Asthma affects 1 in 8 children in the UK. Up to half of these are treated with preventative medicine in the form of low-dose steroids using an inhaler. The National Asthma Treatment Guidelines recommend when this treatment is not working other treatments are started. Studies to support this have taken place in adults but not with children. If patients are instructed how to use inhalers and are given information about asthma, they can control their disease much better. The first part of this study, lasting 4 weeks, will make sure the children and their families understand how to use their inhaler. All children will be given the same steroid inhaler to use and after 4 weeks those still with symptoms will enter the study proper which lasts for 48 weeks. During this part of the study the children will be given one of three treatments. These are:- a steroid inhaler + a dummy tablet, an inhaler containing a steroid and a long-acting reliever + a dummy tablet or a steroid inhaler + an active tablet. In this way the patient, the family and the researchers will not know which of the three treatments the child is taking until the code is broken at the end of the study.

What matters to children is how they feel, are they able to run around and play with friends and are they well enough to go to school. The investigators will assess which of the above treatments best allow these to happen by asking the parents and children to fill in questionnaires on 4 occasions during the study. The investigators will also see which treatment best prevents the need for short courses of steroids tablets during the study. These are commonly given when asthma symptoms worsen.

Most children will be started in the study through their general practitioner clinic. It will take one year to enroll all 900 children. Once enrolled the children will be followed-up in hospital centres. Much of the funding will be required to recruit and follow-up the children, train everyone to the same standard and develop and administer the questionnaires and health economic assessments. Asthma care is an expensive. The investigators will look at the costs and assess which treatment offers most benefit. The team has experience and ability in this field and will ensure the results are well publicised. Any child can withdraw from the study at any time.

Asthma remains the most common medical condition seen in children in primary care and the most frequent cause for medical paediatric hospital admission. It affects 1 in 8 children nationwide, approximately 50% of whom are prescribed low-dose inhaled corticosteroids (ICS). When treatment with low-dose ICS fails to control asthma symptoms, the National Guidelines suggest ensuring compliance, maximising inhaler technique and giving appropriate information about the disease to children and their families. Once these measures have been established and if asthma symptoms persist, the Guidelines recommend changing the treatment (Step 3 of the National Guidelines). The evidence at this step of the Guidelines is much weaker in children than it is in adults. The reasons for this are that few studies have been undertaken in children and most that have taken place have used adult-based outcomes such as lung function measurements. This is unsatisfactory because we know that as a chronic disease entity asthma in children is much more variable than in adults and between periods of symptoms, lung function is often normal. Pharmaceutical company studies have really only been conducted as part of their requirements to obtain a license to market their product. These studies have generally been of short-term duration. They have not added to clinicians' understanding of how and where to use the medications. They have not necessarily selected a representative population due to their entry criteria and their intensive study requirements. Such requirements mean that "real-life" compliance does not occur. In the independent National Dutch Study which attempted to enter patients uncontrolled on low-dose ICS, three treatment groups were employed: inhaled corticosteroids alone, inhaled corticosteroids in double the dose and inhaled corticosteroids + a long-acting beta2 agonist. There was essentially no difference in outcome measures between the three treatment groups as once again the primary outcome measure was that of lung function (FEV1). Comparing this study with a similar adult study6 both of which used lung function as the primary outcome measure, the mean FEV1 on entry into the paediatric study was approximately 89% expected for the children"s heights. In the adult study the mean FEV1 on entry into the study was 74% expected. It is therefore not surprising that the paediatric study was unable to show any differences between the treatment groups.

We do not have the scientific information about how to treat children with asthma who are not well controlled on low-dose ICS therapy. It used to be recommended that when low-dose ICS were not effective their dose should be doubled. Studies in children, however, have investigated this statement and the results are not impressive7 . There is no scientific evidence that when control is poor in children with asthma, the dose of the inhaled steroid should be increased. We have therefore decided not to introduce into this study a treatment limb with a higher ICS dosage. There is anecdotal information, however, from many studies undertaken within the pharmaceutical industry that when children enter a study which is controlled and double-blind in nature, up to 30% of them improve, their symptoms reduce and their lung function increases8. It is therefore surprising that approximately one third of children receiving ICS are prescribed high-dose inhaled steroid therapy (≥ 800micrograms and unlicensed beclometasone dipropionate or equivalent) or they are commenced on "add-on" therapies such as long-acting beta2 agonists (LABA) or leukotriene receptor antagonists (LTRA) in addition to low-dose ICS. Concerns about the safety of high-dose ICS have been raised in relation to growth impairment9, hypoglycaemia10 and suppression of the adrenal cortex11 resulting in warnings on prescribing from the Medicines & Healthcare Products Regulatory Agency (MHRA) in the UK12 and from the Food and Drug Administration (FDA) in the USA. It is therefore unacceptable that approximately one third of children with asthma are being treated with the above regimes. Asthma is a very common condition and the worth of these regimes has not been proven by appropriately devised paediatric studies. The National Guidelines have been developed in a "stepwise" manner, the amount of medication increasing at each step if symptoms are not controlled.

However, as stated above, it may be that childhood asthma differs from that in adults. It seems that relatively poorly controlled asthma in children who exhibit frequent symptoms do not necessarily show abnormal lung function between their periods of symptoms. It is for this reason that in our study we will be concentrating on outcome measures such as exacerbations and quality of life although we will have the opportunity to measure spirometric values at the first (T0) and last (T48) visits in the randomised part of the study. It could be that an increase in medication may only be needed for a short time in children with asthma and there have been suggestions that once control is achieved children should have their add-on therapy reviewed. To incorporate such a step within the present study would make it excessively complicated and would have major implications on the number of patients included. The inclusion of such a step would make the study impractical within the UK. A study is needed which is simple, pragmatic (but placebo-controlled and double-blinded), has outcomes which will be of practical benefit to children and will provide evidence for the use of add-on medications in the most cost effective and efficient way. Children dislike exacerbations. School attendance, daily activity and general well-being increase when asthma is well controlled. Once families understand sufficiently about asthma, inhaler technique has been evaluated and optimised, and compliance issues addressed, one of the reasons why a specific medication may be less effective could be related to the genetic make-up of the patient. In this study we will have the opportunity, through a separate consent process, to collect and store DNA specimens from saliva for later analysis of specific genetic polymorphisms in relation to asthma severity and outcome. This aspect will bring added value to the study.

Inclusion Criteria:

• Children with physician diagnosed asthma aged 6 years -14 years, 11months

• Those requiring frequent short-acting beta2 agonist relief therapy ≥ 7 puffs in the past seven days

• Those with symptoms of asthma (i.e. wheeze, shortness of breath but not cough alone) resulting in:

  i. Nocturnal wakening in the last week because of asthma symptoms and/or ii. Asthma has interfered with usual activities in the last week and/or iii. Those who have had exacerbations, defined as a short course of oral corticosteroids, an unscheduled GP or A&E Department visit or a hospital admission within the previous 6 months

• Fully informed written (proxy) consent and assent, where appropriate

Exclusion Criteria:

• Children receiving long acting beta2-agonists, leukotriene receptor antagonists, regular theophylline therapy or high dose ICS >1000micrograms and unlicensed beclometasone dipropionate or equivalent (at the discretion of the investigator)
• Children with other respiratory diseases, cystic fibrosis, cardiac disease or immunological disorders
• Non-English speaking