A systematic review of randomised, controlled trials of education in asthma concluded that:

The review used a systematic method of searching the literature. The search terms used for Medline (1966–98), Embase (1980–98), and CINAHL (1982–98) were:

asthma OR wheez ^∗

AND education ^∗ OR self management OR self-management

AND placebo ^∗ OR trial ^∗ OR random ^∗ OR double-blind OR double blind OR single-blind OR single blind OR controlled study OR comparative study.

Other databases and journals were also searched and 318 studies were identified. The quality of studies was assessed on the basis of the design, including whether allocation to the intervention or to the control group was known before each subject was included in the study. This criterion for quality was therefore concerned with potential bias: whether all the children in the population to be studied had an equal chance of being included in the sample. Once these criteria were applied, 45 of 318 identified studies were potentially eligible. However, 13 were excluded when other quality criteria were applied, leaving 32 trials totalling 3706 children and adolescents with asthma.

The stricter the criteria used for selecting studies eligible for inclusion in a systematic review, the fewer studies that will be included, so judgements about the criteria to be applied are important in designing such reviews. Guevara et al (2003) summarised the studies they reviewed:

The conclusions that could be reached by the review were therefore limited by the characteristics of the studies that were included. Guevara et al’s review could make limited comments about: non-severe asthma, interventions at different ages, the types of education that is most effective and whether education should be directed at children, parents or both. The conclusion that education should be incorporated into the routine care of children with asthma is welcome but somewhat limited.

Guevara et al recommend further research, and here they are more specific. They suggest that:

These suggestions highlight the importance of understanding the intervention thoroughly. Guevara et al are essentially asking what might be the ‘active ingredient’ of education: is it directed at preventing the occurrence of acute episodes of illness or at improving quality of life? Is the active ingredient individualised education or education provided for groups? Their other question, ‘Should education be based on symptoms or on peak flow measurement?’ can be answered by another study from lower in the hierarchy of study designs. Forty asthmatic children (5–16 years) were asked to:

The authors reported that children’s peak flow results were so unreliable that they should not be trusted:

This illustrates how descriptive studies that add to understanding of how people behave in health care can make important contributions to developing evidence-based practice. (The ‘ p’ value reported by Kamps refers to the probability that the difference was caused by chance, in most studies it is accepted that a result is statistically significant if there was a less than 5% (1 in 20) likelihood that the differences between two values could have been caused by chance.)

Randomised controlled trials (RCTs) have shown that nurses can make important differences to the health of children with asthma by educating them about self-management of their condition:

Madge et al’s study demonstrated both the value and limitations of RCTs. The study showed that children educated by a nurse fared better, with less need for readmission and reduction in asthma symptoms. However, the problem of deciding what the ‘active ingredient’ is evident. One nurse delivered the education – would other nurses have been as effective? The education was delivered as children were discharged from hospital. The authors followed-up children to assess the duration of the effect of education but this analysis was limited by the variation in the period of follow-up from 2–14 months. There is evidence that asthma attacks affect receptiveness to information in adults with asthma:

Was the education effective in Madge et al’s study because it was delivered while the children were being discharged from hospital? In a qualitative study of parents’ and children’s perspectives on asthma it was reported that parents gave more emphasis to acute attacks than children and that they accepted some symptoms as tolerable provided that they did not result in an acute attack (Callery et al 2003). Could education at the time of an attack reinforce parental focus on acute attacks and hospitalisations rather than drawing parents’ attention to controlling low-level symptoms that might affect children’s quality of life? In addition to the educational intervention, the authors reported that parents were given oral steroids and instructions about their use. Were oral steroids one of the ‘active ingredients’ that resulted in improved outcomes for the intervention group? These questions arise because of the difficulty of separating different effects, or confounding variables, on the outcomes. The questions also indicate the importance of a thorough understanding of interventions and how they are seen by those involved, including children as well as parents and health professionals. Therefore descriptive studies, including qualitative studies, are required to complement experimental studies in the development and evaluation of complex interventions.