EDITOR, - The authors of the Grampian asthma study of integrated care failed to provide convincing evidence that routine self monitoring of peak expiratory flow does not improve asthma morbidity.1 Several design flaws in the study related to power, patient selection, the intervention delivered, and outcome measures need discussing.

The power calculation is based on 569 subjects, but data is provided for only 485 (239 monitoring peak expiratory flow, 246 controls); what about the other 84 patients? The physicians' guidelines included two action levels depending on peak expiratory flow: when to start oral steroids (for severe episodes of asthma) and when to seek medical assistance. Given that patients with mild asthma were selected (those possessing peak flow meters were excluded from randomisation) and the weak power, the study design biased evaluation against effective routine self monitoring of peak expiratory flow.

At least four hospital centres (and an unstated number of doctors) participated, and the physicians had the freedom to tailor the self monitoring guidelines “according to their clinical judgment.” The education of patients in self monitoring was not therefore standardised to an acceptable level.

Evaluating the effects of self monitoring is extremely difficult, given the diversity with which health professionals provide information and how patients utilise this to adjust medication and seek medical assistance when appropriate. The outcomes used were mainly doctors' behaviour (number of prescriptions issued) and morbidity (sleep disturbance, restricted activity) - indirect measures of efficacy of self monitoring; the study did not measure changed patient behaviour, which is the aim of education in self monitoring. Measurement of changed behaviour in patients depends on them having episodes of uncontrolled asthma; the mild nature of these patients' asthma biased the study. The most important finding in this study was the 20% increase in consultations for the self monitoring …