House dust mite control measures in the management of asthma: meta-analysisBMJ 1998; 317 doi: https://doi.org/10.1136/bmj.317.7166.1105 (Published 24 October 1998) Cite this as: BMJ 1998;317:1105
- aNordic Cochrane Centre, Rigshospitalet, Department 7112, DK-2200 Copenhagen N, Denmark
- bExecutive Office, Unit of Public Health, Municipality of Gotland, S-62181 Visby, Sweden
- cCentre for Applied Public Health Medicine, University of Wales College of Medicine, Cardiff CF1 3NW
- Correspondence to: Dr Gøtzsche
Objective To determine whether patients with asthma who are sensitive to mites benefit from measures designed to reduce their exposure to house dust mite antigen in the home.
Design Meta-analysis of randomised trials that investigated the effects on asthma patients of chemical or physical measures to control mites, or both, in comparison with an untreated control group. All trials in any language were eligible for inclusion.
Subjects Patients with bronchial asthma as diagnosed by a doctor and sensitisation to mites as determined by skin prick testing, bronchial provocation testing, or serum assays for specific IgE antibodies.
Main outcome measures Number of patients whose allergic symptoms improved, improvement in asthma symptoms, improvement in peak expiratory flow rate. Outcomes measured on different scales were combined using the standardised effect size method (the difference in effect was divided by the standard deviation of the measurements).
Results 23 studies were included in the meta-analysis; 6 studies used chemical methods to reduce exposure to mites, 13 used physical methods, and 4 used a combination. Altogether, 41/113 patients exposed to treatment interventions improved compared with 38/117 in the control groups (odds ratio 1.20, 95% confidence interval 0.66 to 2.18). The standardised mean difference for improvement in asthma symptoms was −0.06 (95% confidence interval −0.54 to 0.41). For peak flow rate measured in the morning the standardised mean difference was −0.03 (−0.25 to 0.19). As measured in the original units this difference between the treatment and the control group corresponds to −3 l/min (95% confidence interval −25 l/min to 19 l/min). The results were similar in the subgroups of trials that reported successful reduction in exposure to mites or had long follow up times.
Conclusion Current chemical and physical methods aimed at reducing exposure to allergens from house dust mites seem to be ineffective and cannot be recommended as prophylactic treatment for asthma patients sensitive to mites.
Current chemical and physical methods aimed at reducing exposure to allergens from house dust mites seem to be ineffective; these methods cannot be recommended as prophylactic treatment for asthma patients who are sensitive to mites
It is unlikely that a worthwhile effect has been overlooked in this meta-analysis since the confidence interval for the peak expiratory flow rate was quite narrow
Future studies should be much larger and more rigorous than those in this meta-analysis and should evaluate other methods of mite control than those used to date
The major allergen in house dust is derived from mites, and a recent review concluded that the environmental control of allergens should be an integral part of the management of sensitised patients.1 Some of the evidence in the review, however, was derived from observational studies. Since clinical trials have shown equivocal results of the effectiveness of measures to reduce exposure to mite antigen, we decided to synthesise the findings of all clinical trials.
Our objective was to determine whether patients with asthma who were sensitised to house dust mites benefited from measures designed to reduce their exposure to mite antigen in the home. All randomised trials in any language performed at any time that compared chemical (acaricidal) or physical measures (such as vacuum cleaning, heating, barrier methods, or air filtration systems) to control mites and analysed their effects on patients with bronchial asthma as compared with an untreated control group were eligible for inclusion in the meta-analysis. Asthma had to have been diagnosed by a doctor and sensitisation to mites had to have been assessed by skin tests, bronchial provocation tests, or serum assays for specific IgE antibodies.
We searched the Asthma and Wheez* databases set up by the Cochrane Airways Group which contain records from the Cumulative Index to Nursing and Allied Health Literature, Medline, and Embase. Mite* in the title, abstract, or keyword (descriptor) field was combined with random*, trial*, placebo, double-blind, double blind, single-blind, single blind, comparative study, or controlled study in all fields. Primary authors were contacted to obtain additional information if necessary. CH searched issues of Respiration (1980-96) and MB searched Clinical and Experimental Allergy (1980-96) by hand.
Extraction of data
Two of the authors (CH and MB) selected the trials for inclusion. Two (PCG and CH) extracted data on the following outcomes: subjective wellbeing, improvement in asthma symptoms, use of drugs to control asthma, number of days of sick leave taken from school or work, number of unscheduled visits made to a doctor or hospital, forced expiratory volume in 1 second, peak expiratory flow rate, provocative concentration that causes a 20% fall in forced expiratory volume in 1 second, and results of skin prick testing. Ambiguities were resolved by discussion.
Review Manager software was used to analyse the data.2 If P<0.10 in the test for heterogeneity a random effects analysis was carried out. Since the results from crossover trials were usually reported as if they had come from a parallel group trial we used the data accordingly and assumed that no carryover effect had occurred. Continuous data were often presented on different scales in different studies (for example, peak expiratory flow rate was given either as absolute values or as a per cent of predicted values). Because of this, we calculated the standardised mean difference in our analysis of these data. With this method, the difference in effect is divided by the standard deviation of the measurements. Since data on wellbeing and improvements in asthma symptoms were closely related we summarised categorical data as the number of patients whose asthma improved; we summarised continuous data in the category of asthma symptoms. In general, the provocative concentration that causes a 20% fall in the forced expiratory volume in 1 second had been analysed after logarithmic transformation because the data were highly skewed. If the mean values and standard deviations had been converted from the logarithmic to the arithmetic scale we reconverted them.3 We excluded data on the provocative concentration from one study which had not used logarithmic transformation (see appendix 1 on the website).
In studies in which the use of several anti-asthma treatments had been reported we used the data on bronchodilators. In studies in which data were recorded at several points in time we used the longest observation period during which patients were still on randomised treatment.
We did not adjust for baseline differences since inequalities occurring despite randomisation would be expected to cancel each other out over a number of trials. Furthermore, baseline values were not always available. If we had made adjustments when possible we would have risked biasing the review since investigators may be inclined to show baseline differences and adjust for them when this procedure favours the experimental treatment. Bias occurring during the analysis of data is common and almost always favours the new treatment.4
Trials included in the analysis
Altogether, 458 references were identified; half of these were irrelevant and the other half were retrieved so that the full study could be examined. Eighteen of the 229 studies met the inclusion criteria.5–22 Another four trials were retrieved from MB's personal archive.23–26 The reference lists of the 229 articles retrieved were also searched but no further appropriate studies were found. One of the papers included in the analysis11 reported on a trial with three arms; this was treated as two separate trials in the meta-analysis. Thus, most of the analyses below refer to 23 trials. (A list of the excluded trials which were not evidently irrelevant and the reasons for their exclusion appear in appendix 2 on the website.)
All studies had used skin prick testing for diagnosis of mite sensitisation. Extracts were from Dermatophagoides pteronyssinus or D farinae except in two trials which had used an unspecified extract of house dust. 2226 Two trials also used a serum assay for specific IgE antibodies. 911 Patients had mostly been recruited from asthma and chest clinics. The specific criteria for the diagnosis of asthma were mentioned only in four papers.11141725
Six studies used chemical methods to reduce exposure to mites, 91112161724 13 used physical methods, 5–8131518–222526 and four used a combination of methods. 10111423 Five studies did not assess the reduction of the population of mites.615182226 Reduction in the exposure to mites occurred in six studies 101119212325 ; reduction was unsuccessful in 12. 57–911–1416172014 The length of follow up, from two weeks to one year, and other characteristics of the studies are shown in the table.
Only one study reported enough information to allow us to determine that allocation had been adequately concealed12; in the remaining articles authors stated that the study was randomised. Thirteen of the studies were double blind or placebo controlled with blind assessment; two studies used blinded assessors and one had blinded participants; nine were crossover studies (table). Altogether, 686 patients were enrolled in the studies.
Results of meta-analysis
The total number of patients who improved after intervention was similar to the total number who improved among the control groups (41/113 in treatment group v 38/117 in control group, odds ratio 1.20, 95% confidence interval 0.66 to 2.18) (fig 1). Improvements in asthma symptoms were heterogeneous (P<0.0001) but there was no indication of an effect. The standardised mean difference of these scores was −0.06 (95% confidence interval −0.54 to 0.41) (fig 2). The result was similar when analysis was done with a fixed effects model (−0.01, −0.28 to 0.26).
The most commonly reported outcome was peak expiratory flow rate in the morning (fig 3). The time of day that peak expiratory flow was measured was not stated in one study19; we assumed that it had been measured in the morning. The standardised mean difference between peak expiratory flow rates was −0.03 (−0.25 to 0.19). In the analysis of chemical methods to reduce the population of house dust mites there was a significant difference between treatment and control groups (−0.50, −0.98 to −0.01) which favoured the control group (one of the two studies had a baseline difference which favoured the control group9). In the analysis of five crossover trials of physical methods the difference was 0.06 (−0.26 to 0.37) (fig 3); for the only parallel group trial the difference was 0.33 (−0.28 to 0.94).19 For the two studies evaluating a combination of methods the difference was 0.02 (−0.46 to 0.50).
Peak expiratory flow in the evening was only reported in six trials 5910152021 ; the difference was −0.13 (−0.48 to 0.22). For the only study that reported on chemical methods the difference was −1.08 (−1.97 to −0.20) in favour of the control group.9 For physical methods of control the difference was 0.06 (−0.35 to 0.47). For the trials evaluating a combination of methods it was −0.03 (−0.96 to 0.89). The difference for forced expiratory volume in 1 second was 0.09 (−0.16 to 0.33).
There was no difference between the treatments in their effects on provocative concentration (standardised mean difference 0.04, −0.32 to 0.23) or use of drugs (−0.14, −0.43 to 0.15). Data for chemical methods were given only in one study in which the use of anti-asthma drugs was significantly higher in the treatment group (0.89, 0.02 to 1.75).9
None of the studies reported on the number of unscheduled visits made to a doctor or hospital. One study reported that three patients missed school during the control period and none missed school during treatment but did not give reasons for these differences.22 Results of skin prick testing after treatment were not reported in any study.
In the subgroup of trials that reported a successful reduction in the population of mites the results were similar to the overall results.101119212325 For measures of morning peak flow rate the difference was 0.11 (−0.22 to 0.45). The only parallel group trial in this subgroup had a baseline difference that favoured the experimental group19; if this trial is excluded the difference becomes 0.02 (−0.39 to 0.42).
We were unable to show any clinical benefit from measures designed to reduce exposure to mites among asthma patients who were sensitive to mites. Since patients with asthma are frequently sensitive to house dust mite allergen, the most likely explanation for our negative findings is that the methods studied did not adequately reduce levels of mite antigens. Those few studies in which exposure to mites was effectively reduced did not have results that were more positive than studies in which exposure to mites was not reduced. This may be because patients with asthma who are sensitive to mites are usually also sensitive to other allergens; the successful elimination of only one allergen may be of limited benefit.
It seems unlikely that the initial levels of mite infestation were already too low for any reduction to be effective. Quite low concentrations of allergen can affect bronchial responsiveness, 1728 and the concentrations in the studies reviewed would usually represent a risk to patients sensitive to mites.
A lack of compliance with the measures to control mites could have played a part in the negative results, but only in one study25 was adherence to protocols evaluated. Adherence to protocols was higher and the amount of mites patients were exposed to was lower in the group that received computer assisted instruction when compared with the group that received conventional instruction. Those subjects who received computer assisted instruction implemented significantly more avoidance measures and had fewer symptoms.
Our meta-analysis did not seem to lack power. The point estimates were close to zero and the confidence interval was narrow for morning peak flow rate, the most commonly used outcome measure, which is related to the severity of the asthma and is sensitive to change. This does not suggest we missed any worthwhile effect. If the difference in morning peak flow is transformed into the most commonly used unit of measurement (l/min) with a standard deviation of 100 l/min (in accordance with fig 3), the difference in peak flow between treatment and control groups is only −3 l/min (−25 l/min to 19 l/min).
Potential sources of bias
Potential sources of bias must be considered. Randomisation methods were not reported except in one study. In several studies researchers or patients were not blinded, and most studies were small. These factors all tend to be associated with an overestimation of reported treatment effects. Further, reporting was variable (for example, one study reported only that there were no significant changes in symptom scores, drug requirements, or peak flow rates13). It is generally safe to assume that unreported data do not favour the intervention. On a few occasions it was necessary to correct the original data; for example, in one study we could not confirm a reported significant effect on mite allergen level.12
We tried to avoid bias while extracting data (for example, by making blinded decisions when several options were available). On a few occasions, however, we had to use data that favoured the experimental interventions (appendix 1 on the website). Finally, the trials we excluded did not have positive results (appendix 2 on the website) and we therefore believe we have not favoured the null hypothesis of no treatment effect in our meta-analysis.
There is a possibility that the results of effective interventions have been diluted by ineffective ones or by trial designs that were insufficiently rigorous. The length of follow up varied but was completely unrelated to the effect of treatment (for example, the three trials with 6-12 months of follow up showed a difference of 0.01 (−0.36 to 0.38) in morning peak flow). This is to be expected since an effect on the reduction of allergens should be noticeable in the short term because mite allergen causes a Type I hypersensitivity reaction. There may be a subset of patients who are highly sensitive to mites who would benefit from mite eradication. It would, however, be difficult to detect such patients and it seems more reasonable to assess the effects of mite eradication on all patients with asthma whose skin prick tests indicate a sensitivity to mites.
Current chemical and physical methods for eradicating mites or reducing exposure to mites seem to be ineffective and cannot be recommended as prophylactic treatment for asthma patients who are sensitive to mites. It is doubtful whether conducting further studies similar to the ones in our meta-analysis would be worthwhile. In particular, several of the trials had used extensive mite eradication and avoidance schemes. We suggest that future studies should be much larger and more rigorous than those analysed here and should use methods to control or eradicate mites other than those used so far. Our review is also published in The Cochrane Library29 where it will be updated when results from additional studies become available.
We would like to thank Professor Paul W Jones, Mr Steve Milan, Ms Anna Bara, and Dr Jane Dennis, of the Cochrane Airways Group; Professor Vinod K Diwan and Professor Martin Bland for helpful support; and Dr Leonardo Antonicelli for providing additional data.
Contributors: CH wrote the draft protocol for the meta-analysis. CH and MB selected trials for inclusion. Trials were reviewed by all authors. Quality assessment of the trials was primarily done by CH, outcome data were extracted primarily by PCG (but discussed in detail with all authors), CH drafted the first manuscript for the Cochrane Library, PCG drafted the manuscript for the journal article. All authors are guarantors for the article, and PCG is guarantor for the statistical calculations.
Funding: Nordic Council of Ministers; Hovedstadens Sygehusfaellesskab, Rigshospitalet, Denmark; Sygekassernes Helsefond, Denmark; the Swedish Heart Lung Foundation (grant 54506).
Conflict of interest: None.