strength describes the evidence, not our endorsement.
1 practices · 0 to start with
| # | Practice | Evidence | Type | Cost | Effort | Results In | Add to plan |
|---|---|---|---|---|---|---|---|
| 1 | Breathwork & HRV Breathing and cough-control exercises, usually taught by a speech therapist, help a cough that lingers for weeks after everything else has cleared. | Moderate | Self-Directed | Free | Easy | Days to Weeks | |
Default order puts the best-supported practices first, with self-directed care ahead of clinical options. Click any row to open the practice.
What It Is
A cough is the body clearing the airway, and most of the time it is doing that well. An acute cough comes on with a cold or chest infection, is almost always viral, and settles on its own. It lasts about 18 days on average, far longer than the week most people expect, and that gap is the main reason people reach for treatments a viral cough does not need. A cough still hanging on at two weeks is usually just running its normal course.
Coughs are grouped by how long they last, and the length points to the likely cause:
- Acute, under three weeks, is nearly always a viral infection running its course.
- Subacute, three to eight weeks, lingers after a viral illness or after whooping cough while the airway stays irritable as it heals, and usually settles without specific treatment.
- Chronic, more than eight weeks, is where the approach changes: the goal becomes finding the cause.
For a chronic cough in a non-smoker with a normal chest x-ray who is not taking an ACE inhibitor, the big three causes are upper-airway cough syndrome (postnasal drip), asthma or eosinophilic bronchitis, and reflux, and often more than one is at play. These overlap with hay fever, asthma, and acid reflux, which have their own pages.
Two special cases are easy to miss. A dry, tickly cough that starts weeks to months after beginning an ACE inhibitor blood pressure drug, the ones ending in -pril, is a side effect of the drug, not a fresh illness. A refractory chronic cough is one that persists after the usual causes have been chased down and treated, driven by an oversensitive cough reflex that outlasts the original trigger; it is more common in women and can last years.
What settles it
You do more for a stubborn cough by finding its cause than by silencing it. A chronic cough almost always has a findable, treatable driver, and the findings below are graded at the strength of their own evidence and ordered by how much difference they make.
For a cough lasting more than eight weeks, finding and treating the cause is what makes it stop.
The diagnostic work pays off. In a prospective study that tested each part of the cough reflex pathway in turn, postnasal drip, asthma, and reflux accounted for 85% of all the causes found, and directed treatment of the identified cause cleared the cough in every patient. The study was small and in older adults, so the exact success rate is not a promise for everyone, but the principle holds: work through the likely causes in order and treat what you find. A new or changing cough in a smoker, or one lasting beyond three weeks with no clear cause, warrants a chest x-ray first.
Honey is the cheap option with better support. In a Cochrane review of six trials in 899 children, honey eased an acute cough more than no treatment and more than a dummy syrup, and about as well as an over-the-counter cough medicine, with most of the benefit in the first three days. A broader review across children and adults found honey improved symptoms and cut cough frequency against usual care, though the comparison with a matched placebo syrup was weaker, so part of the effect may be the soothing of any sweet liquid. It will not cure a viral infection, and it must never be given to a baby under a year old because of the risk of botulism. For anyone over one, it is a sensible first move ahead of a cough medicine.
A dry cough in someone on an -pril blood pressure drug is one of the most missed causes, because the cough looks ordinary. A network meta-analysis of 135 trials in 45,420 patients put the pooled relative risk of cough on an ACE inhibitor at 2.21 against placebo, roughly double, and the cough clears once the drug is stopped or swapped. ACE inhibitors protect the heart and kidneys and remain valuable drugs, so a cough on one is a reason to recognize the side effect and consider a switch, never to stop a heart medicine on your own. If the timing fits, mention it to your prescriber; switching to an angiotensin receptor blocker usually resolves it.
When the usual causes are treated and the cough persists, the target becomes the oversensitive cough reflex itself, and the options escalate in that order:
- Cough-suppression speech and physiotherapy, taught by a trained therapist, improved cough-related quality of life and cut how often people coughed by 41% in a randomized trial, with the benefit held to three months. It is a drug-free option that can come first or pair with a medication trial.
- Gabapentin, a prescription nerve-signal medicine, improved cough-related quality of life more than placebo, with about one in 3.6 people helped, though the benefit fades when the drug is stopped and it needs dose titration under a doctor.
- Gefapixant, a P2X3 antagonist and the newest option, modestly cut coughing at 45 mg twice a day across two phase 3 trials, while the lower dose did not reliably work and altered taste was a common side effect. It is a specialist treatment for refractory cough, reached only after the rest.
The Research & Studies
Everything here is based on the research we have collected and checked, sorted into groups and ordered with the strongest evidence first. Click any claim to open the studies behind it.
Respiratory Infection
Antibiotics did not speed recovery from a chest cough (acute bronchitis), across 17 trials
Antibiotics do not clear a chest cough (acute bronchitis) any faster in a way that matters. Across 17 trials the chance of getting better was the same as with a dummy pill, and the cough was shortened by under half a day. Acute bronchitis is nearly always viral, which antibiotics do not touch.
A Cochrane review of 17 randomized trials in 5,099 people with acute bronchitis, a productive chest cough in people without underlying lung disease, found no difference in the proportion who were clinically improved between antibiotics and placebo (11 studies, 3,841 people, risk ratio 1.07, 95% CI 0.99 to 1.15). Antibiotics shortened mean cough duration by less than half a day (7 studies, mean difference -0.46 days, 95% CI -0.87 to -0.04), a difference too small to matter to most people. Measured in: 5,099 people with a clinical diagnosis of acute bronchitis and no underlying lung disease, across 17 randomized trials of generally good quality. A small subgroup, such as frail older people with several other conditions, may benefit and was under-represented in the trials, so this is about ordinary chest colds rather than everyone. The near-half-day the antibiotic trims off a cough that lasts two to three weeks is not worth the tradeoffs that come with it.
Who this may not transfer to:Both sexes were enrolled across the pooled trials; effects are not reported separately by sex.
For an ordinary chest cough in an otherwise healthy person, an antibiotic is not the answer and is worth declining. The cough of acute bronchitis often lasts two to three weeks and settles on its own; see a doctor if it drags on beyond that or you become breathless or feverish.
The study · 1
Smith et al., antibiotics for acute bronchitis (Cochrane review) · Cochrane Database Syst Rev 2017;6(6):CD000245
Honey eased children's acute cough better than placebo, most benefit in the first three days
In children, a spoonful of honey eased an acute cough better than no treatment or a dummy syrup, and about as well as an over-the-counter cough medicine, with most of the benefit in the first three days. It is cheap and safe over the age of one, but never give honey to a baby under a year old.
A Cochrane review of six randomized trials in 899 children aged 12 months to 18 years compared honey with no treatment, placebo, or common cough medicines for acute cough. On a 7-point symptom scale where a lower score is better, honey probably reduced cough frequency more than no treatment (mean difference -1.05, 95% CI -1.48 to -0.62) and more than placebo (mean difference -1.62, 95% CI -3.02 to -0.22), both moderate-certainty evidence. Honey performed about as well as dextromethorphan and slightly better than diphenhydramine, with most of the benefit seen over the first three days. Measured in: 899 children aged 12 months to 18 years with acute cough in ambulatory settings, across six randomized trials. The trials were short and some were at risk of bias, so this is a modest effect on a symptom score rather than a cure, and honey must never be given to infants under 12 months because of the risk of botulism. It shortens the nuisance of an acute cough that would settle anyway, which is exactly why it is a reasonable first move rather than a cough medicine.
Who this may not transfer to:Both boys and girls were enrolled across the pooled trials; effects are not reported separately by sex.
For a child over one with an ordinary acute cough, honey before bed is a sensible first step ahead of a cough medicine. Under twelve months it is unsafe because of the botulism risk, so avoid it entirely in babies.
The study · 1
Oduwole et al., honey for acute cough in children (Cochrane review) · Cochrane Database Syst Rev 2018;4(4):CD007094
Across 14 studies, honey cut cold-cough frequency and severity versus usual care
Across studies in children and adults, honey eased the symptoms and cut the frequency of a cold-related cough compared with usual care. The comparison against a matched dummy syrup was weaker, so part of the effect may be any sweet syrup, but it is a cheap and safe thing to try for a viral cough.
A systematic review and meta-analysis pooled 14 studies of honey for upper respiratory tract infections in children and adults. Compared with usual care, honey improved a combined symptom score (three studies, mean difference -3.96, 95% CI -5.42 to -2.51), cough frequency (eight studies, standardized mean difference -0.36, 95% CI -0.50 to -0.21) and cough severity (five studies, standardized mean difference -0.44, 95% CI -0.64 to -0.25). The comparison against placebo rather than usual care was smaller and less certain. Measured in: Children and adults with upper respiratory tract infections across 14 studies, overall risk of bias moderate. Most of the benefit is measured against usual care rather than a matched placebo syrup, and the placebo comparison was inconclusive, so some of the effect may be the soothing of any sweet syrup rather than honey specifically. It is still a cheap, low-risk option for the cough of a common cold, and a reasonable alternative to reaching for an antibiotic that a viral infection does not need.
Who this may not transfer to:Both sexes across children and adults were included; the pooled analyzes do not report effects separately by sex.
For the cough of an ordinary cold, honey or a honey drink is a reasonable, low-cost thing to try. It will not cure a viral infection, and it is not a reason to seek an antibiotic, which does nothing for a cold.
The study · 1
Abuelgasim et al., effectiveness of honey for symptomatic relief in upper respiratory tract infections · BMJ Evid Based Med 2021;26(2):57-64
An ordinary acute cough lasts about 18 days, not the week most people expect
An ordinary cough from a cold or chest infection lasts about 18 days on average, far longer than the week most people expect. That gap is why people reach for antibiotics and syrups that do not help. A cough still hanging around after a couple of weeks is usually just running its normal course.
A systematic review of the natural history of acute cough from respiratory infection pooled studies to establish how long an untreated cough actually lasts. The mean duration of cough was close to 18 days across community and trial samples, considerably longer than the roughly one week most people expect. This mismatch between expectation and reality is a major reason people seek antibiotics and cough medicines for a cough that is simply running its normal course. Measured in: Adults with acute cough from respiratory infection, across the studies pooled in a systematic review of cough natural history. This is the average course of an ordinary post-viral cough, not a rule for every cough, and it is exactly why a cough lasting a few weeks is usually reassurance rather than alarm. It does not override the warning signs: coughing up blood, weight loss, breathlessness or a cough dragging on beyond about three to eight weeks still needs assessment.
Who this may not transfer to:Both sexes were included across the pooled studies; duration is not reported separately by sex.
Expect an acute cough to take two to three weeks to settle, and do not read that as treatment failing. Seek help sooner if you cough up blood, become breathless or feverish, lose weight, or if the cough lasts well beyond three weeks.
The study · 1
Ebell et al., how long does a cough last? Comparing patients' expectations with data from a systematic review of the literature · Ann Fam Med 2013;11(1):5-13
Over-the-counter cough syrups did not reliably beat placebo for acute cough, across 29 trials
The common move of buying a cough syrup or suppressant for a chest cold rests on thin ground. Across 29 trials the results were too mixed to add up, and the verdict was that there is no good evidence these medicines work. They are not dangerous for most adults, but they are not a reliable fix either.
A Cochrane review gathered 29 placebo-controlled randomized trials of oral over-the-counter cough preparations in 4,835 people (3,799 adults and 1,036 children) with acute cough. The trials differed so much in the medicines tested, the people studied and how cough was measured that the reviewers judged pooling inappropriate. The individual results were inconsistent: some antitussive, expectorant and combination trials showed a benefit, others showed none, and antihistamines were no better than placebo. The reviewers concluded there is no good evidence for or against the effectiveness of over-the-counter medicines for acute cough. Measured in: 4,835 people (3,799 adults, 1,036 children) with acute cough from upper respiratory infection, across 29 placebo-controlled trials in community settings. This is an absence of reliable evidence rather than proof the medicines do nothing, and the trials were often small and poorly reported. These products cost money, and the antihistamine and dextromethorphan combinations carry a clear potential for harm in young children, so they are not a dependable treatment for a cough that would settle on its own.
Who this may not transfer to:Both sexes across adults and children were included; the review did not pool results, so no by-sex estimate exists.
Do not expect much from an over-the-counter cough syrup for an acute cough, and do not spend heavily on one. For young children, avoid the antihistamine and dextromethorphan products, which can cause harm; honey and fluids are the safer choice.
The study · 1
Smith et al., over-the-counter medications for acute cough in children and adults in community settings (Cochrane review) · Cochrane Database Syst Rev 2014;(11):CD001831
Respiratory
Gabapentin improved refractory chronic cough quality of life, about one in 3.6 helped
For a chronic cough that persists after the usual causes have been treated, gabapentin, a nerve-signal medicine, improved cough-related quality of life more than a dummy pill, with about one in every three or four people helped. It treats the oversensitive cough reflex rather than curing it, and the benefit stops when the drug does.
A randomized, double-blind, placebo-controlled trial gave 62 adults with refractory chronic cough of more than eight weeks, and no active respiratory disease, either gabapentin up to 1,800 mg a day or placebo for ten weeks. Gabapentin improved cough-specific quality of life on the Leicester Cough Questionnaire more than placebo (between-group difference 1.80, 95% CI 0.56 to 3.04; p=0.004), with a number needed to treat of about 3.6. Side effects, mainly nausea and fatigue, affected 31% on gabapentin against 10% on placebo. Measured in: 62 adults with refractory chronic cough (cough persisting despite investigation and treatment) in a single-center randomized trial in Australia. This is a single, modest-sized trial, the benefit is on how the cough affects quality of life rather than a cure, and the effect fades when the drug is stopped. Gabapentin needs a prescription and titration, and its side effects mean it suits refractory cough under medical supervision, not an ordinary cough.
Who this may not transfer to:Randomisation was stratified by sex and both were enrolled; the small size limits any separate by-sex estimate.
If your cough has lasted months and the standard causes have been chased down without success, gabapentin is a reasonable option to discuss with a doctor. It is a prescription drug that needs dose titration and has side effects, so it is a supervised trial rather than something to self-source.
The study · 1
Ryan et al., gabapentin for refractory chronic cough: a randomised, double-blind, placebo-controlled trial · Lancet 2012;380(9853):1583-1589
Cough-suppression speech therapy improved quality of life and cut coughing 41% in refractory cough
For a stubborn chronic cough, learning cough-suppression techniques with a speech or physiotherapist improved cough-related quality of life and cut how often people coughed, in two trials, with the benefit lasting at least three months. It is a drug-free option that treats the oversensitive cough reflex through retraining.
Two randomized controlled trials tested non-drug speech and physiotherapy programs for refractory chronic cough. A multicenter UK trial of 75 patients found a physiotherapy and speech-and-language intervention (education, laryngeal hygiene, cough-suppression techniques and breathing exercises) improved cough-related quality of life on the Leicester Cough Questionnaire by 1.53 points more than a control program (95% CI 0.21 to 2.85; p=0.024) and cut objective cough frequency by 41%, with gains held to three months. An earlier trial of 87 patients found successful outcomes in 88% of the speech-pathology group versus 14% of controls. Measured in: 162 adults with chronic cough persisting despite medical treatment, across two randomized trials (75 in the UK PSALTI trial, 87 in an Australian trial). The trials were modest in size and delivered by trained speech and language therapists, so results depend on access to that skill rather than a leaflet, and the improvement is in cough control and quality of life rather than a cure. It is a drug-free option that pairs well with, and can precede, a medication trial.
Who this may not transfer to:Both sexes were enrolled across the two trials; refractory chronic cough is more common in women, so the samples skew female, and effects are not reported separately by sex.
Ask about cough-suppression or speech therapy for a chronic cough that persists after the causes have been treated. It works best delivered by a trained therapist and sits comfortably alongside or before a medication trial such as gabapentin.
The studies · 2
Chamberlain Mitchell et al., physiotherapy, and speech and language therapy for refractory chronic cough (PSALTI): a multicentre randomised controlled trial · Thorax 2017;72(2):129-136
Vertigan et al., efficacy of speech pathology management for chronic cough: a randomised placebo-controlled trial · Thorax 2006;61(12):1065-1069
Gefapixant at 45 mg modestly cut coughing in refractory chronic cough, across two phase 3 trials
Gefapixant, a new pill that dampens the oversensitive cough reflex, cut how often people with a long-standing unexplained cough coughed, at the higher dose, across two large trials. The benefit over a dummy pill was modest, the lower dose did not reliably work, and altered taste was a common side effect.
Two phase 3 double-blind trials, COUGH-1 (730 participants) and COUGH-2 (1,314 participants), randomized adults with refractory or unexplained chronic cough of at least a year to placebo or the oral P2X3 receptor antagonist gefapixant at 15 mg or 45 mg twice daily. The 45 mg dose produced a significant reduction in 24-hour cough frequency compared with placebo at week 12 in COUGH-1 and week 24 in COUGH-2, while the lower 15 mg dose did not consistently beat placebo. The average reduction over placebo was modest, and taste disturbance was the most common side effect. Measured in: 2,044 adults (roughly three quarters women, mean age about 58, mean cough duration about 11 years) with refractory or unexplained chronic cough, across two international phase 3 trials. The reduction over placebo was significant but modest, the lower dose was not reliably effective, and altered or lost taste was common enough to make people stop, which is why this drug is a specialist option for refractory cough rather than a general remedy. It is the first of a new class aimed at the oversensitive cough reflex.
Who this may not transfer to:About 74% of participants were women, reflecting that refractory chronic cough is more common in women; men were included but are a minority, so the estimate rests mostly on women.
This is an emerging specialist treatment for chronic cough that has resisted everything else, not a first-line remedy. If your cough is refractory, ask a cough specialist whether a P2X3 antagonist is available and suitable, weighing the modest benefit against the taste side effect.
The study · 1
McGarvey et al., efficacy and safety of gefapixant in refractory or unexplained chronic cough (COUGH-1 and COUGH-2): two phase 3 randomised trials · Lancet 2022;399(10328):909-923
Digestion
Acid-suppression pills did not reliably relieve a chronic cough blamed on reflux, across nine trials
Reaching for acid-blocking pills to treat a chronic cough blamed on reflux does not reliably work. Pooled adult trials found no clear improvement in the cough over a dummy pill, and in infants the drug did not help and caused more side effects. Reflux does cause cough in some people, but acid blockers are not a dependable cough cure.
A Cochrane review examined treating gastroesophageal reflux to relieve prolonged non-specific cough, cough not explained by an underlying lung disease. Pooling nine adult trials of proton pump inhibitors versus placebo over two to three months, there was no significant difference in the total resolution of cough (odds ratio 0.46, 95% CI 0.19 to 1.15) and no overall improvement in cough scores; only sensitivity analyzes of cross-over trials showed a small change. In infants, a proton pump inhibitor did not help cough and increased adverse events. Measured in: Children and adults with prolonged cough and reflux, across 19 studies (nine adult PPI-versus-placebo trials pooled). Acid reflux causes cough in some people, but treating everyone with a chronic cough as if reflux is the cause, with acid blockers, does not reliably work and is a common trap. Where reflux is truly the driver, the lifestyle levers and a proper reflux assessment matter more than escalating acid suppression against the cough.
Who this may not transfer to:Both sexes across children and adults were included; the pooled analyzes do not report cough outcomes separately by sex.
If a chronic cough is being treated as reflux, do not assume acid-blocking pills will fix it; the trial evidence for cough is weak. Ask whether reflux is truly the cause and lean on the lifestyle measures, rather than escalating the acid medicine against the cough alone.
The study · 1
Chang et al., gastro-oesophageal reflux treatment for prolonged non-specific cough in children and adults (Cochrane review) · Cochrane Database Syst Rev 2011;(1):CD004823
Measurement And Diagnosis
Finding the cause resolved chronic cough; postnasal drip, asthma and reflux explained 85%
A chronic cough almost always has a findable cause, and treating that cause is what makes it stop. In this study, postnasal drip, asthma and reflux explained about 85% of chronic coughs, and working through them in order cleared the cough in everyone. The lesson is to chase the cause, not just suppress the cough.
A prospective study applied a systematic anatomic diagnostic protocol, developed to test each part of the cough reflex pathway in turn, to 30 older adults with cough lasting at least three weeks. Postnasal drip syndrome, gastroesophageal reflux disease and asthma accounted for 85% of all causes found, and 100% among non-smokers with a normal chest x-ray who were not taking an ACE inhibitor. Directed treatment of the identified cause eliminated the cough in every patient studied. The approach has been used since 1981 across varied adult populations. Measured in: 30 adults aged 64 and over with cough of at least three weeks, prospectively evaluated with an anatomic diagnostic protocol. This particular study was small and in older adults, so the exact 100% success rate is not a promise for everyone, and a smoker or someone on an ACE inhibitor has different likely causes. The principle still holds: a chronic cough is best resolved by systematically finding its cause, most often postnasal drip, asthma or reflux, rather than by suppressing the cough blind.
Who this may not transfer to:Both sexes were included in a small older-adult sample; the causes and success rate may differ in younger people, smokers and those on ACE inhibitors.
If a cough has lasted more than about eight weeks, the goal is to identify its cause, usually postnasal drip, asthma or reflux, and treat that, rather than to mask it. A new or changing cough in a smoker, or one lasting beyond three weeks unexplained, warrants a chest x-ray first.
The study · 1
Smyrnios et al., from a prospective study of chronic cough: diagnostic and therapeutic aspects in older adults · Arch Intern Med 1998;158(11):1222-1228
What Is Oversold
Two of the most common responses to a cough do very little. Across 29 placebo-controlled trials, over-the-counter cough syrups, expectorants, and suppressants gave results too mixed to add up, so they are not a reliable fix for an acute cough, and the antihistamine and dextromethorphan products can harm young children. Antibiotics fare no better for a chest cough. Across 17 trials in people with acute bronchitis, the chance of getting better was the same as with placebo, and the cough was shortened by less than half a day, while side effects were more common, roughly one extra person harmed for every 24 treated. Acute bronchitis is nearly always viral, which antibiotics do not touch, so for an ordinary chest cough an antibiotic is worth declining.
Reflux causes cough in some people, but treating every chronic cough as reflux, with acid suppression, is a common trap. Pooling nine adult trials of proton pump inhibitors against placebo, there was no clear improvement in the cough, and in infants the drug did not help and caused more side effects. Where reflux is truly the driver, a proper reflux assessment and the lifestyle levers matter more than escalating an acid medicine against the cough alone.
Getting a Cough to Settle
For an ordinary acute cough the main work is patience, since it is a viral illness running its course. These steps ease the nuisance while it does, and set out when a lingering cough turns into a job of finding the cause.
An acute cough averages about 18 days, so a cough still there after a week is usually normal and does not mean a treatment has failed. Fluids and rest help you feel better while a viral cough clears on its own.
A spoonful of honey or a honey drink before bed eases a cold-related cough about as well as an over-the-counter medicine and costs far less. It is safe over the age of one, and must never be given to a baby under twelve months.
Antibiotics do not speed recovery from acute bronchitis and add side effects. Save them for the situations a doctor judges they are needed, such as pneumonia rather than an ordinary chest cough.
Smoking is behind many chronic coughs, and stopping lets the airway lining recover. It is also the change that most improves how well any other treatment works.
If you take an ACE inhibitor blood pressure drug and have a nagging dry cough, mention the timing to your prescriber. Switching to a related drug usually resolves it, but never stop a heart medicine on your own.
A cough lasting more than eight weeks is a reason to be seen and worked up for postnasal drip, asthma or reflux, one cause at a time. If it resists everything, ask about cough-suppression speech therapy and, with a specialist, gabapentin or a P2X3 antagonist.
Go Deeper
- Asthma: one of the big three causes of a chronic cough, and a common reason a cough lingers after a cold.
- Acid reflux (GERD): when reflux is truly the driver, why the lifestyle levers matter more than acid-blocking pills for the cough.
- Hay fever and allergic rhinitis: postnasal drip from an allergic nose is the most common single cause of a chronic cough.
- Breathing retraining: the cough-suppression and breathing work that helps a refractory chronic cough, taught by a therapist.
- Whole foods: where honey fits, and why everyday diet is background rather than a cough treatment.
The Chinese Medicine View
The Chinese Medicine View
Chinese medicine reads cough as Ke Sou (咳嗽) and traces it to the Lung, whose Qi should spread and descend. It divides coughs into those that come from outside and those that arise inside. An outside cough is Wind carrying Cold or Heat invading the Lung and disturbing that downward flow. An inside cough is Phlegm gathering, the Lung Yin running dry, or Liver fire flaring upward against the Lung. In this reading, a chronic dry cough and a wet, rattly cough are different states with opposite needs: one to moisten, one to transform Phlegm. This is an interpretive lens, and a practitioner is the right way to work with it, since diagnosis depends on examining your pulse and tongue in person. Read the patterns to see where your cough sits, not as a diagnosis, and note that a cough with any of the warning signs at the end belongs with a doctor first, not with herbs alone.
The acute cough of a cold caught in cold weather: a sudden cough with an itchy throat, thin white phlegm, a stuffy or runny nose, chills and body aches. The classical direction is to release the exterior, scatter Cold and restore the downward movement of Lung Qi.
A hotter acute picture: a harsh cough with thick or yellow phlegm, a sore dry throat, thirst and often a fever. The direction is to release the exterior, clear Heat and free the Lung so its Qi descends.
A productive, rattly, chronic cough with plenty of phlegm. A damp pattern runs cooler with copious white phlegm and a heavy chest; a heat pattern runs thicker and yellow with a fuller, hotter feeling. The direction is to transform Phlegm, and to clear the Lung where there is heat.
The dry, lingering cough that outlasts an infection: little or no phlegm, a dry scratchy throat, a tickle that will not settle, sometimes a trace of blood in the sputum, often worse in the evening. The direction is to nourish Lung Yin and moisten dryness rather than to dry the cough further.
A cough that flares in bouts with stress, frustration or strong emotion, often with a bitter taste, a flushed face and discomfort in the flanks. Fire from the Liver is seen as rising against the Lung. The direction is to clear the Liver, drain fire and redirect the Lung Qi downward.
Cautions
Extra restraintEverything to be aware of is here, in one place. This practice suits most healthy people; a few situations call for real care.
ACE inhibitor blood pressure drugs about doubled the risk of a dry cough (relative risk 2.21)
A network meta-analysis of 135 randomized trials with 45,420 patients taking one of eleven ACE inhibitors quantified the drugs' well-known dry-cough side effect. The pooled relative risk of cough with an ACE inhibitor versus placebo was 2.21 (95% CI 2.05 to 2.39); ACE inhibitors also caused more cough than angiotensin receptor blockers (relative risk 3.2) and calcium channel blockers (relative risk 5.30). The cough typically appears within weeks to months of starting the drug and resolves after it is stopped. ACE inhibitors are valuable drugs that reduce heart and kidney complications, so this is a reason to recognize a specific side effect, not to abandon the class. The cough is a diagnosis worth making because switching to an angiotensin receptor blocker usually resolves it, and the change is a conversation with your prescriber, never a decision to stop a heart medicine on your own.Hu et al., ACE-inhibitor-induced cough compared with placebo and other antihypertensives: a systematic review and network meta-analysis
Antibiotics for a chest cough caused side effects in about one in 24 people treated
In the same Cochrane review of acute bronchitis, people given antibiotics were significantly more likely to report adverse effects than those given placebo (12 studies, 3,496 people, risk ratio 1.20, 95% CI 1.05 to 1.36), with about one extra person harmed for every 24 treated. The commonest problems were nausea, vomiting, diarrhea, rash and headache, set against a benefit on the cough itself that the same review found to be negligible. These are the immediate, individual side effects and do not count the wider cost of driving antibiotic resistance, which makes the balance for a self-limiting cough worse still. The harm is modest per person, but it is a downside stacked against a benefit the trials could not detect.Smith et al., antibiotics for acute bronchitis (Cochrane review)
Coughing up blood is the strongest cough warning sign for lung cancer
A population-based case-control study analyzed the primary-care records of 247 people with lung cancer and 1,235 matched controls over the two years before diagnosis. Seven symptoms were independently associated with lung cancer: coughing up blood (hemoptysis), weight loss, loss of appetite, breathlessness, chest pain, fatigue and cough, along with finger clubbing, abnormal spirometry, a raised platelet count and smoking. Coughing up blood, breathlessness and abnormal spirometry remained associated even after excluding the last six months before diagnosis, marking them as earlier signals. Most people with these symptoms do not have cancer, and cough alone is common and usually benign, so this identifies who warrants investigation rather than predicting cancer. Coughing up blood is the standout: it is uncommon and carries enough weight to need prompt assessment on its own, especially in a smoker or someone over 40.Hamilton et al., what are the clinical features of lung cancer before the diagnosis is made? A population-based case-control study
Start slow, be smart, read the research, and consult a professional if you have any concerns. This is here to inform your choice, not make it for you.
When to See Someone
Most coughs are ordinary: they come with a cold, last a couple of weeks and settle on their own, and honey and time do more than any syrup. These are the features that point away from a simple cough and toward something that needs a doctor, one or two of them the same day:
- Coughing up blood, even a small amount or streaks in the phlegm, which needs prompt assessment and a chest x-ray(seek urgent care)
- Breathlessness, chest pain or tightness, or a cough with a fast heartbeat or swelling in the legs, which can be the heart or lungs rather than a simple cough and needs urgent assessment(seek urgent care)
- A cough with unplanned weight loss, drenching night sweats or a persistent fever, which can point to an infection such as tuberculosis or to something more serious
- A new cough in a smoker, or a long-standing smoking-related cough that changes in character, sound or severity, which is a reason for a chest x-ray
- A cough lasting more than about three weeks with no clear cause, which is a reason to be seen and to have a chest x-ray, since an ordinary cough is usually settling by then
- A hoarse voice that persists for more than three weeks, difficulty swallowing, or choking on food and drink, which can point to a throat or swallowing problem behind the cough
- In a baby or young child, a cough with fast or labored breathing, a whooping sound between coughing fits, or the lips turning blue, which needs urgent medical care(seek urgent care)
None of this is meant to alarm you. The great majority of coughs are self-limiting or have a cause that treats well once it is found. These are the exceptions to watch for, so you can tell an ordinary cough running its course from one that deserves a closer look.
Common Questions
How long should a cough last before I worry?
An ordinary acute cough lasts about 18 days on average, far longer than the week most people expect, so a cough still there after a week or two is usually just running its course. The time to be seen is when it lasts more than about three weeks with no clear cause, or sooner if you cough up blood, become breathless or feverish, or are losing weight without trying.
Do cough syrups actually work?
Not reliably. Across 29 placebo-controlled trials, over-the-counter cough syrups, expectorants, and suppressants gave results too mixed to show a dependable benefit for an acute cough. They are not dangerous for most adults, but they are not a fix either, and honey is the cheaper option with better support. In young children, avoid the antihistamine and dextromethorphan products, which can cause harm.
Do I need antibiotics for a chest cough?
For an ordinary chest cough (acute bronchitis) in an otherwise healthy person, no. Across 17 trials, antibiotics did not improve recovery and shortened the cough by less than half a day, while adding side effects. Acute bronchitis is nearly always viral. See a doctor if the cough drags on well beyond two to three weeks, you become breathless or feverish, or you cough up blood.
My cough will not go away. What could be causing it?
A cough lasting more than eight weeks almost always has a findable cause. The big three in non-smokers are postnasal drip, asthma and reflux, which together explain most chronic coughs, and more than one can be at play. A dry cough that began after starting an ACE inhibitor blood pressure drug is another common and easily missed cause. The path forward is to work through these with a doctor and treat what is found, rather than to suppress the cough blind.
Explore Related
Other pages this one connects to, by the evidence they share, the outcomes they touch, and the ground they cover.
All 14 sources on this page independently checked and cross-referenced.
Thomas Dehli, Founder & Editor, Sacred Lotus
Sacred Lotus has published Chinese medicine reference material since 2001. Integrative pages are held to the same standard as the herb and formula library: cite the source, grade the claim at its real strength, and say where the research has not looked. This page is educational and it is not medical advice. Last reviewed and updated August 9, 2026.
Evidence strength
How confidently the research supports a claim. Strength describes the evidence, not our endorsement.