Seven RCTs were included (n=2,316, range 100 to 937). A single study reported blinded data collection. In one RCT data was collected by those who carried out the intervention. One RCT had limited power. Standard protocols were used in four RCTs. Length of follow-up ranged from four months to two years. In some studies, drop-out differed between intervention and control groups or for different types of participants.
Asthma-related health outcomes: There was a significant decrease in caregiver-reported asthma symptoms in the intervention groups (four RCTs) and in caregiver reports of activity limitations (three RCTs); these persisted for two years after the intervention in one RCT. There were significantly fewer unscheduled asthma-related clinical visits associated with interventions in three of six RCTs. One of four RCTs found a significant benefit on medication use (for reduced under-treatment for children who should be on control medication). Similarly, one of four RCTs reported a significant effect on lung function: a significant improvement in daily nadir peak expiratory flow (PF) and forced expiratory volume (FEV1).
Direct mediators of improved health outcomes: Findings in relation to behaviour change were mixed both within and between studies with improvements for some behaviours but not for others. Mostly the behaviours that improved were related to resources provided as part of the intervention, which included pillow and mattress encasement, vacuum cleaner use, kitchen exhaust fan use, use of caulk and boric acid, and use of doormats or removal of shoes. There was no statistically significant difference between groups in the two studies that assessed knowledge. All seven RCTs measured a variety of allergens and exposures with mixed results. Significant improvements were reported in reduced: cockroach (but not rodent) numbers; dust mite, cat, and dog antigens depending on the location investigated; combined dog, cat, dust mite and cockroach antigens; nitric acid concentration; particulate matter; dust weight; condensation and moisture scores; and composite trigger scores; but not for antigen concentration in dust. Some significant results were not sustained after longer follow-up periods.