Article Text

Download PDFPDF

Cough medications in children
  • Relevant BNF section: 3.9


Cough in children is a common cause for consultations in general practice. Over-the-counter medicines marketed for treating cough are widely available, but relatively few can be prescribed on the NHS. In this article, we review the evidence for the efficacy of cough medications and ask whether their widespread use in children with upper respiratory tract infections is justifiable.

View Full Text

Statistics from

  • Relevant BNF section: 3.9


In children, the commonest cause of upper respiratory tract symptoms is probably viral infection. On average, children have four to eight such infections every year.1 In addition, respiratory allergies often develop during childhood (especially in children who are atopic) and this can complicate diagnosis of the cause of cough.

Making a working diagnosis of the cause of a cough is important since cough itself should not necessarily be the target for treatment. In children with a persistent night-time cough or a persistent exercise-induced cough, chronic illnesses such as asthma or other atopic conditions, cystic fibrosis, or inhalation of a foreign body should be considered. Cough may persist long after an upper respiratory tract infection so there may be no obvious underlying cause. Sometimes cough may be associated with persistent infection when the child may cough up purulent sputum, and sometimes it becomes a behavioural habit. When no specific cause is identified, treatment is usually unnecessary. Reassurance and explanation to the parents to this effect is often sufficient, and backing this up with an information leaflet to take home may be useful. The usual reasons for giving a cough medication to a child is if the cough is thought to be dry and causing discomfort or pain, or if it is severe enough to seriously disturb sleep.

Cough suppresssants

Drugs for cough suppression available on an NHS prescription include codeine phosphate and pholcodine.

Codeine is an opioid which is converted to morphine following oral administration.2 It is thought that codeine blocks the triggering of the cough impulse by binding to both peripheral and CNS (medullary) cough receptors. Paediatric codeine linctus (codeine phosphate 3mg/5mL) is available. The manufacturers' recommended dose for children aged 1-5 years is 5mL, three to four times daily. The recommended dose for children aged 5-12 years is 2.5-5mL of adult codeine linctus (15mg/5mL) three to four times daily.

Pholcodine is structurally related to opioids, and is also thought to bind to receptors in the CNS cough centre. A paediatric pholcodine linctus (pholcodine 2mg/5mL) is available. The manufacturers recommend a dose of 2.5mL three times daily for infants aged 3 months to 1 year, 5mL three times daily for children aged 1-5 years, and 5-10mL three times daily for those aged 6-12 years. An adult linctus (containing pholcodine 5mg/5mL) can be given to children of 5-12 years at a recommended dosage of 2.5-5mL, three to four times daily.

Combination cough medication

Around 200 over-the-counter (OTC) combination products are currently listed in the BNF for the treatment of upper respiratory tract symptoms such as cough.3 Most children's cough preparations contain antihistamines, cough suppressants, stimulant decongestants and analgesics, in variable combinations and doses. The most common active ingredients are diphenhydramine (an antihistamine), dextromethorphan (a cough suppressant) and guaifenesin (an expectorant). Other ingredients include phenylephrine, phenylpropanolamine (decongestants), and chlorpheniramine, pheniramine, pyrilamine and triprolidine (all antihistamines). Demulcent cough preparations, such as paediatric simple linctus (containing citric acid monohydrate 0.625% in a suitable vehicle with anise flavour) are also used with the aim of producing a local soothing effect.

Dextromethorphan is chemically derived from the opiates, but has no analgesic or addictive properties. Like codeine, it is believed to act on the cough centre in the medulla, and raise the cough reflex threshold.2 It is reported to have fewer gastrointestinal unwanted effects.

Diphenhydramine is the antihistamine most commonly included in children's OTC cold preparations. Antihistamines are used as the cough suppressant component of many combination cough medications. However, they often cause sedation which may reflect their main mode of action.3 This allows the child and parents to sleep rather than truly suppressing the cough.

Ephedrine is a sympathomimetic. It constricts blood vessels in nasal mucosa and thereby causes decongestion. It is also a potent CNS stimulant. When used for prolonged periods, tachyphylaxis diminishes its activity.

Pseudoephedrine is a stereoisomer of ephedrine and is also a CNS stimulant. It acts by a similar mechanism but has less pressor activity and fewer CNS effects.

Guaifenesin is believed to reduce the viscosity of sputum and is used as an expectorant. However, there is no evidence that doses of expectorants at a level that does not cause vomiting can specifically facilitate expectoration.3

Clinical studies

Most of the studies have involved branded products that are no longer marketed. However, many of the ingredients of the studied products are used in different combinations in formulations that are currently available. Studies have been hindered by the difficulties in measuring symptoms objectively (especially in children) and in classifying different types of cough. Although codeine and dextromethorphan are effective in suppressing cough in adults, there is little or no evidence for their efficacy in children.4 Also, the dosing guidelines for these agents are based on extrapolation from adult data, without consideration of possible differences in metabolism or unwanted effects in children.4

Placebo-controlled studies

We are aware of five placebo-controlled studies of the use of cough medications in children. In an early study, the efficacy of two cough-suppressant preparations was assessed in 43 children (aged 2 months-12 years; average 3.6 years) presenting principally with a cough, in general practice.5 Participants were assigned to: syrup A (Triaminicol syrup, containing dextromethorphan hydrobromide 15mg/5mL, phenylpropanolamine hydrochloride 12.5mg/5mL, pheniramine maleate 6.25mg/5mL, pyrilamine maleate 6.25mg/5mL and ammonium chloride 90mg/mL); syrup B (Dorcal syrup, containing dextromethorphan hydrobromide 7.5mg/5mL, phenylpropanolamine hydrochloride 8.75mg/5mL, glyceryl guaiacolate 37.5mg/5mL and alcohol 5%); or syrup C (placebo containing no 'active' ingredients but resembling syrup B in colour and taste). It is not clear from the trial report whether children were randomly allocated to treatment. Parents gave the medication as required. The anti-tussive response (as recorded by the parents) was judged satisfactory in 69% of those given syrup A, in 69% of those given syrup B and in 57% of those given placebo. The authors claim these differences were statistically significant. Neither Triamcinicol nor Dorcal is available in the UK.

The second study was a double-blind trial involving 57 patients (aged 18 months-12 years) with night-time cough of less than 14 days' duration (and no underlying lung disease). Patients were randomised either to dextromethorphan (15mg/5mL) plus guaifenesin (100mg/5mL), to codeine (10mg/5mL) plus guaifenesin (100mg/5mL) or to placebo.6 The dose of each medication and placebo was 2.5mL of syrup for those under 5 years old and 5mL for older children, given at bedtime on three consecutive nights. Before the trial, parents had rated the frequency of their child's cough during the night before the visit to the physician. To be randomised, a child had to have a cough rated as occurring "often" (i.e. one prolonged coughing episode or about 10-20 coughs during the night, or both) or "very often" (i.e. more than one prolonged coughing episode or more than 20 coughs during the night, or both). Parents were also asked whether the child had experienced loss of sleep or vomiting because of coughing before the trial started. The same questions were put to the parents on the morning after each of the three night-time doses of trial medication. Neither dextromethorphan plus guaifenesin nor codeine plus guaifenesin was better than placebo in reducing cough on any of the three nights. The authors concluded that by 3 days after a visit to a physician, cough will have improved regardless of what (if any) medication is given. We could not find any commercially available products which matched the combination medications used in this trial.

There are two double-blind trials of Dimetapp in different formulations. The Dimetapp preparation used in the first study contained brompheniramine maleate 4mg/5mL, phenylephrine hydrochloride 5mg/5mL and phenylpropanolamine hydrochloride 5mg/5mL, which is currently available in the UK as Dimotapp elixir. In this study, 96 children (aged 6 months to 5 years) with cold symptoms were given Dimetapp (in an amount providing brompheniramine 0.5-0.75mg/kg/day, in three divided doses, for 2 days), placebo or no medication.7 Compared with either placebo or no treatment, Dimetapp did not produce a clinically significant reduction in the symptoms of upper respiratory tract infection (including cough) after 48 hours of therapy. The second trial, involved 59 children aged 6 months-5 years with an upper respiratory tract infection of less than 7 days' duration. Patients were randomised to Dimetapp elixir or to placebo.8 The Dimetapp elixir contained brompheniramine maleate 2mg/5mL plus phenylpropanolamine hydrochloride 12.5mg/5mL, which is not available in the UK. Children under 2 years of age were given 2.5mL of trial medication and older children were given 5mL, when their parents thought it necessary. No child was allowed medication more often than every 4 hours. Changes in cough, nasal symptoms and sleep status were assessed by means of a standardised questionnaire completed by parents 2 hours after each dose of study medication. Improvements in cough and nasal symptoms were similar in both groups. However, the proportion of children who were asleep 2 hours after receiving the active treatment was significantly higher than in the placebo group (46.6% vs. 26.5%; p=0.01).

The fifth study was a double-blind trial which involved 150 children under 5 years of age. The children were randomised to clemastine fumarate 0.05mg/kg/day twice daily (not available in cold medications in the UK), chlorpheniramine maleate syrup 0.35mg/kg/day three times daily, or placebo.9 Neither antihistamine treatment reduced symptoms of upper respiratory tract infection (including cough) compared with placebo.

Other studies

We know of two studies in which cough medications were not compared with placebo. The first was a randomised, single-blind trial, in general practice, in which 217 children (aged 6-12 years) with acute cough were given Pholcolix (5mL four times daily) or Actifed Compound (7.5mL three times daily).10 The Pholcolix preparation contained pholcodine (5mg/5mL), paracetamol (150mg/5mL) and phenylpropanolamine (12.5mg/5mL). The Actifed Compound contained codeine phosphate (10mg/5mL), triprolidine hydrochloride (1.25mg/5mL) and pseudoephedrine hydrochloride (30mg/5mL). Symptoms were scored at the outset and then 72 hours after treatment was completed. Both treatments alleviated productive cough when compared with baseline assessments. There was no significant difference in efficacy between the two groups. As the trial was not placebo-controlled, it is not possible to tell whether the symptoms resolved spontaneously or whether the treatments benefits were caused by the drug or were largely due to a placebo effect. Pholcolix has been discontinued in the UK, and the Actifed Compound which is now available contains dextromethorphan hydrobromide (10mg/5mL) rather than codeine phosphate.

The second trial was a double-blind study involving 60 children (aged 4 years and above) with symptoms of the common cold and associated cough.11 Patients were randomised to an antihistamine/decongestant/cough-suppressant combination syrup (containing azatadine maleate 1mg/5mL, pseudoephedrine sulphate 60mg/5mL and dextromethorphan hydrobromide 20mg/5mL) or an antihistamine/expectorant combination (containing diphenhydramine hydrochloride 0.205g/100g, ammonium chloride 2g/100g, sodium citrate 1g/100g and menthol 0.017g/100g). Both medications were given at a dose of 2.5mL, three or four times daily. Both treatment groups showed improvements in cough and nasal symptoms. The response was quicker with the antihistamine/decongestant/cough-suppressant combination than with the antihistamine/expectorant combination. Again, the absence of an appropriate control group meant that any placebo effect was not quantified. We were unable to find any commercially available products which matched the ingredients of these combination syrups.

Unwanted effects

In a report of 430 children with acute codeine intoxication (234 of whom had taken more than 5mg/kg body weight), 8 experienced respiratory arrest requiring intubation and mechanical ventilation.12 In all the other children, the intoxication produced at least one of the following symptoms: somnolence, ataxia, miosis, vomiting, rash, swelling, and itching. Children who have received more than 2mg/kg body weight of codeine need close supervision.12 Systemic products containing codeine can cause hyperactivity6 and constipation.2

Case reports of important adverse effects with medications for cough and colds have included: visual hallucinations with pseudoephedrine and triprolidine (given as Actifed);13 a dystonic reaction with a preparation containing dextromethorphan and phenylpropanolamine (given as Triaminic DM);14 and behavioural changes including irritability, extreme agitation, hyperactivity and bad temper following the use of a combined preparation containing phenylpropanolamine and chlorpheniramine.15 All these reactions followed the use of the medications in the usual child doses. Additionally, dextromethorphan can cause 'hyperactive' behaviour and drowsiness6 and respiratory depression4 in overdose.

Cough suppressants containing codeine or other opioid analgesics should be avoided altogether in children under 1 year of age,3 and sympathomimetics (such as pseudoephedrine) are not licensed for use in those under 2 years. The clearance and metabolism of the components of cough mixtures may vary with age and disease state, and this could increase the risk of adverse events in certain children.4

Social pressures

The consequences of cough in a distressed child can leave parents feeling stressed and helpless, which in turn can worsen the child's distress and so further exacerbate the stress and anxiety within the family. This vicious cycle understandably leads parents to seek remedies in the form of cough medications. The lack of randomised controlled trials demonstrating their efficacy, the probability that such medications have a marked placebo effect and the risk of unwanted effects, suggests that a simple cough linctus is probably the most appropriate treatment in these circumstances. Parents should also ensure that the child is drinking enough, that any high temperature is treated (e.g. with paracetamol) and that the room is not too dry.


Cough due to viral upper respiratory tract infections is extremely common during childhood. In most children, the cough does not require any medication. When the parents are worried, explanation and reassurance is usually all that is needed. Sometimes, a cough can lead to stress within the family, especially if it is slow to resolve. In such instances, it is important to look for any underlying cause if symptoms prove persistent.

There is no convincing evidence that brand name cough medications are any more effective than placebo in children. If a medication is felt worthwhile, a 'non-active' simple cough linctus is probably the most suitable choice. This has the advantage of having a single purpose (to soothe the local area), and is very likely to have a strong placebo effect. There is certainly no place for fixed-dose combination products.


[M=meta-analysis; R=randomised controlled trial]

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. R 6.
  7. R 7.
  8. R 8.
  9. R 9.
  10. R 10.
  11. R 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
View Abstract

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.