Updating guidelines on asthma in adults

The changes to the North of England evidence based guidelines on managing asthma in adults in primary care are not very great (see updated guidelines on bmj.com), but they are important and illustrate the need to keep guidelines updated if they are to continue to be useful in guiding doctors' clinical practice.

The new guidelines draw on the reviews produced by the Cochrane airways collaborative review group¹ since the first North of England guideline in 1996, as well as updating the literature searches and meta-analyses originally conducted by the group.²

The principal change in this useful updated guideline is the increasing role for long acting β₂ agonists. These have replaced oral treatments as additional treatment when patients have nocturnal symptoms or need regular short acting β₂ agonists in spite of inhaled steroids equivalent to beclomethasone 1000 μg daily. A recent Cochrane review of six trials comparing long acting β₂ agonists with oral theophylline supports this recommendation.³ Patients taking salmeterol experienced fewer adverse events than those using theophylline (relative risk 0.38; 95% confidence interval 0.25 to 0.57). This has important practical implications: most patients taking oral theophylline are underdosed, and the need for timed measurements of blood concentrations and the potential for drug interactions are added burdens for both patient and general practitioner.

The guideline also incorporates new Cochrane reviews on smoking cessation,⁴ allergen avoidance,⁵ patient education,⁶ and patient self management plans.⁷ Structured antismoking advice and nicotine replacement therapy have both been shown to be effective in reducing smoking. Allergen avoidance is ineffective and is not recommended. What we know about the effect of giving patients information about asthma is limited to patients attending accident and emergency departments, but there it has been shown to reduce subsequent reattendance. Written self management plans, based either on symptoms or peak flow readings, are effective in reducing morbidity in primary care settings.

One of the strengths of guidelines such as those produced by the North of England evidence based guideline development project is that they start with precisely formulated questions rather than being driven by the available evidence. One side effect of this is that they expose limitations in the available evidence.

Randomised controlled trials are combined in systematic reviews, and the strength of the evidence is objectively graded. In a guideline the strength of the recommendation is then explicitly linked to the strength of evidence and its relevance to the question. This process needs to be a continuous one, with reviews and guidelines updated as new trial data emerge. The Cochrane Collaboration has been pivotal in both establishing and maintaining up to date systematic reviews on the Cochrane library.¹

The production of national guidelines, as well as evidence based summaries such as Clinical Evidence, plays an important part in digesting and presenting Cochrane reviews. A cycle can be created whereby relevant questions and high quality evidence synthesis drive evidence based practice and pull research towards the most relevant unanswered questions for day to day patient care. The funding of this work by healthcare agencies world wide is vital in supporting this health service focus, because the other major funder of research—the pharmaceutical industry—concentrates on developing and testing new products, rather than establishing the effectiveness of off patent or non-drug interventions.

The British Thoracic Society guidelines on asthma are probably the most widely known and comprehensive guidelines on asthma available in Britain (OK??), but there is a striking gap in evidence to support many of the practical steps recommended by the British Thoracic Society guidelines.⁸ These include the stepwise sequencing of asthma therapies, the point at which to initiate inhaled corticosteroids, the role of doubling the dose of inhaled corticosteroids in avoiding acute exacerbations, and the role of leukotriene antagonists. This last is a good example of poor research planning. Existing comparisons of leukotriene antagonists with placebo are of little help in deciding how this new treatment compares with existing practice. The NHS spent £465m on drugs and appliances for asthma in 1997, and there are large differences in cost between different devices. There are no reliable data on either the cost effectiveness of different sequencing strategies or different devices. The North of England guidelines adopt the British Thoracic Society consensus that the cheapest device that can be used by a patient should be prescribed. However, there is a need for pragmatic, cost effectiveness trials in this area.

Ten years after the first use of the term evidence based medicine⁹ it is clear that the process by which research influences practice needs to be as rigorous as primary research. Similarly, passive dissemination of guidelines, however evidence based, does not influence practice significantly. Guidelines need to be operationalised using multifaceted interventions such as academic detailing, prompts and reminders, templates, and managed care systems.¹⁰ Primary care organisations need to develop processes whereby updated evidence based guidelines are translated into educational activity, clinical governance, and local computer templates and protocols. But the first step is to do the updating—and that the North of England group have done.